--------------------------------------------------------------------------
WARNING: Linux kernel CMA support was requested via the
btl_vader_single_copy_mechanism MCA variable, but CMA support is
not available due to restrictive ptrace settings.

The vader shared memory BTL will fall back on another single-copy
mechanism if one is available. This may result in lower performance.

  Local host: ip-172-31-29-212
--------------------------------------------------------------------------
[1,1]<stdout>:Using AutoAugment
[1,3]<stdout>:Using AutoAugment
[1,7]<stdout>:Using AutoAugment
[1,6]<stdout>:Using AutoAugment
[1,5]<stdout>:Using AutoAugment
[1,0]<stderr>:INFO:root:Namespace(auto_aug=True, batch_size=128, crop_ratio=0.875, data_nthreads=12, dropblock_prob=0, dtype='float32', eval_frequency=5, hard_weight=0.5, input_size=224, label_smoothing=True, last_gamma=True, log_interval=100, lr=0.05, mixup=True, mixup_alpha=0.2, mixup_off_epoch=0, model='resnest14', momentum=0.9, no_cuda=False, no_wd=True, num_epochs=270, rec_train='/home/ubuntu/data/ILSVRC2012/train.rec', rec_val='/home/ubuntu/data/ILSVRC2012/val.rec', resume_epoch=0, resume_params='', resume_states='', save_dir='params_resnest14', save_frequency=20, teacher=None, temperature=20, use_avd=False, use_pretrained=False, use_rec=True, use_sk=False, warmup_epochs=5, warmup_lr=0.0, wd=0.0001)
[1,2]<stdout>:Using AutoAugment
[1,0]<stdout>:Using AutoAugment
[1,4]<stdout>:Using AutoAugment
[1,0]<stderr>:INFO:root:ResNeSt(
[1,0]<stderr>:  (conv1): HybridSequential(
[1,0]<stderr>:    (0): Conv2D(3 -> 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
[1,0]<stderr>:    (1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=32)
[1,0]<stderr>:    (2): Activation(relu)
[1,0]<stderr>:    (3): Conv2D(32 -> 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
[1,0]<stderr>:    (4): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=32)
[1,0]<stderr>:    (5): Activation(relu)
[1,0]<stderr>:    (6): Conv2D(32 -> 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
[1,0]<stderr>:  )
[1,0]<stderr>:  (bn1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=64)
[1,0]<stderr>:  (relu): Activation(relu)
[1,0]<stderr>:  (maxpool): MaxPool2D(size=(3, 3), stride=(2, 2), padding=(1, 1), ceil_mode=False, global_pool=False, pool_type=max, layout=NCHW)
[1,0]<stderr>:  (layer1): HybridSequential(
[1,0]<stderr>:    (0): Bottleneck(
[1,0]<stderr>:      (conv1): Conv2D(64 -> 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:      (bn1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=64)
[1,0]<stderr>:      (relu1): Activation(relu)
[1,0]<stderr>:      (conv2): SplitAttentionConv(
[1,0]<stderr>:        (conv): Conv2D(32 -> 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=2, bias=False)
[1,0]<stderr>:        (bn): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=128)
[1,0]<stderr>:        (relu): Activation(relu)
[1,0]<stderr>:        (fc1): Conv2D(64 -> 32, kernel_size=(1, 1), stride=(1, 1))
[1,0]<stderr>:        (bn1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=32)
[1,0]<stderr>:        (relu1): Activation(relu)
[1,0]<stderr>:        (fc2): Conv2D(32 -> 128, kernel_size=(1, 1), stride=(1, 1))
[1,0]<stderr>:      )
[1,0]<stderr>:      (conv3): Conv2D(64 -> 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:      (bn3): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=256)
[1,0]<stderr>:      (relu3): Activation(relu)
[1,0]<stderr>:      (downsample): HybridSequential(
[1,0]<stderr>:        (0): AvgPool2D(size=(1, 1), stride=(1, 1), padding=(0, 0), ceil_mode=True, global_pool=False, pool_type=avg, layout=NCHW)
[1,0]<stderr>:        (1): Conv2D(64 -> 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:        (2): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=256)
[1,0]<stderr>:      )
[1,0]<stderr>:    )
[1,0]<stderr>:  )
[1,0]<stderr>:  (layer2): HybridSequential(
[1,0]<stderr>:    (0): Bottleneck(
[1,0]<stderr>:      (conv1): Conv2D(256 -> 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:      (bn1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=128)
[1,0]<stderr>:      (relu1): Activation(relu)
[1,0]<stderr>:      (conv2): SplitAttentionConv(
[1,0]<stderr>:        (conv): Conv2D(64 -> 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=2, bias=False)
[1,0]<stderr>:        (bn): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=256)
[1,0]<stderr>:        (relu): Activation(relu)
[1,0]<stderr>:        (fc1): Conv2D(128 -> 64, kernel_size=(1, 1), stride=(1, 1))
[1,0]<stderr>:        (bn1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=64)
[1,0]<stderr>:        (relu1): Activation(relu)
[1,0]<stderr>:        (fc2): Conv2D(64 -> 256, kernel_size=(1, 1), stride=(1, 1))
[1,0]<stderr>:      )
[1,0]<stderr>:      (conv3): Conv2D(128 -> 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:      (bn3): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=512)
[1,0]<stderr>:      (avd_layer): AvgPool2D(size=(3, 3), stride=(2, 2), padding=(1, 1), ceil_mode=False, global_pool=False, pool_type=avg, layout=NCHW)
[1,0]<stderr>:      (relu3): Activation(relu)
[1,0]<stderr>:      (downsample): HybridSequential(
[1,0]<stderr>:        (0): AvgPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=True, global_pool=False, pool_type=avg, layout=NCHW)
[1,0]<stderr>:        (1): Conv2D(256 -> 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:        (2): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=512)
[1,0]<stderr>:      )
[1,0]<stderr>:    )
[1,0]<stderr>:  )
[1,0]<stderr>:  (layer3): HybridSequential(
[1,0]<stderr>:    (0): Bottleneck(
[1,0]<stderr>:      (conv1): Conv2D(512 -> 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:      (bn1): BatchNorm(axis=1, eps=1e-05, mo[1,0]<stderr>:mentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=256)
[1,0]<stderr>:      (relu1): Activation(relu)
[1,0]<stderr>:      (conv2): SplitAttentionConv(
[1,0]<stderr>:        (conv): Conv2D(128 -> 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=2, bias=False)
[1,0]<stderr>:        (bn): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=512)
[1,0]<stderr>:        (relu): Activation(relu)
[1,0]<stderr>:        (fc1): Conv2D(256 -> 128, kernel_size=(1, 1), stride=(1, 1))
[1,0]<stderr>:        (bn1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=128)
[1,0]<stderr>:        (relu1): Activation(relu)
[1,0]<stderr>:        (fc2): Conv2D(128 -> 512, kernel_size=(1, 1), stride=(1, 1))
[1,0]<stderr>:      )
[1,0]<stderr>:      (conv3): Conv2D(256 -> 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:      (bn3): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=1024)
[1,0]<stderr>:      (avd_layer): AvgPool2D(size=(3, 3), stride=(2, 2), padding=(1, 1), ceil_mode=False, global_pool=False, pool_type=avg, layout=NCHW)
[1,0]<stderr>:      (relu3): Activation(relu)
[1,0]<stderr>:      (downsample): HybridSequential(
[1,0]<stderr>:        (0): AvgPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=True, global_pool=False, pool_type=avg, layout=NCHW)
[1,0]<stderr>:        (1): Conv2D(512 -> 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:        (2): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=1024)
[1,0]<stderr>:      )
[1,0]<stderr>:    )
[1,0]<stderr>:  )
[1,0]<stderr>:  (layer4): HybridSequential(
[1,0]<stderr>:    (0): Bottleneck(
[1,0]<stderr>:      (conv1): Conv2D(1024 -> 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:      (bn1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=512)
[1,0]<stderr>:      (relu1): Activation(relu)
[1,0]<stderr>:      (conv2): SplitAttentionConv(
[1,0]<stderr>:        (conv): Conv2D(256 -> 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=2, bias=False)
[1,0]<stderr>:        (bn): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=1024)
[1,0]<stderr>:        (relu): Activation(relu)
[1,0]<stderr>:        (fc1): Conv2D(512 -> 256, kernel_size=(1, 1), stride=(1, 1))
[1,0]<stderr>:        (bn1): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=256)
[1,0]<stderr>:        (relu1): Activation(relu)
[1,0]<stderr>:        (fc2): Conv2D(256 -> 1024, kernel_size=(1, 1), stride=(1, 1))
[1,0]<stderr>:      )
[1,0]<stderr>:      (conv3): Conv2D(512 -> 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:      (bn3): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=2048)
[1,0]<stderr>:      (avd_layer): AvgPool2D(size=(3, 3), stride=(2, 2), padding=(1, 1), ceil_mode=False, global_pool=False, pool_type=avg, layout=NCHW)
[1,0]<stderr>:      (relu3): Activation(relu)
[1,0]<stderr>:      (downsample): HybridSequential(
[1,0]<stderr>:        (0): AvgPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=True, global_pool=False, pool_type=avg, layout=NCHW)
[1,0]<stderr>:        (1): Conv2D(1024 -> 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
[1,0]<stderr>:        (2): BatchNorm(axis=1, eps=1e-05, momentum=0.9, fix_gamma=False, use_global_stats=False, in_channels=2048)
[1,0]<stderr>:      )
[1,0]<stderr>:    )
[1,0]<stderr>:  )
[1,0]<stderr>:  (avgpool): GlobalAvgPool2D(size=(1, 1), stride=(1, 1), padding=(0, 0), ceil_mode=True, global_pool=True, pool_type=avg, layout=NCHW)
[1,0]<stderr>:  (flat): Flatten
[1,0]<stderr>:  (fc): Dense(2048 -> 1000, linear)
[1,0]<stderr>:)
[ip-172-31-29-212:52930] 7 more processes have sent help message help-btl-vader.txt / cma-permission-denied
[ip-172-31-29-212:52930] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages
[1,7]<stderr>:[02:55:16] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,1]<stderr>:[02:55:17] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,2]<stderr>:[02:55:17] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,6]<stderr>:[02:55:18] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,4]<stderr>:[02:55:18] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,3]<stderr>:[02:55:18] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,0]<stderr>:[02:55:19] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,5]<stderr>:[02:55:20] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,0]<stderr>:INFO:root:Epoch[0] Batch[100] Loss[6.968]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[100]	rmse=0.026110	lr=0.006390
[1,0]<stderr>:INFO:root:Epoch[0] Batch[200] Loss[6.902]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[200]	rmse=0.025904	lr=0.012780
[1,0]<stderr>:INFO:root:Epoch[0] Batch[300] Loss[6.920]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[300]	rmse=0.025836	lr=0.019169
[1,0]<stderr>:INFO:root:Epoch[0] Batch[400] Loss[6.819]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[400]	rmse=0.025832	lr=0.025559
[1,0]<stderr>:INFO:root:Epoch[0] Batch[500] Loss[6.645]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[500]	rmse=0.025945	lr=0.031949
[1,0]<stderr>:INFO:root:Epoch[0] Batch[600] Loss[6.698]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[600]	rmse=0.025930	lr=0.038339
[1,0]<stderr>:INFO:root:Epoch[0] Batch[700] Loss[6.723]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[700]	rmse=0.025995	lr=0.044728
[1,0]<stderr>:INFO:root:Epoch[0] Batch[800] Loss[6.681]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[800]	rmse=0.026082	lr=0.051118
[1,0]<stderr>:INFO:root:Epoch[0] Batch[900] Loss[6.368]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[900]	rmse=0.026126	lr=0.057508
[1,0]<stderr>:INFO:root:Epoch[0] Batch[1000] Loss[6.377]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[1000]	rmse=0.026119	lr=0.063898
[1,0]<stderr>:INFO:root:Epoch[0] Batch[1100] Loss[6.606]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[1100]	rmse=0.026145	lr=0.070288
[1,0]<stderr>:INFO:root:Epoch[0] Batch[1200] Loss[6.162]
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[1200]	rmse=0.026096	lr=0.076677
[1,0]<stderr>:INFO:root:Epoch[0] Rank[0] Batch[1251]	Time cost=411.25	Train-metric=0.026114
[1,0]<stderr>:INFO:root:Epoch[0]	Speed: 3114.92 samples/sec
[1,0]<stderr>:INFO:root:Epoch[1] Batch[100] Loss[6.538]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[100]	rmse=0.026010	lr=0.086326
[1,0]<stderr>:INFO:root:Epoch[1] Batch[200] Loss[6.005]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[200]	rmse=0.026021	lr=0.092716
[1,0]<stderr>:INFO:root:Epoch[1] Batch[300] Loss[6.032]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[300]	rmse=0.026005	lr=0.099105
[1,0]<stderr>:INFO:root:Epoch[1] Batch[400] Loss[6.288]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[400]	rmse=0.025951	lr=0.105495
[1,0]<stderr>:INFO:root:Epoch[1] Batch[500] Loss[5.896]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[500]	rmse=0.025931	lr=0.111885
[1,0]<stderr>:INFO:root:Epoch[1] Batch[600] Loss[5.924]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[600]	rmse=0.026036	lr=0.118275
[1,0]<stderr>:INFO:root:Epoch[1] Batch[700] Loss[5.784]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[700]	rmse=0.026068	lr=0.124665
[1,0]<stderr>:INFO:root:Epoch[1] Batch[800] Loss[6.021]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[800]	rmse=0.026045	lr=0.131054
[1,0]<stderr>:INFO:root:Epoch[1] Batch[900] Loss[6.072]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[900]	rmse=0.026026	lr=0.137444
[1,0]<stderr>:INFO:root:Epoch[1] Batch[1000] Loss[5.878]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[1000]	rmse=0.026021	lr=0.143834
[1,0]<stderr>:INFO:root:Epoch[1] Batch[1100] Loss[6.373]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[1100]	rmse=0.025989	lr=0.150224
[1,0]<stderr>:INFO:root:Epoch[1] Batch[1200] Loss[6.210]
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[1200]	rmse=0.025946	lr=0.156613
[1,0]<stderr>:INFO:root:Epoch[1] Rank[0] Batch[1251]	Time cost=403.06	Train-metric=0.025975
[1,0]<stderr>:INFO:root:Epoch[1]	Speed: 3178.23 samples/sec
[1,0]<stderr>:INFO:root:Epoch[2] Batch[100] Loss[6.393]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[100]	rmse=0.025684	lr=0.166262
[1,0]<stderr>:INFO:root:Epoch[2] Batch[200] Loss[5.491]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[200]	rmse=0.025817	lr=0.172652
[1,0]<stderr>:INFO:root:Epoch[2] Batch[300] Loss[5.498]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[300]	rmse=0.025850	lr=0.179042
[1,0]<stderr>:INFO:root:Epoch[2] Batch[400] Loss[5.816]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[400]	rmse=0.025743	lr=0.185431
[1,0]<stderr>:INFO:root:Epoch[2] Batch[500] Loss[6.468]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[500]	rmse=0.025642	lr=0.191821
[1,0]<stderr>:INFO:root:Epoch[2] Batch[600] Loss[5.075]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[600]	rmse=0.025579	lr=0.198211
[1,0]<stderr>:INFO:root:Epoch[2] Batch[700] Loss[5.993]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[700]	rmse=0.025634	lr=0.204601
[1,0]<stderr>:INFO:root:Epoch[2] Batch[800] Loss[5.910]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[800]	rmse=0.025607	lr=0.210990
[1,0]<stderr>:INFO:root:Epoch[2] Batch[900] Loss[4.931]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[900]	rmse=0.025564	lr=0.217380
[1,0]<stderr>:INFO:root:Epoch[2] Batch[1000] Loss[5.203]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[1000]	rmse=0.025557	lr=0.223770
[1,0]<stderr>:INFO:root:Epoch[2] Batch[1100] Loss[6.421]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[1100]	rmse=0.025573	lr=0.230160
[1,0]<stderr>:INFO:root:Epoch[2] Batch[1200] Loss[5.375]
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[1200]	rmse=0.025520	lr=0.236550
[1,0]<stderr>:INFO:root:Epoch[2] Rank[0] Batch[1251]	Time cost=399.89	Train-metric=0.025506
[1,0]<stderr>:INFO:root:Epoch[2]	Speed: 3203.44 samples/sec
[1,0]<stderr>:INFO:root:Epoch[3] Batch[100] Loss[4.938]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[100]	rmse=0.025412	lr=0.246198
[1,0]<stderr>:INFO:root:Epoch[3] Batch[200] Loss[4.823]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[200]	rmse=0.025167	lr=0.252588
[1,0]<stderr>:INFO:root:Epoch[3] Batch[300] Loss[4.837]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[300]	rmse=0.025393	lr=0.258978
[1,0]<stderr>:INFO:root:Epoch[3] Batch[400] Loss[5.892]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[400]	rmse=0.025473	lr=0.265367
[1,0]<stderr>:INFO:root:Epoch[3] Batch[500] Loss[6.228]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[500]	rmse=0.025425	lr=0.271757
[1,0]<stderr>:INFO:root:Epoch[3] Batch[600] Loss[4.889]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[600]	rmse=0.025333	lr=0.278147
[1,0]<stderr>:INFO:root:Epoch[3] Batch[700] Loss[5.523]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[700]	rmse=0.025333	lr=0.284537
[1,0]<stderr>:INFO:root:Epoch[3] Batch[800] Loss[4.585]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[800]	rmse=0.025308	lr=0.290927
[1,0]<stderr>:INFO:root:Epoch[3] Batch[900] Loss[4.672]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[900]	rmse=0.025262	lr=0.297316
[1,0]<stderr>:INFO:root:Epoch[3] Batch[1000] Loss[4.370]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[1000]	rmse=0.025265	lr=0.303706
[1,0]<stderr>:INFO:root:Epoch[3] Batch[1100] Loss[4.562]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[1100]	rmse=0.025291	lr=0.310096
[1,0]<stderr>:INFO:root:Epoch[3] Batch[1200] Loss[4.394]
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[1200]	rmse=0.025231	lr=0.316486
[1,0]<stderr>:INFO:root:Epoch[3] Rank[0] Batch[1251]	Time cost=402.00	Train-metric=0.025214
[1,0]<stderr>:INFO:root:Epoch[3]	Speed: 3186.64 samples/sec
[1,0]<stderr>:INFO:root:Epoch[4] Batch[100] Loss[5.590]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[100]	rmse=0.024661	lr=0.326134
[1,0]<stderr>:INFO:root:Epoch[4] Batch[200] Loss[4.566]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[200]	rmse=0.024546	lr=0.332524
[1,0]<stderr>:INFO:root:Epoch[4] Batch[300] Loss[4.503]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[300]	rmse=0.024636	lr=0.338914
[1,0]<stderr>:INFO:root:Epoch[4] Batch[400] Loss[4.048]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[400]	rmse=0.024755	lr=0.345304
[1,0]<stderr>:INFO:root:Epoch[4] Batch[500] Loss[4.986]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[500]	rmse=0.024727	lr=0.351693
[1,0]<stderr>:INFO:root:Epoch[4] Batch[600] Loss[4.287]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[600]	rmse=0.024665	lr=0.358083
[1,0]<stderr>:INFO:root:Epoch[4] Batch[700] Loss[4.859]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[700]	rmse=0.024630	lr=0.364473
[1,0]<stderr>:INFO:root:Epoch[4] Batch[800] Loss[4.753]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[800]	rmse=0.024636	lr=0.370863
[1,0]<stderr>:INFO:root:Epoch[4] Batch[900] Loss[3.926]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[900]	rmse=0.024622	lr=0.377252
[1,0]<stderr>:INFO:root:Epoch[4] Batch[1000] Loss[4.337]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[1000]	rmse=0.024565	lr=0.383642
[1,0]<stderr>:INFO:root:Epoch[4] Batch[1100] Loss[5.897]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[1100]	rmse=0.024542	lr=0.390032
[1,0]<stderr>:INFO:root:Epoch[4] Batch[1200] Loss[4.200]
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[1200]	rmse=0.024544	lr=0.396422
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0] Batch[1251]	Time cost=400.81	Train-metric=0.024516
[1,0]<stderr>:INFO:root:Epoch[4]	Speed: 3196.10 samples/sec
[1,1]<stderr>:[03:28:57] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,0]<stderr>:[03:28:57] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,6]<stderr>:[03:28:57] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,4]<stderr>:[03:28:57] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,2]<stderr>:[03:28:57] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,7]<stderr>:[03:28:58] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,3]<stderr>:[03:28:58] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,5]<stderr>:[03:28:58] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[1,0]<stderr>:INFO:root:Epoch[4] Rank[0]	Validation-accuracy=0.329120	Validation-top_k_accuracy_5=0.591300
[1,0]<stderr>:INFO:root:Epoch[5] Batch[100] Loss[6.020]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[100]	rmse=0.023842	lr=0.400000
[1,0]<stderr>:INFO:root:Epoch[5] Batch[200] Loss[4.476]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[200]	rmse=0.023963	lr=0.400000
[1,0]<stderr>:INFO:root:Epoch[5] Batch[300] Loss[5.560]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[300]	rmse=0.023967	lr=0.399999
[1,0]<stderr>:INFO:root:Epoch[5] Batch[400] Loss[5.731]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[400]	rmse=0.024017	lr=0.399999
[1,0]<stderr>:INFO:root:Epoch[5] Batch[500] Loss[4.411]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[500]	rmse=0.024053	lr=0.399998
[1,0]<stderr>:INFO:root:Epoch[5] Batch[600] Loss[3.651]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[600]	rmse=0.024110	lr=0.399997
[1,0]<stderr>:INFO:root:Epoch[5] Batch[700] Loss[4.031]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[700]	rmse=0.024078	lr=0.399996
[1,0]<stderr>:INFO:root:Epoch[5] Batch[800] Loss[3.611]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[800]	rmse=0.024064	lr=0.399994
[1,0]<stderr>:INFO:root:Epoch[5] Batch[900] Loss[3.908]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[900]	rmse=0.024084	lr=0.399993
[1,0]<stderr>:INFO:root:Epoch[5] Batch[1000] Loss[4.317]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[1000]	rmse=0.024048	lr=0.399991
[1,0]<stderr>:INFO:root:Epoch[5] Batch[1100] Loss[4.801]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[1100]	rmse=0.024017	lr=0.399989
[1,0]<stderr>:INFO:root:Epoch[5] Batch[1200] Loss[4.117]
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[1200]	rmse=0.024014	lr=0.399987
[1,0]<stderr>:INFO:root:Epoch[5] Rank[0] Batch[1251]	Time cost=396.90	Train-metric=0.024009
[1,0]<stderr>:INFO:root:Epoch[5]	Speed: 3227.59 samples/sec
[1,0]<stderr>:INFO:root:Epoch[6] Batch[100] Loss[5.530]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[100]	rmse=0.023413	lr=0.399984
[1,0]<stderr>:INFO:root:Epoch[6] Batch[200] Loss[3.681]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[200]	rmse=0.023487	lr=0.399981
[1,0]<stderr>:INFO:root:Epoch[6] Batch[300] Loss[5.505]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[300]	rmse=0.023505	lr=0.399979
[1,0]<stderr>:INFO:root:Epoch[6] Batch[400] Loss[5.287]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[400]	rmse=0.023518	lr=0.399976
[1,0]<stderr>:INFO:root:Epoch[6] Batch[500] Loss[5.637]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[500]	rmse=0.023517	lr=0.399973
[1,0]<stderr>:INFO:root:Epoch[6] Batch[600] Loss[3.670]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[600]	rmse=0.023561	lr=0.399969
[1,0]<stderr>:INFO:root:Epoch[6] Batch[700] Loss[4.074]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[700]	rmse=0.023520	lr=0.399966
[1,0]<stderr>:INFO:root:Epoch[6] Batch[800] Loss[3.888]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[800]	rmse=0.023505	lr=0.399962
[1,0]<stderr>:INFO:root:Epoch[6] Batch[900] Loss[5.590]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[900]	rmse=0.023511	lr=0.399959
[1,0]<stderr>:INFO:root:Epoch[6] Batch[1000] Loss[4.002]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[1000]	rmse=0.023506	lr=0.399955
[1,0]<stderr>:INFO:root:Epoch[6] Batch[1100] Loss[3.459]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[1100]	rmse=0.023486	lr=0.399951
[1,0]<stderr>:INFO:root:Epoch[6] Batch[1200] Loss[3.580]
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[1200]	rmse=0.023465	lr=0.399946
[1,0]<stderr>:INFO:root:Epoch[6] Rank[0] Batch[1251]	Time cost=401.24	Train-metric=0.023467
[1,0]<stderr>:INFO:root:Epoch[6]	Speed: 3192.67 samples/sec
[1,0]<stderr>:INFO:root:Epoch[7] Batch[100] Loss[5.722]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[100]	rmse=0.023108	lr=0.399940
[1,0]<stderr>:INFO:root:Epoch[7] Batch[200] Loss[3.651]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[200]	rmse=0.023077	lr=0.399935
[1,0]<stderr>:INFO:root:Epoch[7] Batch[300] Loss[5.294]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[300]	rmse=0.023188	lr=0.399930
[1,0]<stderr>:INFO:root:Epoch[7] Batch[400] Loss[3.546]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[400]	rmse=0.023169	lr=0.399925
[1,0]<stderr>:INFO:root:Epoch[7] Batch[500] Loss[3.698]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[500]	rmse=0.023181	lr=0.399919
[1,0]<stderr>:INFO:root:Epoch[7] Batch[600] Loss[4.370]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[600]	rmse=0.023200	lr=0.399914
[1,0]<stderr>:INFO:root:Epoch[7] Batch[700] Loss[3.864]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[700]	rmse=0.023206	lr=0.399908
[1,0]<stderr>:INFO:root:Epoch[7] Batch[800] Loss[5.142]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[800]	rmse=0.023172	lr=0.399903
[1,0]<stderr>:INFO:root:Epoch[7] Batch[900] Loss[4.452]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[900]	rmse=0.023170	lr=0.399897
[1,0]<stderr>:INFO:root:Epoch[7] Batch[1000] Loss[3.645]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[1000]	rmse=0.023207	lr=0.399890
[1,0]<stderr>:INFO:root:Epoch[7] Batch[1100] Loss[3.806]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[1100]	rmse=0.023203	lr=0.399884
[1,0]<stderr>:INFO:root:Epoch[7] Batch[1200] Loss[3.898]
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[1200]	rmse=0.023223	lr=0.399877
[1,0]<stderr>:INFO:root:Epoch[7] Rank[0] Batch[1251]	Time cost=402.08	Train-metric=0.023212
[1,0]<stderr>:INFO:root:Epoch[7]	Speed: 3186.02 samples/sec
[1,0]<stderr>:INFO:root:Epoch[8] Batch[100] Loss[3.634]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[100]	rmse=0.022877	lr=0.399867
[1,0]<stderr>:INFO:root:Epoch[8] Batch[200] Loss[3.400]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[200]	rmse=0.022901	lr=0.399860
[1,0]<stderr>:INFO:root:Epoch[8] Batch[300] Loss[4.747]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[300]	rmse=0.022917	lr=0.399853
[1,0]<stderr>:INFO:root:Epoch[8] Batch[400] Loss[5.370]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[400]	rmse=0.022890	lr=0.399846
[1,0]<stderr>:INFO:root:Epoch[8] Batch[500] Loss[3.378]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[500]	rmse=0.022935	lr=0.399838
[1,0]<stderr>:INFO:root:Epoch[8] Batch[600] Loss[5.494]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[600]	rmse=0.022949	lr=0.399831
[1,0]<stderr>:INFO:root:Epoch[8] Batch[700] Loss[4.168]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[700]	rmse=0.022956	lr=0.399823
[1,0]<stderr>:INFO:root:Epoch[8] Batch[800] Loss[4.942]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[800]	rmse=0.022968	lr=0.399815
[1,0]<stderr>:INFO:root:Epoch[8] Batch[900] Loss[3.396]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[900]	rmse=0.022944	lr=0.399806
[1,0]<stderr>:INFO:root:Epoch[8] Batch[1000] Loss[3.228]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[1000]	rmse=0.022935	lr=0.399798
[1,0]<stderr>:INFO:root:Epoch[8] Batch[1100] Loss[3.025]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[1100]	rmse=0.022899	lr=0.399789
[1,0]<stderr>:INFO:root:Epoch[8] Batch[1200] Loss[3.342]
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[1200]	rmse=0.022874	lr=0.399781
[1,0]<stderr>:INFO:root:Epoch[8] Rank[0] Batch[1251]	Time cost=399.20	Train-metric=0.022872
[1,0]<stderr>:INFO:root:Epoch[8]	Speed: 3208.99 samples/sec
[1,0]<stderr>:INFO:root:Epoch[9] Batch[100] Loss[3.327]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[100]	rmse=0.022718	lr=0.399767
[1,0]<stderr>:INFO:root:Epoch[9] Batch[200] Loss[4.662]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[200]	rmse=0.022743	lr=0.399758
[1,0]<stderr>:INFO:root:Epoch[9] Batch[300] Loss[5.136]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[300]	rmse=0.022730	lr=0.399748
[1,0]<stderr>:INFO:root:Epoch[9] Batch[400] Loss[3.182]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[400]	rmse=0.022725	lr=0.399739
[1,0]<stderr>:INFO:root:Epoch[9] Batch[500] Loss[5.391]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[500]	rmse=0.022673	lr=0.399729
[1,0]<stderr>:INFO:root:Epoch[9] Batch[600] Loss[4.219]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[600]	rmse=0.022653	lr=0.399719
[1,0]<stderr>:INFO:root:Epoch[9] Batch[700] Loss[5.428]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[700]	rmse=0.022639	lr=0.399709
[1,0]<stderr>:INFO:root:Epoch[9] Batch[800] Loss[4.167]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[800]	rmse=0.022641	lr=0.399699
[1,0]<stderr>:INFO:root:Epoch[9] Batch[900] Loss[3.790]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[900]	rmse=0.022647	lr=0.399688
[1,0]<stderr>:INFO:root:Epoch[9] Batch[1000] Loss[4.589]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[1000]	rmse=0.022653	lr=0.399677
[1,0]<stderr>:INFO:root:Epoch[9] Batch[1100] Loss[5.346]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[1100]	rmse=0.022666	lr=0.399667
[1,0]<stderr>:INFO:root:Epoch[9] Batch[1200] Loss[3.867]
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[1200]	rmse=0.022685	lr=0.399656
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0] Batch[1251]	Time cost=398.21	Train-metric=0.022700
[1,0]<stderr>:INFO:root:Epoch[9]	Speed: 3216.93 samples/sec
[1,0]<stderr>:INFO:root:Epoch[9] Rank[0]	Validation-accuracy=0.477440	Validation-top_k_accuracy_5=0.736700
[1,0]<stderr>:INFO:root:Epoch[10] Batch[100] Loss[3.537]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[100]	rmse=0.022510	lr=0.399639
[1,0]<stderr>:INFO:root:Epoch[10] Batch[200] Loss[3.451]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[200]	rmse=0.022540	lr=0.399627
[1,0]<stderr>:INFO:root:Epoch[10] Batch[300] Loss[3.211]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[300]	rmse=0.022558	lr=0.399615
[1,0]<stderr>:INFO:root:Epoch[10] Batch[400] Loss[3.461]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[400]	rmse=0.022615	lr=0.399604
[1,0]<stderr>:INFO:root:Epoch[10] Batch[500] Loss[3.595]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[500]	rmse=0.022633	lr=0.399592
[1,0]<stderr>:INFO:root:Epoch[10] Batch[600] Loss[4.230]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[600]	rmse=0.022666	lr=0.399579
[1,0]<stderr>:INFO:root:Epoch[10] Batch[700] Loss[3.641]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[700]	rmse=0.022649	lr=0.399567
[1,0]<stderr>:INFO:root:Epoch[10] Batch[800] Loss[3.503]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[800]	rmse=0.022657	lr=0.399555
[1,0]<stderr>:INFO:root:Epoch[10] Batch[900] Loss[3.414]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[900]	rmse=0.022651	lr=0.399542
[1,0]<stderr>:INFO:root:Epoch[10] Batch[1000] Loss[3.552]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[1000]	rmse=0.022616	lr=0.399529
[1,0]<stderr>:INFO:root:Epoch[10] Batch[1100] Loss[3.272]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[1100]	rmse=0.022630	lr=0.399516
[1,0]<stderr>:INFO:root:Epoch[10] Batch[1200] Loss[5.053]
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[1200]	rmse=0.022605	lr=0.399503
[1,0]<stderr>:INFO:root:Epoch[10] Rank[0] Batch[1251]	Time cost=401.45	Train-metric=0.022596
[1,0]<stderr>:INFO:root:Epoch[10]	Speed: 3190.99 samples/sec
[1,0]<stderr>:INFO:root:Epoch[11] Batch[100] Loss[3.316]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[100]	rmse=0.022474	lr=0.399482
[1,0]<stderr>:INFO:root:Epoch[11] Batch[200] Loss[3.661]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[200]	rmse=0.022433	lr=0.399469
[1,0]<stderr>:INFO:root:Epoch[11] Batch[300] Loss[3.185]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[300]	rmse=0.022512	lr=0.399455
[1,0]<stderr>:INFO:root:Epoch[11] Batch[400] Loss[3.373]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[400]	rmse=0.022498	lr=0.399441
[1,0]<stderr>:INFO:root:Epoch[11] Batch[500] Loss[3.382]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[500]	rmse=0.022454	lr=0.399426
[1,0]<stderr>:INFO:root:Epoch[11] Batch[600] Loss[3.227]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[600]	rmse=0.022425	lr=0.399412
[1,0]<stderr>:INFO:root:Epoch[11] Batch[700] Loss[5.718]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[700]	rmse=0.022460	lr=0.399397
[1,0]<stderr>:INFO:root:Epoch[11] Batch[800] Loss[3.314]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[800]	rmse=0.022441	lr=0.399383
[1,0]<stderr>:INFO:root:Epoch[11] Batch[900] Loss[3.478]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[900]	rmse=0.022465	lr=0.399368
[1,0]<stderr>:INFO:root:Epoch[11] Batch[1000] Loss[3.298]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[1000]	rmse=0.022444	lr=0.399352
[1,0]<stderr>:INFO:root:Epoch[11] Batch[1100] Loss[3.302]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[1100]	rmse=0.022459	lr=0.399337
[1,0]<stderr>:INFO:root:Epoch[11] Batch[1200] Loss[5.339]
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[1200]	rmse=0.022443	lr=0.399322
[1,0]<stderr>:INFO:root:Epoch[11] Rank[0] Batch[1251]	Time cost=398.20	Train-metric=0.022444
[1,0]<stderr>:INFO:root:Epoch[11]	Speed: 3217.05 samples/sec
[1,0]<stderr>:INFO:root:Epoch[12] Batch[100] Loss[3.303]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[100]	rmse=0.022407	lr=0.399298
[1,0]<stderr>:INFO:root:Epoch[12] Batch[200] Loss[3.812]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[200]	rmse=0.022429	lr=0.399282
[1,0]<stderr>:INFO:root:Epoch[12] Batch[300] Loss[3.042]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[300]	rmse=0.022403	lr=0.399266
[1,0]<stderr>:INFO:root:Epoch[12] Batch[400] Loss[3.392]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[400]	rmse=0.022333	lr=0.399249
[1,0]<stderr>:INFO:root:Epoch[12] Batch[500] Loss[3.049]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[500]	rmse=0.022420	lr=0.399233
[1,0]<stderr>:INFO:root:Epoch[12] Batch[600] Loss[3.536]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[600]	rmse=0.022398	lr=0.399216
[1,0]<stderr>:INFO:root:Epoch[12] Batch[700] Loss[3.481]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[700]	rmse=0.022381	lr=0.399200
[1,0]<stderr>:INFO:root:Epoch[12] Batch[800] Loss[3.579]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[800]	rmse=0.022397	lr=0.399182
[1,0]<stderr>:INFO:root:Epoch[12] Batch[900] Loss[5.378]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[900]	rmse=0.022395	lr=0.399165
[1,0]<stderr>:INFO:root:Epoch[12] Batch[1000] Loss[3.516]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[1000]	rmse=0.022421	lr=0.399148
[1,0]<stderr>:INFO:root:Epoch[12] Batch[1100] Loss[3.380]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[1100]	rmse=0.022440	lr=0.399130
[1,0]<stderr>:INFO:root:Epoch[12] Batch[1200] Loss[3.275]
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[1200]	rmse=0.022440	lr=0.399113
[1,0]<stderr>:INFO:root:Epoch[12] Rank[0] Batch[1251]	Time cost=398.92	Train-metric=0.022427
[1,0]<stderr>:INFO:root:Epoch[12]	Speed: 3211.23 samples/sec
[1,0]<stderr>:INFO:root:Epoch[13] Batch[100] Loss[5.545]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[100]	rmse=0.022518	lr=0.399086
[1,0]<stderr>:INFO:root:Epoch[13] Batch[200] Loss[3.106]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[200]	rmse=0.022432	lr=0.399067
[1,0]<stderr>:INFO:root:Epoch[13] Batch[300] Loss[4.162]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[300]	rmse=0.022425	lr=0.399049
[1,0]<stderr>:INFO:root:Epoch[13] Batch[400] Loss[3.275]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[400]	rmse=0.022408	lr=0.399030
[1,0]<stderr>:INFO:root:Epoch[13] Batch[500] Loss[3.476]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[500]	rmse=0.022400	lr=0.399012
[1,0]<stderr>:INFO:root:Epoch[13] Batch[600] Loss[3.433]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[600]	rmse=0.022388	lr=0.398993
[1,0]<stderr>:INFO:root:Epoch[13] Batch[700] Loss[5.330]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[700]	rmse=0.022357	lr=0.398974
[1,0]<stderr>:INFO:root:Epoch[13] Batch[800] Loss[3.449]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[800]	rmse=0.022392	lr=0.398955
[1,0]<stderr>:INFO:root:Epoch[13] Batch[900] Loss[4.113]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[900]	rmse=0.022427	lr=0.398935
[1,0]<stderr>:INFO:root:Epoch[13] Batch[1000] Loss[5.247]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[1000]	rmse=0.022413	lr=0.398916
[1,0]<stderr>:INFO:root:Epoch[13] Batch[1100] Loss[4.209]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[1100]	rmse=0.022366	lr=0.398896
[1,0]<stderr>:INFO:root:Epoch[13] Batch[1200] Loss[3.249]
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[1200]	rmse=0.022372	lr=0.398876
[1,0]<stderr>:INFO:root:Epoch[13] Rank[0] Batch[1251]	Time cost=399.78	Train-metric=0.022372
[1,0]<stderr>:INFO:root:Epoch[13]	Speed: 3204.29 samples/sec
[1,0]<stderr>:INFO:root:Epoch[14] Batch[100] Loss[3.286]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[100]	rmse=0.022107	lr=0.398845
[1,0]<stderr>:INFO:root:Epoch[14] Batch[200] Loss[3.825]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[200]	rmse=0.022128	lr=0.398825
[1,0]<stderr>:INFO:root:Epoch[14] Batch[300] Loss[3.802]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[300]	rmse=0.022189	lr=0.398804
[1,0]<stderr>:INFO:root:Epoch[14] Batch[400] Loss[4.039]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[400]	rmse=0.022144	lr=0.398784
[1,0]<stderr>:INFO:root:Epoch[14] Batch[500] Loss[3.356]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[500]	rmse=0.022214	lr=0.398763
[1,0]<stderr>:INFO:root:Epoch[14] Batch[600] Loss[3.274]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[600]	rmse=0.022236	lr=0.398741
[1,0]<stderr>:INFO:root:Epoch[14] Batch[700] Loss[3.259]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[700]	rmse=0.022243	lr=0.398720
[1,0]<stderr>:INFO:root:Epoch[14] Batch[800] Loss[3.648]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[800]	rmse=0.022238	lr=0.398699
[1,0]<stderr>:INFO:root:Epoch[14] Batch[900] Loss[3.723]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[900]	rmse=0.022239	lr=0.398677
[1,0]<stderr>:INFO:root:Epoch[14] Batch[1000] Loss[5.552]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[1000]	rmse=0.022243	lr=0.398655
[1,0]<stderr>:INFO:root:Epoch[14] Batch[1100] Loss[3.233]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[1100]	rmse=0.022212	lr=0.398633
[1,0]<stderr>:INFO:root:Epoch[14] Batch[1200] Loss[5.406]
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[1200]	rmse=0.022197	lr=0.398611
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0] Batch[1251]	Time cost=399.10	Train-metric=0.022203
[1,0]<stderr>:INFO:root:Epoch[14]	Speed: 3209.79 samples/sec
[1,0]<stderr>:INFO:root:Epoch[14] Rank[0]	Validation-accuracy=0.515760	Validation-top_k_accuracy_5=0.768100
[1,0]<stderr>:INFO:root:Epoch[15] Batch[100] Loss[3.201]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[100]	rmse=0.022317	lr=0.398577
[1,0]<stderr>:INFO:root:Epoch[15] Batch[200] Loss[3.376]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[200]	rmse=0.022214	lr=0.398554
[1,0]<stderr>:INFO:root:Epoch[15] Batch[300] Loss[3.470]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[300]	rmse=0.022243	lr=0.398532
[1,0]<stderr>:INFO:root:Epoch[15] Batch[400] Loss[3.489]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[400]	rmse=0.022248	lr=0.398509
[1,0]<stderr>:INFO:root:Epoch[15] Batch[500] Loss[3.095]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[500]	rmse=0.022209	lr=0.398485
[1,0]<stderr>:INFO:root:Epoch[15] Batch[600] Loss[4.500]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[600]	rmse=0.022174	lr=0.398462
[1,0]<stderr>:INFO:root:Epoch[15] Batch[700] Loss[3.170]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[700]	rmse=0.022161	lr=0.398439
[1,0]<stderr>:INFO:root:Epoch[15] Batch[800] Loss[3.654]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[800]	rmse=0.022142	lr=0.398415
[1,0]<stderr>:INFO:root:Epoch[15] Batch[900] Loss[3.215]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[900]	rmse=0.022121	lr=0.398391
[1,0]<stderr>:INFO:root:Epoch[15] Batch[1000] Loss[3.682]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[1000]	rmse=0.022135	lr=0.398367
[1,0]<stderr>:INFO:root:Epoch[15] Batch[1100] Loss[5.445]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[1100]	rmse=0.022123	lr=0.398343
[1,0]<stderr>:INFO:root:Epoch[15] Batch[1200] Loss[3.763]
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[1200]	rmse=0.022127	lr=0.398318
[1,0]<stderr>:INFO:root:Epoch[15] Rank[0] Batch[1251]	Time cost=401.09	Train-metric=0.022139
[1,0]<stderr>:INFO:root:Epoch[15]	Speed: 3193.85 samples/sec
[1,0]<stderr>:INFO:root:Epoch[16] Batch[100] Loss[3.150]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[100]	rmse=0.022021	lr=0.398281
[1,0]<stderr>:INFO:root:Epoch[16] Batch[200] Loss[3.884]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[200]	rmse=0.021990	lr=0.398256
[1,0]<stderr>:INFO:root:Epoch[16] Batch[300] Loss[5.601]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[300]	rmse=0.022036	lr=0.398231
[1,0]<stderr>:INFO:root:Epoch[16] Batch[400] Loss[3.017]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[400]	rmse=0.022083	lr=0.398206
[1,0]<stderr>:INFO:root:Epoch[16] Batch[500] Loss[4.853]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[500]	rmse=0.022101	lr=0.398181
[1,0]<stderr>:INFO:root:Epoch[16] Batch[600] Loss[3.438]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[600]	rmse=0.022071	lr=0.398155
[1,0]<stderr>:INFO:root:Epoch[16] Batch[700] Loss[3.504]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[700]	rmse=0.022116	lr=0.398129
[1,0]<stderr>:INFO:root:Epoch[16] Batch[800] Loss[4.341]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[800]	rmse=0.022123	lr=0.398103
[1,0]<stderr>:INFO:root:Epoch[16] Batch[900] Loss[3.437]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[900]	rmse=0.022145	lr=0.398077
[1,0]<stderr>:INFO:root:Epoch[16] Batch[1000] Loss[5.461]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[1000]	rmse=0.022107	lr=0.398051
[1,0]<stderr>:INFO:root:Epoch[16] Batch[1100] Loss[3.206]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[1100]	rmse=0.022097	lr=0.398024
[1,0]<stderr>:INFO:root:Epoch[16] Batch[1200] Loss[3.420]
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[1200]	rmse=0.022095	lr=0.397998
[1,0]<stderr>:INFO:root:Epoch[16] Rank[0] Batch[1251]	Time cost=400.70	Train-metric=0.022101
[1,0]<stderr>:INFO:root:Epoch[16]	Speed: 3196.96 samples/sec
[1,0]<stderr>:INFO:root:Epoch[17] Batch[100] Loss[3.242]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[100]	rmse=0.021988	lr=0.397957
[1,0]<stderr>:INFO:root:Epoch[17] Batch[200] Loss[3.223]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[200]	rmse=0.022069	lr=0.397930
[1,0]<stderr>:INFO:root:Epoch[17] Batch[300] Loss[4.792]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[300]	rmse=0.022083	lr=0.397903
[1,0]<stderr>:INFO:root:Epoch[17] Batch[400] Loss[4.180]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[400]	rmse=0.022067	lr=0.397875
[1,0]<stderr>:INFO:root:Epoch[17] Batch[500] Loss[4.909]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[500]	rmse=0.022020	lr=0.397848
[1,0]<stderr>:INFO:root:Epoch[17] Batch[600] Loss[3.771]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[600]	rmse=0.021950	lr=0.397820
[1,0]<stderr>:INFO:root:Epoch[17] Batch[700] Loss[3.207]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[700]	rmse=0.021988	lr=0.397792
[1,0]<stderr>:INFO:root:Epoch[17] Batch[800] Loss[5.239]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[800]	rmse=0.022004	lr=0.397764
[1,0]<stderr>:INFO:root:Epoch[17] Batch[900] Loss[3.770]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[900]	rmse=0.021979	lr=0.397736
[1,0]<stderr>:INFO:root:Epoch[17] Batch[1000] Loss[3.259]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[1000]	rmse=0.021994	lr=0.397707
[1,0]<stderr>:INFO:root:Epoch[17] Batch[1100] Loss[3.649]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[1100]	rmse=0.021984	lr=0.397678
[1,0]<stderr>:INFO:root:Epoch[17] Batch[1200] Loss[3.263]
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[1200]	rmse=0.021987	lr=0.397650
[1,0]<stderr>:INFO:root:Epoch[17] Rank[0] Batch[1251]	Time cost=399.92	Train-metric=0.022010
[1,0]<stderr>:INFO:root:Epoch[17]	Speed: 3203.17 samples/sec
[1,0]<stderr>:INFO:root:Epoch[18] Batch[100] Loss[3.043]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[100]	rmse=0.021863	lr=0.397606
[1,0]<stderr>:INFO:root:Epoch[18] Batch[200] Loss[3.053]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[200]	rmse=0.022005	lr=0.397576
[1,0]<stderr>:INFO:root:Epoch[18] Batch[300] Loss[2.985]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[300]	rmse=0.021939	lr=0.397547
[1,0]<stderr>:INFO:root:Epoch[18] Batch[400] Loss[4.045]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[400]	rmse=0.021948	lr=0.397517
[1,0]<stderr>:INFO:root:Epoch[18] Batch[500] Loss[5.039]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[500]	rmse=0.021975	lr=0.397487
[1,0]<stderr>:INFO:root:Epoch[18] Batch[600] Loss[2.967]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[600]	rmse=0.021975	lr=0.397457
[1,0]<stderr>:INFO:root:Epoch[18] Batch[700] Loss[3.036]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[700]	rmse=0.021987	lr=0.397427
[1,0]<stderr>:INFO:root:Epoch[18] Batch[800] Loss[3.749]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[800]	rmse=0.021974	lr=0.397397
[1,0]<stderr>:INFO:root:Epoch[18] Batch[900] Loss[3.229]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[900]	rmse=0.021985	lr=0.397366
[1,0]<stderr>:INFO:root:Epoch[18] Batch[1000] Loss[5.431]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[1000]	rmse=0.022002	lr=0.397336
[1,0]<stderr>:INFO:root:Epoch[18] Batch[1100] Loss[3.071]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[1100]	rmse=0.022003	lr=0.397305
[1,0]<stderr>:INFO:root:Epoch[18] Batch[1200] Loss[4.766]
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[1200]	rmse=0.021982	lr=0.397274
[1,0]<stderr>:INFO:root:Epoch[18] Rank[0] Batch[1251]	Time cost=398.49	Train-metric=0.021985
[1,0]<stderr>:INFO:root:Epoch[18]	Speed: 3214.67 samples/sec
[1,0]<stderr>:INFO:root:Epoch[19] Batch[100] Loss[4.358]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[100]	rmse=0.021834	lr=0.397226
[1,0]<stderr>:INFO:root:Epoch[19] Batch[200] Loss[3.482]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[200]	rmse=0.021800	lr=0.397195
[1,0]<stderr>:INFO:root:Epoch[19] Batch[300] Loss[5.413]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[300]	rmse=0.021915	lr=0.397163
[1,0]<stderr>:INFO:root:Epoch[19] Batch[400] Loss[3.350]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[400]	rmse=0.021878	lr=0.397131
[1,0]<stderr>:INFO:root:Epoch[19] Batch[500] Loss[3.488]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[500]	rmse=0.021988	lr=0.397099
[1,0]<stderr>:INFO:root:Epoch[19] Batch[600] Loss[4.054]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[600]	rmse=0.021979	lr=0.397067
[1,0]<stderr>:INFO:root:Epoch[19] Batch[700] Loss[4.175]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[700]	rmse=0.021950	lr=0.397035
[1,0]<stderr>:INFO:root:Epoch[19] Batch[800] Loss[5.061]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[800]	rmse=0.021959	lr=0.397002
[1,0]<stderr>:INFO:root:Epoch[19] Batch[900] Loss[3.687]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[900]	rmse=0.021938	lr=0.396969
[1,0]<stderr>:INFO:root:Epoch[19] Batch[1000] Loss[4.217]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[1000]	rmse=0.021934	lr=0.396936
[1,0]<stderr>:INFO:root:Epoch[19] Batch[1100] Loss[3.307]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[1100]	rmse=0.021942	lr=0.396903
[1,0]<stderr>:INFO:root:Epoch[19] Batch[1200] Loss[5.301]
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[1200]	rmse=0.021950	lr=0.396870
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0] Batch[1251]	Time cost=398.56	Train-metric=0.021955
[1,0]<stderr>:INFO:root:Epoch[19]	Speed: 3214.15 samples/sec
[1,0]<stderr>:INFO:root:Epoch[19] Rank[0]	Validation-accuracy=0.520360	Validation-top_k_accuracy_5=0.771700
[1,0]<stderr>:INFO:root:Epoch[20] Batch[100] Loss[5.320]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[100]	rmse=0.021871	lr=0.396819
[1,0]<stderr>:INFO:root:Epoch[20] Batch[200] Loss[4.256]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[200]	rmse=0.021810	lr=0.396786
[1,0]<stderr>:INFO:root:Epoch[20] Batch[300] Loss[3.095]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[300]	rmse=0.021807	lr=0.396752
[1,0]<stderr>:INFO:root:Epoch[20] Batch[400] Loss[4.537]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[400]	rmse=0.021855	lr=0.396718
[1,0]<stderr>:INFO:root:Epoch[20] Batch[500] Loss[3.472]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[500]	rmse=0.021861	lr=0.396683
[1,0]<stderr>:INFO:root:Epoch[20] Batch[600] Loss[3.335]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[600]	rmse=0.021881	lr=0.396649
[1,0]<stderr>:INFO:root:Epoch[20] Batch[700] Loss[3.083]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[700]	rmse=0.021902	lr=0.396614
[1,0]<stderr>:INFO:root:Epoch[20] Batch[800] Loss[2.994]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[800]	rmse=0.021921	lr=0.396579
[1,0]<stderr>:INFO:root:Epoch[20] Batch[900] Loss[3.718]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[900]	rmse=0.021925	lr=0.396544
[1,0]<stderr>:INFO:root:Epoch[20] Batch[1000] Loss[5.136]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[1000]	rmse=0.021916	lr=0.396509
[1,0]<stderr>:INFO:root:Epoch[20] Batch[1100] Loss[2.796]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[1100]	rmse=0.021905	lr=0.396474
[1,0]<stderr>:INFO:root:Epoch[20] Batch[1200] Loss[3.127]
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[1200]	rmse=0.021907	lr=0.396439
[1,0]<stderr>:INFO:root:Epoch[20] Rank[0] Batch[1251]	Time cost=398.88	Train-metric=0.021911
[1,0]<stderr>:INFO:root:Epoch[20]	Speed: 3211.57 samples/sec
[1,0]<stderr>:INFO:root:Epoch[21] Batch[100] Loss[2.865]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[100]	rmse=0.021725	lr=0.396385
[1,0]<stderr>:INFO:root:Epoch[21] Batch[200] Loss[4.129]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[200]	rmse=0.021709	lr=0.396349
[1,0]<stderr>:INFO:root:Epoch[21] Batch[300] Loss[3.403]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[300]	rmse=0.021724	lr=0.396313
[1,0]<stderr>:INFO:root:Epoch[21] Batch[400] Loss[3.354]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[400]	rmse=0.021786	lr=0.396276
[1,0]<stderr>:INFO:root:Epoch[21] Batch[500] Loss[4.422]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[500]	rmse=0.021802	lr=0.396240
[1,0]<stderr>:INFO:root:Epoch[21] Batch[600] Loss[4.552]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[600]	rmse=0.021832	lr=0.396203
[1,0]<stderr>:INFO:root:Epoch[21] Batch[700] Loss[5.124]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[700]	rmse=0.021880	lr=0.396166
[1,0]<stderr>:INFO:root:Epoch[21] Batch[800] Loss[3.032]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[800]	rmse=0.021924	lr=0.396129
[1,0]<stderr>:INFO:root:Epoch[21] Batch[900] Loss[4.796]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[900]	rmse=0.021964	lr=0.396092
[1,0]<stderr>:INFO:root:Epoch[21] Batch[1000] Loss[3.081]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[1000]	rmse=0.021949	lr=0.396055
[1,0]<stderr>:INFO:root:Epoch[21] Batch[1100] Loss[4.386]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[1100]	rmse=0.021940	lr=0.396017
[1,0]<stderr>:INFO:root:Epoch[21] Batch[1200] Loss[3.084]
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[1200]	rmse=0.021947	lr=0.395980
[1,0]<stderr>:INFO:root:Epoch[21] Rank[0] Batch[1251]	Time cost=399.05	Train-metric=0.021936
[1,0]<stderr>:INFO:root:Epoch[21]	Speed: 3210.16 samples/sec
[1,0]<stderr>:INFO:root:Epoch[22] Batch[100] Loss[4.694]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[100]	rmse=0.021702	lr=0.395922
[1,0]<stderr>:INFO:root:Epoch[22] Batch[200] Loss[3.172]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[200]	rmse=0.021783	lr=0.395884
[1,0]<stderr>:INFO:root:Epoch[22] Batch[300] Loss[5.290]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[300]	rmse=0.021784	lr=0.395846
[1,0]<stderr>:INFO:root:Epoch[22] Batch[400] Loss[5.300]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[400]	rmse=0.021820	lr=0.395807
[1,0]<stderr>:INFO:root:Epoch[22] Batch[500] Loss[5.284]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[500]	rmse=0.021811	lr=0.395769
[1,0]<stderr>:INFO:root:Epoch[22] Batch[600] Loss[2.996]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[600]	rmse=0.021833	lr=0.395730
[1,0]<stderr>:INFO:root:Epoch[22] Batch[700] Loss[5.554]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[700]	rmse=0.021843	lr=0.395691
[1,0]<stderr>:INFO:root:Epoch[22] Batch[800] Loss[3.022]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[800]	rmse=0.021845	lr=0.395652
[1,0]<stderr>:INFO:root:Epoch[22] Batch[900] Loss[5.453]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[900]	rmse=0.021843	lr=0.395612
[1,0]<stderr>:INFO:root:Epoch[22] Batch[1000] Loss[2.950]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[1000]	rmse=0.021851	lr=0.395573
[1,0]<stderr>:INFO:root:Epoch[22] Batch[1100] Loss[3.818]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[1100]	rmse=0.021845	lr=0.395533
[1,0]<stderr>:INFO:root:Epoch[22] Batch[1200] Loss[3.454]
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[1200]	rmse=0.021867	lr=0.395493
[1,0]<stderr>:INFO:root:Epoch[22] Rank[0] Batch[1251]	Time cost=398.56	Train-metric=0.021867
[1,0]<stderr>:INFO:root:Epoch[22]	Speed: 3214.13 samples/sec
[1,0]<stderr>:INFO:root:Epoch[23] Batch[100] Loss[2.999]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[100]	rmse=0.022039	lr=0.395433
[1,0]<stderr>:INFO:root:Epoch[23] Batch[200] Loss[5.396]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[200]	rmse=0.021939	lr=0.395392
[1,0]<stderr>:INFO:root:Epoch[23] Batch[300] Loss[3.292]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[300]	rmse=0.021934	lr=0.395352
[1,0]<stderr>:INFO:root:Epoch[23] Batch[400] Loss[3.039]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[400]	rmse=0.021894	lr=0.395311
[1,0]<stderr>:INFO:root:Epoch[23] Batch[500] Loss[3.190]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[500]	rmse=0.021860	lr=0.395270
[1,0]<stderr>:INFO:root:Epoch[23] Batch[600] Loss[3.744]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[600]	rmse=0.021849	lr=0.395229
[1,0]<stderr>:INFO:root:Epoch[23] Batch[700] Loss[3.085]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[700]	rmse=0.021849	lr=0.395188
[1,0]<stderr>:INFO:root:Epoch[23] Batch[800] Loss[5.211]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[800]	rmse=0.021861	lr=0.395147
[1,0]<stderr>:INFO:root:Epoch[23] Batch[900] Loss[3.249]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[900]	rmse=0.021846	lr=0.395105
[1,0]<stderr>:INFO:root:Epoch[23] Batch[1000] Loss[3.691]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[1000]	rmse=0.021847	lr=0.395063
[1,0]<stderr>:INFO:root:Epoch[23] Batch[1100] Loss[3.050]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[1100]	rmse=0.021846	lr=0.395022
[1,0]<stderr>:INFO:root:Epoch[23] Batch[1200] Loss[3.155]
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[1200]	rmse=0.021863	lr=0.394980
[1,0]<stderr>:INFO:root:Epoch[23] Rank[0] Batch[1251]	Time cost=399.75	Train-metric=0.021868
[1,0]<stderr>:INFO:root:Epoch[23]	Speed: 3204.53 samples/sec
[1,0]<stderr>:INFO:root:Epoch[24] Batch[100] Loss[5.036]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[100]	rmse=0.021684	lr=0.394916
[1,0]<stderr>:INFO:root:Epoch[24] Batch[200] Loss[3.987]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[200]	rmse=0.021698	lr=0.394873
[1,0]<stderr>:INFO:root:Epoch[24] Batch[300] Loss[3.061]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[300]	rmse=0.021754	lr=0.394830
[1,0]<stderr>:INFO:root:Epoch[24] Batch[400] Loss[5.363]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[400]	rmse=0.021748	lr=0.394788
[1,0]<stderr>:INFO:root:Epoch[24] Batch[500] Loss[3.378]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[500]	rmse=0.021795	lr=0.394745
[1,0]<stderr>:INFO:root:Epoch[24] Batch[600] Loss[3.448]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[600]	rmse=0.021809	lr=0.394701
[1,0]<stderr>:INFO:root:Epoch[24] Batch[700] Loss[3.718]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[700]	rmse=0.021783	lr=0.394658
[1,0]<stderr>:INFO:root:Epoch[24] Batch[800] Loss[3.143]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[800]	rmse=0.021799	lr=0.394614
[1,0]<stderr>:INFO:root:Epoch[24] Batch[900] Loss[3.238]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[900]	rmse=0.021799	lr=0.394571
[1,0]<stderr>:INFO:root:Epoch[24] Batch[1000] Loss[3.088]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[1000]	rmse=0.021813	lr=0.394527
[1,0]<stderr>:INFO:root:Epoch[24] Batch[1100] Loss[4.933]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[1100]	rmse=0.021819	lr=0.394483
[1,0]<stderr>:INFO:root:Epoch[24] Batch[1200] Loss[3.643]
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[1200]	rmse=0.021828	lr=0.394438
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0] Batch[1251]	Time cost=397.96	Train-metric=0.021828
[1,0]<stderr>:INFO:root:Epoch[24]	Speed: 3218.94 samples/sec
[1,0]<stderr>:INFO:root:Epoch[24] Rank[0]	Validation-accuracy=0.531960	Validation-top_k_accuracy_5=0.786180
[1,0]<stderr>:INFO:root:Epoch[25] Batch[100] Loss[3.232]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[100]	rmse=0.021801	lr=0.394371
[1,0]<stderr>:INFO:root:Epoch[25] Batch[200] Loss[3.391]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[200]	rmse=0.021717	lr=0.394326
[1,0]<stderr>:INFO:root:Epoch[25] Batch[300] Loss[5.224]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[300]	rmse=0.021775	lr=0.394282
[1,0]<stderr>:INFO:root:Epoch[25] Batch[400] Loss[3.211]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[400]	rmse=0.021759	lr=0.394237
[1,0]<stderr>:INFO:root:Epoch[25] Batch[500] Loss[3.630]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[500]	rmse=0.021729	lr=0.394191
[1,0]<stderr>:INFO:root:Epoch[25] Batch[600] Loss[4.279]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[600]	rmse=0.021727	lr=0.394146
[1,0]<stderr>:INFO:root:Epoch[25] Batch[700] Loss[3.589]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[700]	rmse=0.021715	lr=0.394100
[1,0]<stderr>:INFO:root:Epoch[25] Batch[800] Loss[4.696]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[800]	rmse=0.021719	lr=0.394055
[1,0]<stderr>:INFO:root:Epoch[25] Batch[900] Loss[3.439]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[900]	rmse=0.021745	lr=0.394009
[1,0]<stderr>:INFO:root:Epoch[25] Batch[1000] Loss[3.385]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[1000]	rmse=0.021767	lr=0.393963
[1,0]<stderr>:INFO:root:Epoch[25] Batch[1100] Loss[3.410]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[1100]	rmse=0.021769	lr=0.393916
[1,0]<stderr>:INFO:root:Epoch[25] Batch[1200] Loss[3.544]
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[1200]	rmse=0.021775	lr=0.393870
[1,0]<stderr>:INFO:root:Epoch[25] Rank[0] Batch[1251]	Time cost=400.61	Train-metric=0.021783
[1,0]<stderr>:INFO:root:Epoch[25]	Speed: 3197.65 samples/sec
[1,0]<stderr>:INFO:root:Epoch[26] Batch[100] Loss[4.576]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[100]	rmse=0.021780	lr=0.393799
[1,0]<stderr>:INFO:root:Epoch[26] Batch[200] Loss[4.336]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[200]	rmse=0.021708	lr=0.393753
[1,0]<stderr>:INFO:root:Epoch[26] Batch[300] Loss[3.887]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[300]	rmse=0.021696	lr=0.393706
[1,0]<stderr>:INFO:root:Epoch[26] Batch[400] Loss[3.050]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[400]	rmse=0.021691	lr=0.393658
[1,0]<stderr>:INFO:root:Epoch[26] Batch[500] Loss[3.177]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[500]	rmse=0.021727	lr=0.393611
[1,0]<stderr>:INFO:root:Epoch[26] Batch[600] Loss[3.246]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[600]	rmse=0.021715	lr=0.393563
[1,0]<stderr>:INFO:root:Epoch[26] Batch[700] Loss[3.092]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[700]	rmse=0.021703	lr=0.393516
[1,0]<stderr>:INFO:root:Epoch[26] Batch[800] Loss[4.335]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[800]	rmse=0.021717	lr=0.393468
[1,0]<stderr>:INFO:root:Epoch[26] Batch[900] Loss[3.080]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[900]	rmse=0.021710	lr=0.393420
[1,0]<stderr>:INFO:root:Epoch[26] Batch[1000] Loss[4.177]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[1000]	rmse=0.021726	lr=0.393371
[1,0]<stderr>:INFO:root:Epoch[26] Batch[1100] Loss[3.319]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[1100]	rmse=0.021732	lr=0.393323
[1,0]<stderr>:INFO:root:Epoch[26] Batch[1200] Loss[5.379]
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[1200]	rmse=0.021722	lr=0.393274
[1,0]<stderr>:INFO:root:Epoch[26] Rank[0] Batch[1251]	Time cost=401.40	Train-metric=0.021729
[1,0]<stderr>:INFO:root:Epoch[26]	Speed: 3191.40 samples/sec
[1,0]<stderr>:INFO:root:Epoch[27] Batch[100] Loss[3.422]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[100]	rmse=0.021691	lr=0.393201
[1,0]<stderr>:INFO:root:Epoch[27] Batch[200] Loss[3.013]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[200]	rmse=0.021716	lr=0.393151
[1,0]<stderr>:INFO:root:Epoch[27] Batch[300] Loss[4.942]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[300]	rmse=0.021790	lr=0.393102
[1,0]<stderr>:INFO:root:Epoch[27] Batch[400] Loss[2.949]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[400]	rmse=0.021801	lr=0.393053
[1,0]<stderr>:INFO:root:Epoch[27] Batch[500] Loss[5.223]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[500]	rmse=0.021791	lr=0.393003
[1,0]<stderr>:INFO:root:Epoch[27] Batch[600] Loss[3.337]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[600]	rmse=0.021785	lr=0.392954
[1,0]<stderr>:INFO:root:Epoch[27] Batch[700] Loss[4.276]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[700]	rmse=0.021793	lr=0.392904
[1,0]<stderr>:INFO:root:Epoch[27] Batch[800] Loss[3.160]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[800]	rmse=0.021771	lr=0.392854
[1,0]<stderr>:INFO:root:Epoch[27] Batch[900] Loss[3.163]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[900]	rmse=0.021785	lr=0.392803
[1,0]<stderr>:INFO:root:Epoch[27] Batch[1000] Loss[4.486]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[1000]	rmse=0.021788	lr=0.392753
[1,0]<stderr>:INFO:root:Epoch[27] Batch[1100] Loss[3.190]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[1100]	rmse=0.021798	lr=0.392702
[1,0]<stderr>:INFO:root:Epoch[27] Batch[1200] Loss[3.578]
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[1200]	rmse=0.021798	lr=0.392651
[1,0]<stderr>:INFO:root:Epoch[27] Rank[0] Batch[1251]	Time cost=401.40	Train-metric=0.021795
[1,0]<stderr>:INFO:root:Epoch[27]	Speed: 3191.42 samples/sec
[1,0]<stderr>:INFO:root:Epoch[28] Batch[100] Loss[5.269]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[100]	rmse=0.021677	lr=0.392574
[1,0]<stderr>:INFO:root:Epoch[28] Batch[200] Loss[2.825]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[200]	rmse=0.021690	lr=0.392523
[1,0]<stderr>:INFO:root:Epoch[28] Batch[300] Loss[3.059]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[300]	rmse=0.021718	lr=0.392472
[1,0]<stderr>:INFO:root:Epoch[28] Batch[400] Loss[2.957]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[400]	rmse=0.021699	lr=0.392420
[1,0]<stderr>:INFO:root:Epoch[28] Batch[500] Loss[5.123]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[500]	rmse=0.021707	lr=0.392369
[1,0]<stderr>:INFO:root:Epoch[28] Batch[600] Loss[3.092]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[600]	rmse=0.021719	lr=0.392317
[1,0]<stderr>:INFO:root:Epoch[28] Batch[700] Loss[2.837]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[700]	rmse=0.021707	lr=0.392265
[1,0]<stderr>:INFO:root:Epoch[28] Batch[800] Loss[3.152]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[800]	rmse=0.021726	lr=0.392212
[1,0]<stderr>:INFO:root:Epoch[28] Batch[900] Loss[4.948]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[900]	rmse=0.021745	lr=0.392160
[1,0]<stderr>:INFO:root:Epoch[28] Batch[1000] Loss[3.132]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[1000]	rmse=0.021750	lr=0.392107
[1,0]<stderr>:INFO:root:Epoch[28] Batch[1100] Loss[3.140]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[1100]	rmse=0.021757	lr=0.392055
[1,0]<stderr>:INFO:root:Epoch[28] Batch[1200] Loss[4.988]
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[1200]	rmse=0.021743	lr=0.392002
[1,0]<stderr>:INFO:root:Epoch[28] Rank[0] Batch[1251]	Time cost=399.29	Train-metric=0.021753
[1,0]<stderr>:INFO:root:Epoch[28]	Speed: 3208.22 samples/sec
[1,0]<stderr>:INFO:root:Epoch[29] Batch[100] Loss[3.385]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[100]	rmse=0.021609	lr=0.391921
[1,0]<stderr>:INFO:root:Epoch[29] Batch[200] Loss[3.850]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[200]	rmse=0.021547	lr=0.391868
[1,0]<stderr>:INFO:root:Epoch[29] Batch[300] Loss[3.472]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[300]	rmse=0.021542	lr=0.391815
[1,0]<stderr>:INFO:root:Epoch[29] Batch[400] Loss[4.938]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[400]	rmse=0.021587	lr=0.391761
[1,0]<stderr>:INFO:root:Epoch[29] Batch[500] Loss[2.865]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[500]	rmse=0.021614	lr=0.391707
[1,0]<stderr>:INFO:root:Epoch[29] Batch[600] Loss[3.397]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[600]	rmse=0.021649	lr=0.391653
[1,0]<stderr>:INFO:root:Epoch[29] Batch[700] Loss[2.957]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[700]	rmse=0.021642	lr=0.391599
[1,0]<stderr>:INFO:root:Epoch[29] Batch[800] Loss[5.390]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[800]	rmse=0.021642	lr=0.391544
[1,0]<stderr>:INFO:root:Epoch[29] Batch[900] Loss[3.185]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[900]	rmse=0.021666	lr=0.391490
[1,0]<stderr>:INFO:root:Epoch[29] Batch[1000] Loss[4.188]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[1000]	rmse=0.021668	lr=0.391435
[1,0]<stderr>:INFO:root:Epoch[29] Batch[1100] Loss[3.405]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[1100]	rmse=0.021664	lr=0.391380
[1,0]<stderr>:INFO:root:Epoch[29] Batch[1200] Loss[5.060]
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[1200]	rmse=0.021672	lr=0.391325
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0] Batch[1251]	Time cost=399.33	Train-metric=0.021676
[1,0]<stderr>:INFO:root:Epoch[29]	Speed: 3207.90 samples/sec
[1,0]<stderr>:INFO:root:Epoch[29] Rank[0]	Validation-accuracy=0.528540	Validation-top_k_accuracy_5=0.779280
[1,0]<stderr>:INFO:root:Epoch[30] Batch[100] Loss[3.367]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[100]	rmse=0.021622	lr=0.391241
[1,0]<stderr>:INFO:root:Epoch[30] Batch[200] Loss[3.122]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[200]	rmse=0.021615	lr=0.391186
[1,0]<stderr>:INFO:root:Epoch[30] Batch[300] Loss[3.229]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[300]	rmse=0.021658	lr=0.391130
[1,0]<stderr>:INFO:root:Epoch[30] Batch[400] Loss[3.281]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[400]	rmse=0.021680	lr=0.391074
[1,0]<stderr>:INFO:root:Epoch[30] Batch[500] Loss[2.971]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[500]	rmse=0.021671	lr=0.391018
[1,0]<stderr>:INFO:root:Epoch[30] Batch[600] Loss[3.153]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[600]	rmse=0.021644	lr=0.390962
[1,0]<stderr>:INFO:root:Epoch[30] Batch[700] Loss[3.551]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[700]	rmse=0.021703	lr=0.390906
[1,0]<stderr>:INFO:root:Epoch[30] Batch[800] Loss[5.289]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[800]	rmse=0.021697	lr=0.390849
[1,0]<stderr>:INFO:root:Epoch[30] Batch[900] Loss[2.833]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[900]	rmse=0.021680	lr=0.390793
[1,0]<stderr>:INFO:root:Epoch[30] Batch[1000] Loss[4.397]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[1000]	rmse=0.021655	lr=0.390736
[1,0]<stderr>:INFO:root:Epoch[30] Batch[1100] Loss[2.828]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[1100]	rmse=0.021675	lr=0.390679
[1,0]<stderr>:INFO:root:Epoch[30] Batch[1200] Loss[3.132]
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[1200]	rmse=0.021669	lr=0.390621
[1,0]<stderr>:INFO:root:Epoch[30] Rank[0] Batch[1251]	Time cost=399.77	Train-metric=0.021684
[1,0]<stderr>:INFO:root:Epoch[30]	Speed: 3204.37 samples/sec
[1,0]<stderr>:INFO:root:Epoch[31] Batch[100] Loss[5.485]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[100]	rmse=0.021377	lr=0.390535
[1,0]<stderr>:INFO:root:Epoch[31] Batch[200] Loss[3.068]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[200]	rmse=0.021335	lr=0.390477
[1,0]<stderr>:INFO:root:Epoch[31] Batch[300] Loss[3.336]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[300]	rmse=0.021432	lr=0.390419
[1,0]<stderr>:INFO:root:Epoch[31] Batch[400] Loss[4.108]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[400]	rmse=0.021479	lr=0.390361
[1,0]<stderr>:INFO:root:Epoch[31] Batch[500] Loss[5.506]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[500]	rmse=0.021523	lr=0.390303
[1,0]<stderr>:INFO:root:Epoch[31] Batch[600] Loss[3.279]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[600]	rmse=0.021554	lr=0.390245
[1,0]<stderr>:INFO:root:Epoch[31] Batch[700] Loss[3.336]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[700]	rmse=0.021600	lr=0.390186
[1,0]<stderr>:INFO:root:Epoch[31] Batch[800] Loss[3.369]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[800]	rmse=0.021625	lr=0.390127
[1,0]<stderr>:INFO:root:Epoch[31] Batch[900] Loss[5.516]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[900]	rmse=0.021647	lr=0.390069
[1,0]<stderr>:INFO:root:Epoch[31] Batch[1000] Loss[2.971]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[1000]	rmse=0.021668	lr=0.390010
[1,0]<stderr>:INFO:root:Epoch[31] Batch[1100] Loss[5.113]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[1100]	rmse=0.021671	lr=0.389950
[1,0]<stderr>:INFO:root:Epoch[31] Batch[1200] Loss[3.086]
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[1200]	rmse=0.021680	lr=0.389891
[1,0]<stderr>:INFO:root:Epoch[31] Rank[0] Batch[1251]	Time cost=398.77	Train-metric=0.021670
[1,0]<stderr>:INFO:root:Epoch[31]	Speed: 3212.43 samples/sec
[1,0]<stderr>:INFO:root:Epoch[32] Batch[100] Loss[2.988]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[100]	rmse=0.021437	lr=0.389801
[1,0]<stderr>:INFO:root:Epoch[32] Batch[200] Loss[5.326]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[200]	rmse=0.021553	lr=0.389741
[1,0]<stderr>:INFO:root:Epoch[32] Batch[300] Loss[3.390]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[300]	rmse=0.021707	lr=0.389681
[1,0]<stderr>:INFO:root:Epoch[32] Batch[400] Loss[3.074]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[400]	rmse=0.021672	lr=0.389621
[1,0]<stderr>:INFO:root:Epoch[32] Batch[500] Loss[3.141]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[500]	rmse=0.021618	lr=0.389561
[1,0]<stderr>:INFO:root:Epoch[32] Batch[600] Loss[4.619]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[600]	rmse=0.021597	lr=0.389500
[1,0]<stderr>:INFO:root:Epoch[32] Batch[700] Loss[4.071]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[700]	rmse=0.021616	lr=0.389440
[1,0]<stderr>:INFO:root:Epoch[32] Batch[800] Loss[3.171]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[800]	rmse=0.021625	lr=0.389379
[1,0]<stderr>:INFO:root:Epoch[32] Batch[900] Loss[3.087]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[900]	rmse=0.021655	lr=0.389318
[1,0]<stderr>:INFO:root:Epoch[32] Batch[1000] Loss[3.725]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[1000]	rmse=0.021665	lr=0.389257
[1,0]<stderr>:INFO:root:Epoch[32] Batch[1100] Loss[3.324]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[1100]	rmse=0.021667	lr=0.389196
[1,0]<stderr>:INFO:root:Epoch[32] Batch[1200] Loss[3.290]
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[1200]	rmse=0.021673	lr=0.389134
[1,0]<stderr>:INFO:root:Epoch[32] Rank[0] Batch[1251]	Time cost=399.25	Train-metric=0.021692
[1,0]<stderr>:INFO:root:Epoch[32]	Speed: 3208.57 samples/sec
[1,0]<stderr>:INFO:root:Epoch[33] Batch[100] Loss[3.613]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[100]	rmse=0.021639	lr=0.389041
[1,0]<stderr>:INFO:root:Epoch[33] Batch[200] Loss[3.054]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[200]	rmse=0.021776	lr=0.388979
[1,0]<stderr>:INFO:root:Epoch[33] Batch[300] Loss[2.739]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[300]	rmse=0.021620	lr=0.388917
[1,0]<stderr>:INFO:root:Epoch[33] Batch[400] Loss[3.896]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[400]	rmse=0.021627	lr=0.388855
[1,0]<stderr>:INFO:root:Epoch[33] Batch[500] Loss[3.448]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[500]	rmse=0.021665	lr=0.388792
[1,0]<stderr>:INFO:root:Epoch[33] Batch[600] Loss[2.614]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[600]	rmse=0.021650	lr=0.388730
[1,0]<stderr>:INFO:root:Epoch[33] Batch[700] Loss[3.200]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[700]	rmse=0.021649	lr=0.388667
[1,0]<stderr>:INFO:root:Epoch[33] Batch[800] Loss[4.616]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[800]	rmse=0.021658	lr=0.388604
[1,0]<stderr>:INFO:root:Epoch[33] Batch[900] Loss[3.096]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[900]	rmse=0.021667	lr=0.388541
[1,0]<stderr>:INFO:root:Epoch[33] Batch[1000] Loss[3.212]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[1000]	rmse=0.021645	lr=0.388478
[1,0]<stderr>:INFO:root:Epoch[33] Batch[1100] Loss[4.747]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[1100]	rmse=0.021634	lr=0.388414
[1,0]<stderr>:INFO:root:Epoch[33] Batch[1200] Loss[5.062]
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[1200]	rmse=0.021640	lr=0.388351
[1,0]<stderr>:INFO:root:Epoch[33] Rank[0] Batch[1251]	Time cost=400.03	Train-metric=0.021633
[1,0]<stderr>:INFO:root:Epoch[33]	Speed: 3202.30 samples/sec
[1,0]<stderr>:INFO:root:Epoch[34] Batch[100] Loss[3.578]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[100]	rmse=0.021445	lr=0.388254
[1,0]<stderr>:INFO:root:Epoch[34] Batch[200] Loss[3.028]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[200]	rmse=0.021628	lr=0.388190
[1,0]<stderr>:INFO:root:Epoch[34] Batch[300] Loss[5.139]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[300]	rmse=0.021581	lr=0.388126
[1,0]<stderr>:INFO:root:Epoch[34] Batch[400] Loss[3.252]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[400]	rmse=0.021562	lr=0.388062
[1,0]<stderr>:INFO:root:Epoch[34] Batch[500] Loss[2.988]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[500]	rmse=0.021583	lr=0.387997
[1,0]<stderr>:INFO:root:Epoch[34] Batch[600] Loss[3.402]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[600]	rmse=0.021617	lr=0.387932
[1,0]<stderr>:INFO:root:Epoch[34] Batch[700] Loss[3.462]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[700]	rmse=0.021659	lr=0.387867
[1,0]<stderr>:INFO:root:Epoch[34] Batch[800] Loss[4.733]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[800]	rmse=0.021652	lr=0.387802
[1,0]<stderr>:INFO:root:Epoch[34] Batch[900] Loss[2.974]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[900]	rmse=0.021657	lr=0.387737
[1,0]<stderr>:INFO:root:Epoch[34] Batch[1000] Loss[3.122]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[1000]	rmse=0.021657	lr=0.387672
[1,0]<stderr>:INFO:root:Epoch[34] Batch[1100] Loss[2.892]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[1100]	rmse=0.021660	lr=0.387606
[1,0]<stderr>:INFO:root:Epoch[34] Batch[1200] Loss[3.926]
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[1200]	rmse=0.021658	lr=0.387541
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0] Batch[1251]	Time cost=398.93	Train-metric=0.021650
[1,0]<stderr>:INFO:root:Epoch[34]	Speed: 3211.12 samples/sec
[1,0]<stderr>:INFO:root:Epoch[34] Rank[0]	Validation-accuracy=0.527620	Validation-top_k_accuracy_5=0.777500
[1,0]<stderr>:INFO:root:Epoch[35] Batch[100] Loss[3.109]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[100]	rmse=0.021580	lr=0.387441
[1,0]<stderr>:INFO:root:Epoch[35] Batch[200] Loss[3.179]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[200]	rmse=0.021598	lr=0.387375
[1,0]<stderr>:INFO:root:Epoch[35] Batch[300] Loss[4.290]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[300]	rmse=0.021584	lr=0.387309
[1,0]<stderr>:INFO:root:Epoch[35] Batch[400] Loss[5.357]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[400]	rmse=0.021572	lr=0.387242
[1,0]<stderr>:INFO:root:Epoch[35] Batch[500] Loss[4.644]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[500]	rmse=0.021592	lr=0.387175
[1,0]<stderr>:INFO:root:Epoch[35] Batch[600] Loss[4.162]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[600]	rmse=0.021612	lr=0.387109
[1,0]<stderr>:INFO:root:Epoch[35] Batch[700] Loss[3.281]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[700]	rmse=0.021628	lr=0.387042
[1,0]<stderr>:INFO:root:Epoch[35] Batch[800] Loss[2.848]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[800]	rmse=0.021617	lr=0.386975
[1,0]<stderr>:INFO:root:Epoch[35] Batch[900] Loss[3.067]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[900]	rmse=0.021625	lr=0.386907
[1,0]<stderr>:INFO:root:Epoch[35] Batch[1000] Loss[5.481]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[1000]	rmse=0.021626	lr=0.386840
[1,0]<stderr>:INFO:root:Epoch[35] Batch[1100] Loss[5.499]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[1100]	rmse=0.021634	lr=0.386772
[1,0]<stderr>:INFO:root:Epoch[35] Batch[1200] Loss[2.835]
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[1200]	rmse=0.021643	lr=0.386704
[1,0]<stderr>:INFO:root:Epoch[35] Rank[0] Batch[1251]	Time cost=398.47	Train-metric=0.021646
[1,0]<stderr>:INFO:root:Epoch[35]	Speed: 3214.83 samples/sec
[1,0]<stderr>:INFO:root:Epoch[36] Batch[100] Loss[3.050]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[100]	rmse=0.021742	lr=0.386602
[1,0]<stderr>:INFO:root:Epoch[36] Batch[200] Loss[3.186]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[200]	rmse=0.021698	lr=0.386533
[1,0]<stderr>:INFO:root:Epoch[36] Batch[300] Loss[3.596]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[300]	rmse=0.021570	lr=0.386465
[1,0]<stderr>:INFO:root:Epoch[36] Batch[400] Loss[3.081]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[400]	rmse=0.021617	lr=0.386396
[1,0]<stderr>:INFO:root:Epoch[36] Batch[500] Loss[3.158]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[500]	rmse=0.021630	lr=0.386328
[1,0]<stderr>:INFO:root:Epoch[36] Batch[600] Loss[3.006]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[600]	rmse=0.021616	lr=0.386259
[1,0]<stderr>:INFO:root:Epoch[36] Batch[700] Loss[2.765]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[700]	rmse=0.021599	lr=0.386190
[1,0]<stderr>:INFO:root:Epoch[36] Batch[800] Loss[3.239]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[800]	rmse=0.021606	lr=0.386120
[1,0]<stderr>:INFO:root:Epoch[36] Batch[900] Loss[4.161]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[900]	rmse=0.021610	lr=0.386051
[1,0]<stderr>:INFO:root:Epoch[36] Batch[1000] Loss[3.293]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[1000]	rmse=0.021631	lr=0.385981
[1,0]<stderr>:INFO:root:Epoch[36] Batch[1100] Loss[5.489]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[1100]	rmse=0.021611	lr=0.385912
[1,0]<stderr>:INFO:root:Epoch[36] Batch[1200] Loss[3.092]
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[1200]	rmse=0.021618	lr=0.385842
[1,0]<stderr>:INFO:root:Epoch[36] Rank[0] Batch[1251]	Time cost=398.76	Train-metric=0.021616
[1,0]<stderr>:INFO:root:Epoch[36]	Speed: 3212.56 samples/sec
[1,0]<stderr>:INFO:root:Epoch[37] Batch[100] Loss[4.797]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[100]	rmse=0.021470	lr=0.385736
[1,0]<stderr>:INFO:root:Epoch[37] Batch[200] Loss[3.284]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[200]	rmse=0.021568	lr=0.385666
[1,0]<stderr>:INFO:root:Epoch[37] Batch[300] Loss[3.616]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[300]	rmse=0.021561	lr=0.385595
[1,0]<stderr>:INFO:root:Epoch[37] Batch[400] Loss[3.180]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[400]	rmse=0.021581	lr=0.385525
[1,0]<stderr>:INFO:root:Epoch[37] Batch[500] Loss[2.869]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[500]	rmse=0.021610	lr=0.385454
[1,0]<stderr>:INFO:root:Epoch[37] Batch[600] Loss[5.556]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[600]	rmse=0.021579	lr=0.385383
[1,0]<stderr>:INFO:root:Epoch[37] Batch[700] Loss[4.711]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[700]	rmse=0.021568	lr=0.385312
[1,0]<stderr>:INFO:root:Epoch[37] Batch[800] Loss[3.531]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[800]	rmse=0.021571	lr=0.385240
[1,0]<stderr>:INFO:root:Epoch[37] Batch[900] Loss[4.009]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[900]	rmse=0.021541	lr=0.385169
[1,0]<stderr>:INFO:root:Epoch[37] Batch[1000] Loss[4.705]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[1000]	rmse=0.021554	lr=0.385097
[1,0]<stderr>:INFO:root:Epoch[37] Batch[1100] Loss[3.148]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[1100]	rmse=0.021562	lr=0.385025
[1,0]<stderr>:INFO:root:Epoch[37] Batch[1200] Loss[3.411]
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[1200]	rmse=0.021562	lr=0.384953
[1,0]<stderr>:INFO:root:Epoch[37] Rank[0] Batch[1251]	Time cost=398.62	Train-metric=0.021567
[1,0]<stderr>:INFO:root:Epoch[37]	Speed: 3213.61 samples/sec
[1,0]<stderr>:INFO:root:Epoch[38] Batch[100] Loss[3.171]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[100]	rmse=0.021414	lr=0.384844
[1,0]<stderr>:INFO:root:Epoch[38] Batch[200] Loss[2.977]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[200]	rmse=0.021451	lr=0.384772
[1,0]<stderr>:INFO:root:Epoch[38] Batch[300] Loss[3.006]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[300]	rmse=0.021467	lr=0.384699
[1,0]<stderr>:INFO:root:Epoch[38] Batch[400] Loss[5.053]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[400]	rmse=0.021507	lr=0.384627
[1,0]<stderr>:INFO:root:Epoch[38] Batch[500] Loss[3.713]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[500]	rmse=0.021541	lr=0.384554
[1,0]<stderr>:INFO:root:Epoch[38] Batch[600] Loss[3.996]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[600]	rmse=0.021569	lr=0.384481
[1,0]<stderr>:INFO:root:Epoch[38] Batch[700] Loss[4.842]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[700]	rmse=0.021568	lr=0.384407
[1,0]<stderr>:INFO:root:Epoch[38] Batch[800] Loss[2.947]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[800]	rmse=0.021574	lr=0.384334
[1,0]<stderr>:INFO:root:Epoch[38] Batch[900] Loss[3.508]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[900]	rmse=0.021572	lr=0.384260
[1,0]<stderr>:INFO:root:Epoch[38] Batch[1000] Loss[3.033]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[1000]	rmse=0.021563	lr=0.384187
[1,0]<stderr>:INFO:root:Epoch[38] Batch[1100] Loss[2.931]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[1100]	rmse=0.021568	lr=0.384113
[1,0]<stderr>:INFO:root:Epoch[38] Batch[1200] Loss[3.100]
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[1200]	rmse=0.021598	lr=0.384039
[1,0]<stderr>:INFO:root:Epoch[38] Rank[0] Batch[1251]	Time cost=399.39	Train-metric=0.021599
[1,0]<stderr>:INFO:root:Epoch[38]	Speed: 3207.46 samples/sec
[1,0]<stderr>:INFO:root:Epoch[39] Batch[100] Loss[3.038]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[100]	rmse=0.021666	lr=0.383927
[1,0]<stderr>:INFO:root:Epoch[39] Batch[200] Loss[5.365]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[200]	rmse=0.021607	lr=0.383852
[1,0]<stderr>:INFO:root:Epoch[39] Batch[300] Loss[5.126]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[300]	rmse=0.021585	lr=0.383778
[1,0]<stderr>:INFO:root:Epoch[39] Batch[400] Loss[4.682]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[400]	rmse=0.021574	lr=0.383703
[1,0]<stderr>:INFO:root:Epoch[39] Batch[500] Loss[2.973]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[500]	rmse=0.021565	lr=0.383628
[1,0]<stderr>:INFO:root:Epoch[39] Batch[600] Loss[2.814]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[600]	rmse=0.021541	lr=0.383553
[1,0]<stderr>:INFO:root:Epoch[39] Batch[700] Loss[4.271]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[700]	rmse=0.021570	lr=0.383477
[1,0]<stderr>:INFO:root:Epoch[39] Batch[800] Loss[3.317]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[800]	rmse=0.021594	lr=0.383402
[1,0]<stderr>:INFO:root:Epoch[39] Batch[900] Loss[3.161]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[900]	rmse=0.021617	lr=0.383326
[1,0]<stderr>:INFO:root:Epoch[39] Batch[1000] Loss[3.235]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[1000]	rmse=0.021613	lr=0.383251
[1,0]<stderr>:INFO:root:Epoch[39] Batch[1100] Loss[2.713]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[1100]	rmse=0.021617	lr=0.383175
[1,0]<stderr>:INFO:root:Epoch[39] Batch[1200] Loss[4.568]
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[1200]	rmse=0.021607	lr=0.383099
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0] Batch[1251]	Time cost=398.50	Train-metric=0.021606
[1,0]<stderr>:INFO:root:Epoch[39]	Speed: 3214.60 samples/sec
[1,0]<stderr>:INFO:root:Epoch[39] Rank[0]	Validation-accuracy=0.546820	Validation-top_k_accuracy_5=0.792020
[1,0]<stderr>:INFO:root:Epoch[40] Batch[100] Loss[3.030]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[100]	rmse=0.021646	lr=0.382983
[1,0]<stderr>:INFO:root:Epoch[40] Batch[200] Loss[3.205]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[200]	rmse=0.021604	lr=0.382907
[1,0]<stderr>:INFO:root:Epoch[40] Batch[300] Loss[4.347]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[300]	rmse=0.021652	lr=0.382830
[1,0]<stderr>:INFO:root:Epoch[40] Batch[400] Loss[3.302]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[400]	rmse=0.021608	lr=0.382753
[1,0]<stderr>:INFO:root:Epoch[40] Batch[500] Loss[5.204]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[500]	rmse=0.021590	lr=0.382676
[1,0]<stderr>:INFO:root:Epoch[40] Batch[600] Loss[3.464]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[600]	rmse=0.021593	lr=0.382599
[1,0]<stderr>:INFO:root:Epoch[40] Batch[700] Loss[2.883]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[700]	rmse=0.021577	lr=0.382522
[1,0]<stderr>:INFO:root:Epoch[40] Batch[800] Loss[3.143]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[800]	rmse=0.021576	lr=0.382444
[1,0]<stderr>:INFO:root:Epoch[40] Batch[900] Loss[3.218]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[900]	rmse=0.021579	lr=0.382367
[1,0]<stderr>:INFO:root:Epoch[40] Batch[1000] Loss[3.611]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[1000]	rmse=0.021614	lr=0.382289
[1,0]<stderr>:INFO:root:Epoch[40] Batch[1100] Loss[3.869]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[1100]	rmse=0.021619	lr=0.382211
[1,0]<stderr>:INFO:root:Epoch[40] Batch[1200] Loss[2.914]
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[1200]	rmse=0.021609	lr=0.382133
[1,0]<stderr>:INFO:root:Epoch[40] Rank[0] Batch[1251]	Time cost=396.76	Train-metric=0.021618
[1,0]<stderr>:INFO:root:Epoch[40]	Speed: 3228.72 samples/sec
[1,0]<stderr>:INFO:root:Epoch[41] Batch[100] Loss[2.943]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[100]	rmse=0.021391	lr=0.382014
[1,0]<stderr>:INFO:root:Epoch[41] Batch[200] Loss[5.351]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[200]	rmse=0.021457	lr=0.381936
[1,0]<stderr>:INFO:root:Epoch[41] Batch[300] Loss[2.813]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[300]	rmse=0.021496	lr=0.381857
[1,0]<stderr>:INFO:root:Epoch[41] Batch[400] Loss[2.963]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[400]	rmse=0.021478	lr=0.381778
[1,0]<stderr>:INFO:root:Epoch[41] Batch[500] Loss[2.965]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[500]	rmse=0.021481	lr=0.381699
[1,0]<stderr>:INFO:root:Epoch[41] Batch[600] Loss[2.932]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[600]	rmse=0.021453	lr=0.381620
[1,0]<stderr>:INFO:root:Epoch[41] Batch[700] Loss[5.574]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[700]	rmse=0.021466	lr=0.381540
[1,0]<stderr>:INFO:root:Epoch[41] Batch[800] Loss[4.149]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[800]	rmse=0.021488	lr=0.381461
[1,0]<stderr>:INFO:root:Epoch[41] Batch[900] Loss[5.206]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[900]	rmse=0.021520	lr=0.381381
[1,0]<stderr>:INFO:root:Epoch[41] Batch[1000] Loss[5.245]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[1000]	rmse=0.021528	lr=0.381301
[1,0]<stderr>:INFO:root:Epoch[41] Batch[1100] Loss[3.717]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[1100]	rmse=0.021550	lr=0.381221
[1,0]<stderr>:INFO:root:Epoch[41] Batch[1200] Loss[4.106]
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[1200]	rmse=0.021551	lr=0.381141
[1,0]<stderr>:INFO:root:Epoch[41] Rank[0] Batch[1251]	Time cost=398.49	Train-metric=0.021566
[1,0]<stderr>:INFO:root:Epoch[41]	Speed: 3214.71 samples/sec
[1,0]<stderr>:INFO:root:Epoch[42] Batch[100] Loss[3.573]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[100]	rmse=0.021386	lr=0.381020
[1,0]<stderr>:INFO:root:Epoch[42] Batch[200] Loss[3.900]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[200]	rmse=0.021486	lr=0.380939
[1,0]<stderr>:INFO:root:Epoch[42] Batch[300] Loss[5.588]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[300]	rmse=0.021357	lr=0.380858
[1,0]<stderr>:INFO:root:Epoch[42] Batch[400] Loss[5.327]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[400]	rmse=0.021389	lr=0.380777
[1,0]<stderr>:INFO:root:Epoch[42] Batch[500] Loss[3.437]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[500]	rmse=0.021454	lr=0.380696
[1,0]<stderr>:INFO:root:Epoch[42] Batch[600] Loss[2.931]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[600]	rmse=0.021503	lr=0.380615
[1,0]<stderr>:INFO:root:Epoch[42] Batch[700] Loss[4.870]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[700]	rmse=0.021508	lr=0.380533
[1,0]<stderr>:INFO:root:Epoch[42] Batch[800] Loss[3.482]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[800]	rmse=0.021486	lr=0.380452
[1,0]<stderr>:INFO:root:Epoch[42] Batch[900] Loss[5.467]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[900]	rmse=0.021488	lr=0.380370
[1,0]<stderr>:INFO:root:Epoch[42] Batch[1000] Loss[3.004]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[1000]	rmse=0.021493	lr=0.380288
[1,0]<stderr>:INFO:root:Epoch[42] Batch[1100] Loss[4.107]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[1100]	rmse=0.021506	lr=0.380206
[1,0]<stderr>:INFO:root:Epoch[42] Batch[1200] Loss[3.040]
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[1200]	rmse=0.021516	lr=0.380124
[1,0]<stderr>:INFO:root:Epoch[42] Rank[0] Batch[1251]	Time cost=398.89	Train-metric=0.021521
[1,0]<stderr>:INFO:root:Epoch[42]	Speed: 3211.48 samples/sec
[1,0]<stderr>:INFO:root:Epoch[43] Batch[100] Loss[2.961]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[100]	rmse=0.021440	lr=0.380000
[1,0]<stderr>:INFO:root:Epoch[43] Batch[200] Loss[4.058]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[200]	rmse=0.021408	lr=0.379917
[1,0]<stderr>:INFO:root:Epoch[43] Batch[300] Loss[3.181]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[300]	rmse=0.021437	lr=0.379834
[1,0]<stderr>:INFO:root:Epoch[43] Batch[400] Loss[3.092]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[400]	rmse=0.021387	lr=0.379751
[1,0]<stderr>:INFO:root:Epoch[43] Batch[500] Loss[5.258]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[500]	rmse=0.021352	lr=0.379668
[1,0]<stderr>:INFO:root:Epoch[43] Batch[600] Loss[3.839]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[600]	rmse=0.021354	lr=0.379585
[1,0]<stderr>:INFO:root:Epoch[43] Batch[700] Loss[2.943]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[700]	rmse=0.021367	lr=0.379501
[1,0]<stderr>:INFO:root:Epoch[43] Batch[800] Loss[3.506]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[800]	rmse=0.021400	lr=0.379418
[1,0]<stderr>:INFO:root:Epoch[43] Batch[900] Loss[3.147]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[900]	rmse=0.021414	lr=0.379334
[1,0]<stderr>:INFO:root:Epoch[43] Batch[1000] Loss[4.874]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[1000]	rmse=0.021430	lr=0.379250
[1,0]<stderr>:INFO:root:Epoch[43] Batch[1100] Loss[3.064]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[1100]	rmse=0.021459	lr=0.379166
[1,0]<stderr>:INFO:root:Epoch[43] Batch[1200] Loss[3.948]
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[1200]	rmse=0.021452	lr=0.379082
[1,0]<stderr>:INFO:root:Epoch[43] Rank[0] Batch[1251]	Time cost=399.33	Train-metric=0.021457
[1,0]<stderr>:INFO:root:Epoch[43]	Speed: 3207.93 samples/sec
[1,0]<stderr>:INFO:root:Epoch[44] Batch[100] Loss[3.116]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[100]	rmse=0.021577	lr=0.378954
[1,0]<stderr>:INFO:root:Epoch[44] Batch[200] Loss[3.157]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[200]	rmse=0.021563	lr=0.378870
[1,0]<stderr>:INFO:root:Epoch[44] Batch[300] Loss[2.808]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[300]	rmse=0.021562	lr=0.378785
[1,0]<stderr>:INFO:root:Epoch[44] Batch[400] Loss[3.001]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[400]	rmse=0.021532	lr=0.378700
[1,0]<stderr>:INFO:root:Epoch[44] Batch[500] Loss[3.273]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[500]	rmse=0.021536	lr=0.378615
[1,0]<stderr>:INFO:root:Epoch[44] Batch[600] Loss[3.010]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[600]	rmse=0.021531	lr=0.378529
[1,0]<stderr>:INFO:root:Epoch[44] Batch[700] Loss[3.252]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[700]	rmse=0.021532	lr=0.378444
[1,0]<stderr>:INFO:root:Epoch[44] Batch[800] Loss[3.991]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[800]	rmse=0.021553	lr=0.378358
[1,0]<stderr>:INFO:root:Epoch[44] Batch[900] Loss[2.941]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[900]	rmse=0.021576	lr=0.378273
[1,0]<stderr>:INFO:root:Epoch[44] Batch[1000] Loss[3.222]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[1000]	rmse=0.021581	lr=0.378187
[1,0]<stderr>:INFO:root:Epoch[44] Batch[1100] Loss[5.558]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[1100]	rmse=0.021566	lr=0.378101
[1,0]<stderr>:INFO:root:Epoch[44] Batch[1200] Loss[4.737]
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[1200]	rmse=0.021569	lr=0.378014
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0] Batch[1251]	Time cost=402.95	Train-metric=0.021561
[1,0]<stderr>:INFO:root:Epoch[44]	Speed: 3179.12 samples/sec
[1,0]<stderr>:INFO:root:Epoch[44] Rank[0]	Validation-accuracy=0.559860	Validation-top_k_accuracy_5=0.803180
[1,0]<stderr>:INFO:root:Epoch[45] Batch[100] Loss[5.134]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[100]	rmse=0.021405	lr=0.377884
[1,0]<stderr>:INFO:root:Epoch[45] Batch[200] Loss[3.143]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[200]	rmse=0.021526	lr=0.377797
[1,0]<stderr>:INFO:root:Epoch[45] Batch[300] Loss[3.217]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[300]	rmse=0.021417	lr=0.377710
[1,0]<stderr>:INFO:root:Epoch[45] Batch[400] Loss[2.983]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[400]	rmse=0.021445	lr=0.377623
[1,0]<stderr>:INFO:root:Epoch[45] Batch[500] Loss[5.234]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[500]	rmse=0.021445	lr=0.377536
[1,0]<stderr>:INFO:root:Epoch[45] Batch[600] Loss[4.147]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[600]	rmse=0.021464	lr=0.377449
[1,0]<stderr>:INFO:root:Epoch[45] Batch[700] Loss[3.080]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[700]	rmse=0.021471	lr=0.377362
[1,0]<stderr>:INFO:root:Epoch[45] Batch[800] Loss[5.345]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[800]	rmse=0.021463	lr=0.377274
[1,0]<stderr>:INFO:root:Epoch[45] Batch[900] Loss[3.278]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[900]	rmse=0.021493	lr=0.377186
[1,0]<stderr>:INFO:root:Epoch[45] Batch[1000] Loss[4.636]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[1000]	rmse=0.021510	lr=0.377098
[1,0]<stderr>:INFO:root:Epoch[45] Batch[1100] Loss[3.025]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[1100]	rmse=0.021522	lr=0.377010
[1,0]<stderr>:INFO:root:Epoch[45] Batch[1200] Loss[3.111]
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[1200]	rmse=0.021530	lr=0.376922
[1,0]<stderr>:INFO:root:Epoch[45] Rank[0] Batch[1251]	Time cost=401.76	Train-metric=0.021539
[1,0]<stderr>:INFO:root:Epoch[45]	Speed: 3188.54 samples/sec
[1,0]<stderr>:INFO:root:Epoch[46] Batch[100] Loss[3.287]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[100]	rmse=0.021619	lr=0.376788
[1,0]<stderr>:INFO:root:Epoch[46] Batch[200] Loss[2.989]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[200]	rmse=0.021562	lr=0.376700
[1,0]<stderr>:INFO:root:Epoch[46] Batch[300] Loss[2.953]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[300]	rmse=0.021494	lr=0.376611
[1,0]<stderr>:INFO:root:Epoch[46] Batch[400] Loss[3.285]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[400]	rmse=0.021470	lr=0.376522
[1,0]<stderr>:INFO:root:Epoch[46] Batch[500] Loss[2.985]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[500]	rmse=0.021423	lr=0.376433
[1,0]<stderr>:INFO:root:Epoch[46] Batch[600] Loss[2.977]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[600]	rmse=0.021443	lr=0.376344
[1,0]<stderr>:INFO:root:Epoch[46] Batch[700] Loss[3.409]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[700]	rmse=0.021462	lr=0.376254
[1,0]<stderr>:INFO:root:Epoch[46] Batch[800] Loss[3.309]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[800]	rmse=0.021468	lr=0.376165
[1,0]<stderr>:INFO:root:Epoch[46] Batch[900] Loss[5.404]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[900]	rmse=0.021460	lr=0.376075
[1,0]<stderr>:INFO:root:Epoch[46] Batch[1000] Loss[3.478]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[1000]	rmse=0.021467	lr=0.375985
[1,0]<stderr>:INFO:root:Epoch[46] Batch[1100] Loss[2.819]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[1100]	rmse=0.021467	lr=0.375895
[1,0]<stderr>:INFO:root:Epoch[46] Batch[1200] Loss[2.872]
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[1200]	rmse=0.021461	lr=0.375805
[1,0]<stderr>:INFO:root:Epoch[46] Rank[0] Batch[1251]	Time cost=399.89	Train-metric=0.021458
[1,0]<stderr>:INFO:root:Epoch[46]	Speed: 3203.41 samples/sec
[1,0]<stderr>:INFO:root:Epoch[47] Batch[100] Loss[2.889]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[100]	rmse=0.021394	lr=0.375668
[1,0]<stderr>:INFO:root:Epoch[47] Batch[200] Loss[3.033]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[200]	rmse=0.021405	lr=0.375578
[1,0]<stderr>:INFO:root:Epoch[47] Batch[300] Loss[3.407]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[300]	rmse=0.021360	lr=0.375487
[1,0]<stderr>:INFO:root:Epoch[47] Batch[400] Loss[3.249]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[400]	rmse=0.021469	lr=0.375396
[1,0]<stderr>:INFO:root:Epoch[47] Batch[500] Loss[3.583]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[500]	rmse=0.021469	lr=0.375305
[1,0]<stderr>:INFO:root:Epoch[47] Batch[600] Loss[3.115]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[600]	rmse=0.021438	lr=0.375214
[1,0]<stderr>:INFO:root:Epoch[47] Batch[700] Loss[3.313]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[700]	rmse=0.021469	lr=0.375122
[1,0]<stderr>:INFO:root:Epoch[47] Batch[800] Loss[5.394]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[800]	rmse=0.021456	lr=0.375031
[1,0]<stderr>:INFO:root:Epoch[47] Batch[900] Loss[3.361]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[900]	rmse=0.021452	lr=0.374939
[1,0]<stderr>:INFO:root:Epoch[47] Batch[1000] Loss[2.930]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[1000]	rmse=0.021473	lr=0.374847
[1,0]<stderr>:INFO:root:Epoch[47] Batch[1100] Loss[5.359]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[1100]	rmse=0.021450	lr=0.374755
[1,0]<stderr>:INFO:root:Epoch[47] Batch[1200] Loss[3.957]
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[1200]	rmse=0.021451	lr=0.374663
[1,0]<stderr>:INFO:root:Epoch[47] Rank[0] Batch[1251]	Time cost=400.99	Train-metric=0.021450
[1,0]<stderr>:INFO:root:Epoch[47]	Speed: 3194.64 samples/sec
[1,0]<stderr>:INFO:root:Epoch[48] Batch[100] Loss[5.597]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[100]	rmse=0.021429	lr=0.374523
[1,0]<stderr>:INFO:root:Epoch[48] Batch[200] Loss[3.221]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[200]	rmse=0.021416	lr=0.374431
[1,0]<stderr>:INFO:root:Epoch[48] Batch[300] Loss[3.159]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[300]	rmse=0.021405	lr=0.374338
[1,0]<stderr>:INFO:root:Epoch[48] Batch[400] Loss[5.407]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[400]	rmse=0.021393	lr=0.374245
[1,0]<stderr>:INFO:root:Epoch[48] Batch[500] Loss[3.365]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[500]	rmse=0.021391	lr=0.374152
[1,0]<stderr>:INFO:root:Epoch[48] Batch[600] Loss[3.179]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[600]	rmse=0.021397	lr=0.374059
[1,0]<stderr>:INFO:root:Epoch[48] Batch[700] Loss[3.207]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[700]	rmse=0.021427	lr=0.373966
[1,0]<stderr>:INFO:root:Epoch[48] Batch[800] Loss[3.164]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[800]	rmse=0.021408	lr=0.373872
[1,0]<stderr>:INFO:root:Epoch[48] Batch[900] Loss[3.408]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[900]	rmse=0.021408	lr=0.373779
[1,0]<stderr>:INFO:root:Epoch[48] Batch[1000] Loss[3.332]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[1000]	rmse=0.021434	lr=0.373685
[1,0]<stderr>:INFO:root:Epoch[48] Batch[1100] Loss[3.193]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[1100]	rmse=0.021445	lr=0.373591
[1,0]<stderr>:INFO:root:Epoch[48] Batch[1200] Loss[2.761]
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[1200]	rmse=0.021445	lr=0.373497
[1,0]<stderr>:INFO:root:Epoch[48] Rank[0] Batch[1251]	Time cost=399.09	Train-metric=0.021440
[1,0]<stderr>:INFO:root:Epoch[48]	Speed: 3209.86 samples/sec
[1,0]<stderr>:INFO:root:Epoch[49] Batch[100] Loss[2.995]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[100]	rmse=0.021526	lr=0.373354
[1,0]<stderr>:INFO:root:Epoch[49] Batch[200] Loss[3.143]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[200]	rmse=0.021344	lr=0.373260
[1,0]<stderr>:INFO:root:Epoch[49] Batch[300] Loss[3.390]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[300]	rmse=0.021314	lr=0.373165
[1,0]<stderr>:INFO:root:Epoch[49] Batch[400] Loss[3.693]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[400]	rmse=0.021277	lr=0.373070
[1,0]<stderr>:INFO:root:Epoch[49] Batch[500] Loss[2.969]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[500]	rmse=0.021337	lr=0.372975
[1,0]<stderr>:INFO:root:Epoch[49] Batch[600] Loss[3.016]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[600]	rmse=0.021336	lr=0.372880
[1,0]<stderr>:INFO:root:Epoch[49] Batch[700] Loss[4.827]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[700]	rmse=0.021363	lr=0.372785
[1,0]<stderr>:INFO:root:Epoch[49] Batch[800] Loss[3.006]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[800]	rmse=0.021361	lr=0.372689
[1,0]<stderr>:INFO:root:Epoch[49] Batch[900] Loss[3.194]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[900]	rmse=0.021399	lr=0.372594
[1,0]<stderr>:INFO:root:Epoch[49] Batch[1000] Loss[4.188]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[1000]	rmse=0.021424	lr=0.372498
[1,0]<stderr>:INFO:root:Epoch[49] Batch[1100] Loss[3.034]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[1100]	rmse=0.021424	lr=0.372402
[1,0]<stderr>:INFO:root:Epoch[49] Batch[1200] Loss[2.904]
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[1200]	rmse=0.021448	lr=0.372306
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0] Batch[1251]	Time cost=398.81	Train-metric=0.021458
[1,0]<stderr>:INFO:root:Epoch[49]	Speed: 3212.13 samples/sec
[1,0]<stderr>:INFO:root:Epoch[49] Rank[0]	Validation-accuracy=0.559560	Validation-top_k_accuracy_5=0.803140
[1,0]<stderr>:INFO:root:Epoch[50] Batch[100] Loss[4.187]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[100]	rmse=0.021244	lr=0.372161
[1,0]<stderr>:INFO:root:Epoch[50] Batch[200] Loss[2.888]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[200]	rmse=0.021419	lr=0.372064
[1,0]<stderr>:INFO:root:Epoch[50] Batch[300] Loss[3.508]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[300]	rmse=0.021350	lr=0.371967
[1,0]<stderr>:INFO:root:Epoch[50] Batch[400] Loss[2.932]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[400]	rmse=0.021356	lr=0.371871
[1,0]<stderr>:INFO:root:Epoch[50] Batch[500] Loss[2.681]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[500]	rmse=0.021370	lr=0.371774
[1,0]<stderr>:INFO:root:Epoch[50] Batch[600] Loss[4.528]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[600]	rmse=0.021372	lr=0.371677
[1,0]<stderr>:INFO:root:Epoch[50] Batch[700] Loss[2.739]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[700]	rmse=0.021392	lr=0.371579
[1,0]<stderr>:INFO:root:Epoch[50] Batch[800] Loss[3.713]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[800]	rmse=0.021392	lr=0.371482
[1,0]<stderr>:INFO:root:Epoch[50] Batch[900] Loss[4.326]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[900]	rmse=0.021384	lr=0.371385
[1,0]<stderr>:INFO:root:Epoch[50] Batch[1000] Loss[3.005]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[1000]	rmse=0.021413	lr=0.371287
[1,0]<stderr>:INFO:root:Epoch[50] Batch[1100] Loss[3.070]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[1100]	rmse=0.021432	lr=0.371189
[1,0]<stderr>:INFO:root:Epoch[50] Batch[1200] Loss[3.900]
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[1200]	rmse=0.021465	lr=0.371091
[1,0]<stderr>:INFO:root:Epoch[50] Rank[0] Batch[1251]	Time cost=398.02	Train-metric=0.021468
[1,0]<stderr>:INFO:root:Epoch[50]	Speed: 3218.52 samples/sec
[1,0]<stderr>:INFO:root:Epoch[51] Batch[100] Loss[2.760]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[100]	rmse=0.021453	lr=0.370943
[1,0]<stderr>:INFO:root:Epoch[51] Batch[200] Loss[5.153]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[200]	rmse=0.021375	lr=0.370844
[1,0]<stderr>:INFO:root:Epoch[51] Batch[300] Loss[3.499]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[300]	rmse=0.021428	lr=0.370746
[1,0]<stderr>:INFO:root:Epoch[51] Batch[400] Loss[2.961]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[400]	rmse=0.021423	lr=0.370647
[1,0]<stderr>:INFO:root:Epoch[51] Batch[500] Loss[4.979]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[500]	rmse=0.021422	lr=0.370548
[1,0]<stderr>:INFO:root:Epoch[51] Batch[600] Loss[3.580]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[600]	rmse=0.021444	lr=0.370449
[1,0]<stderr>:INFO:root:Epoch[51] Batch[700] Loss[2.961]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[700]	rmse=0.021472	lr=0.370350
[1,0]<stderr>:INFO:root:Epoch[51] Batch[800] Loss[2.974]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[800]	rmse=0.021463	lr=0.370251
[1,0]<stderr>:INFO:root:Epoch[51] Batch[900] Loss[2.937]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[900]	rmse=0.021477	lr=0.370151
[1,0]<stderr>:INFO:root:Epoch[51] Batch[1000] Loss[3.612]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[1000]	rmse=0.021494	lr=0.370052
[1,0]<stderr>:INFO:root:Epoch[51] Batch[1100] Loss[2.980]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[1100]	rmse=0.021489	lr=0.369952
[1,0]<stderr>:INFO:root:Epoch[51] Batch[1200] Loss[4.338]
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[1200]	rmse=0.021486	lr=0.369852
[1,0]<stderr>:INFO:root:Epoch[51] Rank[0] Batch[1251]	Time cost=399.30	Train-metric=0.021483
[1,0]<stderr>:INFO:root:Epoch[51]	Speed: 3208.14 samples/sec
[1,0]<stderr>:INFO:root:Epoch[52] Batch[100] Loss[5.323]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[100]	rmse=0.021240	lr=0.369701
[1,0]<stderr>:INFO:root:Epoch[52] Batch[200] Loss[3.812]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[200]	rmse=0.021346	lr=0.369601
[1,0]<stderr>:INFO:root:Epoch[52] Batch[300] Loss[2.935]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[300]	rmse=0.021372	lr=0.369500
[1,0]<stderr>:INFO:root:Epoch[52] Batch[400] Loss[3.090]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[400]	rmse=0.021400	lr=0.369400
[1,0]<stderr>:INFO:root:Epoch[52] Batch[500] Loss[4.143]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[500]	rmse=0.021414	lr=0.369299
[1,0]<stderr>:INFO:root:Epoch[52] Batch[600] Loss[2.985]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[600]	rmse=0.021437	lr=0.369198
[1,0]<stderr>:INFO:root:Epoch[52] Batch[700] Loss[3.073]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[700]	rmse=0.021437	lr=0.369097
[1,0]<stderr>:INFO:root:Epoch[52] Batch[800] Loss[2.866]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[800]	rmse=0.021410	lr=0.368996
[1,0]<stderr>:INFO:root:Epoch[52] Batch[900] Loss[3.232]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[900]	rmse=0.021409	lr=0.368894
[1,0]<stderr>:INFO:root:Epoch[52] Batch[1000] Loss[2.910]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[1000]	rmse=0.021395	lr=0.368793
[1,0]<stderr>:INFO:root:Epoch[52] Batch[1100] Loss[3.109]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[1100]	rmse=0.021408	lr=0.368691
[1,0]<stderr>:INFO:root:Epoch[52] Batch[1200] Loss[3.235]
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[1200]	rmse=0.021415	lr=0.368589
[1,0]<stderr>:INFO:root:Epoch[52] Rank[0] Batch[1251]	Time cost=400.41	Train-metric=0.021414
[1,0]<stderr>:INFO:root:Epoch[52]	Speed: 3199.31 samples/sec
[1,0]<stderr>:INFO:root:Epoch[53] Batch[100] Loss[3.150]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[100]	rmse=0.021228	lr=0.368435
[1,0]<stderr>:INFO:root:Epoch[53] Batch[200] Loss[3.012]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[200]	rmse=0.021246	lr=0.368333
[1,0]<stderr>:INFO:root:Epoch[53] Batch[300] Loss[2.943]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[300]	rmse=0.021248	lr=0.368231
[1,0]<stderr>:INFO:root:Epoch[53] Batch[400] Loss[2.799]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[400]	rmse=0.021259	lr=0.368128
[1,0]<stderr>:INFO:root:Epoch[53] Batch[500] Loss[2.981]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[500]	rmse=0.021322	lr=0.368026
[1,0]<stderr>:INFO:root:Epoch[53] Batch[600] Loss[4.510]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[600]	rmse=0.021356	lr=0.367923
[1,0]<stderr>:INFO:root:Epoch[53] Batch[700] Loss[2.966]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[700]	rmse=0.021371	lr=0.367820
[1,0]<stderr>:INFO:root:Epoch[53] Batch[800] Loss[2.694]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[800]	rmse=0.021404	lr=0.367717
[1,0]<stderr>:INFO:root:Epoch[53] Batch[900] Loss[3.332]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[900]	rmse=0.021415	lr=0.367614
[1,0]<stderr>:INFO:root:Epoch[53] Batch[1000] Loss[5.288]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[1000]	rmse=0.021427	lr=0.367510
[1,0]<stderr>:INFO:root:Epoch[53] Batch[1100] Loss[4.044]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[1100]	rmse=0.021426	lr=0.367407
[1,0]<stderr>:INFO:root:Epoch[53] Batch[1200] Loss[3.047]
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[1200]	rmse=0.021431	lr=0.367303
[1,0]<stderr>:INFO:root:Epoch[53] Rank[0] Batch[1251]	Time cost=399.45	Train-metric=0.021420
[1,0]<stderr>:INFO:root:Epoch[53]	Speed: 3206.94 samples/sec
[1,0]<stderr>:INFO:root:Epoch[54] Batch[100] Loss[3.494]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[100]	rmse=0.021105	lr=0.367146
[1,0]<stderr>:INFO:root:Epoch[54] Batch[200] Loss[3.101]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[200]	rmse=0.021275	lr=0.367042
[1,0]<stderr>:INFO:root:Epoch[54] Batch[300] Loss[3.000]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[300]	rmse=0.021321	lr=0.366938
[1,0]<stderr>:INFO:root:Epoch[54] Batch[400] Loss[2.786]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[400]	rmse=0.021318	lr=0.366834
[1,0]<stderr>:INFO:root:Epoch[54] Batch[500] Loss[2.781]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[500]	rmse=0.021368	lr=0.366729
[1,0]<stderr>:INFO:root:Epoch[54] Batch[600] Loss[2.730]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[600]	rmse=0.021381	lr=0.366624
[1,0]<stderr>:INFO:root:Epoch[54] Batch[700] Loss[3.842]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[700]	rmse=0.021409	lr=0.366520
[1,0]<stderr>:INFO:root:Epoch[54] Batch[800] Loss[2.861]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[800]	rmse=0.021406	lr=0.366415
[1,0]<stderr>:INFO:root:Epoch[54] Batch[900] Loss[3.286]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[900]	rmse=0.021408	lr=0.366309
[1,0]<stderr>:INFO:root:Epoch[54] Batch[1000] Loss[3.896]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[1000]	rmse=0.021407	lr=0.366204
[1,0]<stderr>:INFO:root:Epoch[54] Batch[1100] Loss[4.758]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[1100]	rmse=0.021405	lr=0.366099
[1,0]<stderr>:INFO:root:Epoch[54] Batch[1200] Loss[4.291]
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[1200]	rmse=0.021411	lr=0.365993
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0] Batch[1251]	Time cost=399.09	Train-metric=0.021410
[1,0]<stderr>:INFO:root:Epoch[54]	Speed: 3209.84 samples/sec
[1,0]<stderr>:INFO:root:Epoch[54] Rank[0]	Validation-accuracy=0.574500	Validation-top_k_accuracy_5=0.817740
[1,0]<stderr>:INFO:root:Epoch[55] Batch[100] Loss[3.326]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[100]	rmse=0.021402	lr=0.365834
[1,0]<stderr>:INFO:root:Epoch[55] Batch[200] Loss[2.659]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[200]	rmse=0.021429	lr=0.365728
[1,0]<stderr>:INFO:root:Epoch[55] Batch[300] Loss[3.375]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[300]	rmse=0.021426	lr=0.365621
[1,0]<stderr>:INFO:root:Epoch[55] Batch[400] Loss[3.161]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[400]	rmse=0.021417	lr=0.365515
[1,0]<stderr>:INFO:root:Epoch[55] Batch[500] Loss[3.356]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[500]	rmse=0.021462	lr=0.365409
[1,0]<stderr>:INFO:root:Epoch[55] Batch[600] Loss[2.820]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[600]	rmse=0.021462	lr=0.365302
[1,0]<stderr>:INFO:root:Epoch[55] Batch[700] Loss[3.510]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[700]	rmse=0.021444	lr=0.365196
[1,0]<stderr>:INFO:root:Epoch[55] Batch[800] Loss[3.459]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[800]	rmse=0.021457	lr=0.365089
[1,0]<stderr>:INFO:root:Epoch[55] Batch[900] Loss[3.194]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[900]	rmse=0.021428	lr=0.364982
[1,0]<stderr>:INFO:root:Epoch[55] Batch[1000] Loss[5.408]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[1000]	rmse=0.021445	lr=0.364875
[1,0]<stderr>:INFO:root:Epoch[55] Batch[1100] Loss[4.200]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[1100]	rmse=0.021459	lr=0.364767
[1,0]<stderr>:INFO:root:Epoch[55] Batch[1200] Loss[3.078]
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[1200]	rmse=0.021449	lr=0.364660
[1,0]<stderr>:INFO:root:Epoch[55] Rank[0] Batch[1251]	Time cost=398.13	Train-metric=0.021463
[1,0]<stderr>:INFO:root:Epoch[55]	Speed: 3217.59 samples/sec
[1,0]<stderr>:INFO:root:Epoch[56] Batch[100] Loss[3.431]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[100]	rmse=0.021501	lr=0.364498
[1,0]<stderr>:INFO:root:Epoch[56] Batch[200] Loss[4.922]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[200]	rmse=0.021337	lr=0.364390
[1,0]<stderr>:INFO:root:Epoch[56] Batch[300] Loss[4.310]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[300]	rmse=0.021301	lr=0.364282
[1,0]<stderr>:INFO:root:Epoch[56] Batch[400] Loss[3.055]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[400]	rmse=0.021360	lr=0.364174
[1,0]<stderr>:INFO:root:Epoch[56] Batch[500] Loss[3.255]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[500]	rmse=0.021342	lr=0.364066
[1,0]<stderr>:INFO:root:Epoch[56] Batch[600] Loss[3.059]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[600]	rmse=0.021350	lr=0.363957
[1,0]<stderr>:INFO:root:Epoch[56] Batch[700] Loss[2.831]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[700]	rmse=0.021382	lr=0.363849
[1,0]<stderr>:INFO:root:Epoch[56] Batch[800] Loss[3.433]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[800]	rmse=0.021390	lr=0.363740
[1,0]<stderr>:INFO:root:Epoch[56] Batch[900] Loss[3.015]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[900]	rmse=0.021403	lr=0.363631
[1,0]<stderr>:INFO:root:Epoch[56] Batch[1000] Loss[2.951]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[1000]	rmse=0.021408	lr=0.363522
[1,0]<stderr>:INFO:root:Epoch[56] Batch[1100] Loss[3.184]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[1100]	rmse=0.021423	lr=0.363413
[1,0]<stderr>:INFO:root:Epoch[56] Batch[1200] Loss[5.117]
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[1200]	rmse=0.021422	lr=0.363304
[1,0]<stderr>:INFO:root:Epoch[56] Rank[0] Batch[1251]	Time cost=399.77	Train-metric=0.021427
[1,0]<stderr>:INFO:root:Epoch[56]	Speed: 3204.37 samples/sec
[1,0]<stderr>:INFO:root:Epoch[57] Batch[100] Loss[3.545]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[100]	rmse=0.021303	lr=0.363139
[1,0]<stderr>:INFO:root:Epoch[57] Batch[200] Loss[4.619]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[200]	rmse=0.021364	lr=0.363029
[1,0]<stderr>:INFO:root:Epoch[57] Batch[300] Loss[3.832]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[300]	rmse=0.021396	lr=0.362919
[1,0]<stderr>:INFO:root:Epoch[57] Batch[400] Loss[3.274]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[400]	rmse=0.021432	lr=0.362809
[1,0]<stderr>:INFO:root:Epoch[57] Batch[500] Loss[3.223]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[500]	rmse=0.021478	lr=0.362699
[1,0]<stderr>:INFO:root:Epoch[57] Batch[600] Loss[5.003]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[600]	rmse=0.021448	lr=0.362589
[1,0]<stderr>:INFO:root:Epoch[57] Batch[700] Loss[3.111]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[700]	rmse=0.021464	lr=0.362479
[1,0]<stderr>:INFO:root:Epoch[57] Batch[800] Loss[2.980]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[800]	rmse=0.021464	lr=0.362368
[1,0]<stderr>:INFO:root:Epoch[57] Batch[900] Loss[3.786]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[900]	rmse=0.021463	lr=0.362257
[1,0]<stderr>:INFO:root:Epoch[57] Batch[1000] Loss[3.056]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[1000]	rmse=0.021461	lr=0.362147
[1,0]<stderr>:INFO:root:Epoch[57] Batch[1100] Loss[3.055]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[1100]	rmse=0.021457	lr=0.362036
[1,0]<stderr>:INFO:root:Epoch[57] Batch[1200] Loss[5.030]
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[1200]	rmse=0.021426	lr=0.361925
[1,0]<stderr>:INFO:root:Epoch[57] Rank[0] Batch[1251]	Time cost=398.82	Train-metric=0.021414
[1,0]<stderr>:INFO:root:Epoch[57]	Speed: 3212.04 samples/sec
[1,0]<stderr>:INFO:root:Epoch[58] Batch[100] Loss[2.821]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[100]	rmse=0.021200	lr=0.361757
[1,0]<stderr>:INFO:root:Epoch[58] Batch[200] Loss[2.828]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[200]	rmse=0.021227	lr=0.361645
[1,0]<stderr>:INFO:root:Epoch[58] Batch[300] Loss[3.512]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[300]	rmse=0.021245	lr=0.361534
[1,0]<stderr>:INFO:root:Epoch[58] Batch[400] Loss[3.224]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[400]	rmse=0.021320	lr=0.361422
[1,0]<stderr>:INFO:root:Epoch[58] Batch[500] Loss[3.170]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[500]	rmse=0.021330	lr=0.361310
[1,0]<stderr>:INFO:root:Epoch[58] Batch[600] Loss[3.089]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[600]	rmse=0.021303	lr=0.361198
[1,0]<stderr>:INFO:root:Epoch[58] Batch[700] Loss[3.276]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[700]	rmse=0.021340	lr=0.361086
[1,0]<stderr>:INFO:root:Epoch[58] Batch[800] Loss[4.920]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[800]	rmse=0.021333	lr=0.360973
[1,0]<stderr>:INFO:root:Epoch[58] Batch[900] Loss[3.149]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[900]	rmse=0.021364	lr=0.360861
[1,0]<stderr>:INFO:root:Epoch[58] Batch[1000] Loss[2.859]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[1000]	rmse=0.021354	lr=0.360748
[1,0]<stderr>:INFO:root:Epoch[58] Batch[1100] Loss[2.819]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[1100]	rmse=0.021372	lr=0.360636
[1,0]<stderr>:INFO:root:Epoch[58] Batch[1200] Loss[3.076]
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[1200]	rmse=0.021390	lr=0.360523
[1,0]<stderr>:INFO:root:Epoch[58] Rank[0] Batch[1251]	Time cost=400.37	Train-metric=0.021412
[1,0]<stderr>:INFO:root:Epoch[58]	Speed: 3199.58 samples/sec
[1,0]<stderr>:INFO:root:Epoch[59] Batch[100] Loss[2.911]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[100]	rmse=0.021265	lr=0.360352
[1,0]<stderr>:INFO:root:Epoch[59] Batch[200] Loss[4.113]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[200]	rmse=0.021257	lr=0.360239
[1,0]<stderr>:INFO:root:Epoch[59] Batch[300] Loss[3.808]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[300]	rmse=0.021247	lr=0.360125
[1,0]<stderr>:INFO:root:Epoch[59] Batch[400] Loss[3.013]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[400]	rmse=0.021198	lr=0.360012
[1,0]<stderr>:INFO:root:Epoch[59] Batch[500] Loss[5.018]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[500]	rmse=0.021209	lr=0.359898
[1,0]<stderr>:INFO:root:Epoch[59] Batch[600] Loss[3.308]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[600]	rmse=0.021236	lr=0.359784
[1,0]<stderr>:INFO:root:Epoch[59] Batch[700] Loss[2.795]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[700]	rmse=0.021277	lr=0.359670
[1,0]<stderr>:INFO:root:Epoch[59] Batch[800] Loss[4.162]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[800]	rmse=0.021282	lr=0.359556
[1,0]<stderr>:INFO:root:Epoch[59] Batch[900] Loss[3.241]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[900]	rmse=0.021304	lr=0.359442
[1,0]<stderr>:INFO:root:Epoch[59] Batch[1000] Loss[4.570]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[1000]	rmse=0.021317	lr=0.359328
[1,0]<stderr>:INFO:root:Epoch[59] Batch[1100] Loss[3.232]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[1100]	rmse=0.021333	lr=0.359213
[1,0]<stderr>:INFO:root:Epoch[59] Batch[1200] Loss[4.192]
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[1200]	rmse=0.021332	lr=0.359098
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0] Batch[1251]	Time cost=399.21	Train-metric=0.021328
[1,0]<stderr>:INFO:root:Epoch[59]	Speed: 3208.92 samples/sec
[1,0]<stderr>:INFO:root:Epoch[59] Rank[0]	Validation-accuracy=0.556400	Validation-top_k_accuracy_5=0.801860
[1,0]<stderr>:INFO:root:Epoch[60] Batch[100] Loss[3.155]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[100]	rmse=0.021360	lr=0.358925
[1,0]<stderr>:INFO:root:Epoch[60] Batch[200] Loss[4.769]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[200]	rmse=0.021251	lr=0.358810
[1,0]<stderr>:INFO:root:Epoch[60] Batch[300] Loss[3.065]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[300]	rmse=0.021256	lr=0.358695
[1,0]<stderr>:INFO:root:Epoch[60] Batch[400] Loss[5.331]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[400]	rmse=0.021224	lr=0.358579
[1,0]<stderr>:INFO:root:Epoch[60] Batch[500] Loss[5.443]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[500]	rmse=0.021249	lr=0.358464
[1,0]<stderr>:INFO:root:Epoch[60] Batch[600] Loss[3.131]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[600]	rmse=0.021286	lr=0.358348
[1,0]<stderr>:INFO:root:Epoch[60] Batch[700] Loss[3.459]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[700]	rmse=0.021319	lr=0.358233
[1,0]<stderr>:INFO:root:Epoch[60] Batch[800] Loss[3.142]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[800]	rmse=0.021335	lr=0.358117
[1,0]<stderr>:INFO:root:Epoch[60] Batch[900] Loss[3.263]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[900]	rmse=0.021330	lr=0.358001
[1,0]<stderr>:INFO:root:Epoch[60] Batch[1000] Loss[3.634]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[1000]	rmse=0.021326	lr=0.357884
[1,0]<stderr>:INFO:root:Epoch[60] Batch[1100] Loss[3.191]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[1100]	rmse=0.021351	lr=0.357768
[1,0]<stderr>:INFO:root:Epoch[60] Batch[1200] Loss[3.449]
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[1200]	rmse=0.021359	lr=0.357652
[1,0]<stderr>:INFO:root:Epoch[60] Rank[0] Batch[1251]	Time cost=399.37	Train-metric=0.021361
[1,0]<stderr>:INFO:root:Epoch[60]	Speed: 3207.63 samples/sec
[1,0]<stderr>:INFO:root:Epoch[61] Batch[100] Loss[3.552]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[100]	rmse=0.021141	lr=0.357475
[1,0]<stderr>:INFO:root:Epoch[61] Batch[200] Loss[3.986]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[200]	rmse=0.021239	lr=0.357359
[1,0]<stderr>:INFO:root:Epoch[61] Batch[300] Loss[3.646]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[300]	rmse=0.021225	lr=0.357242
[1,0]<stderr>:INFO:root:Epoch[61] Batch[400] Loss[3.115]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[400]	rmse=0.021240	lr=0.357125
[1,0]<stderr>:INFO:root:Epoch[61] Batch[500] Loss[4.354]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[500]	rmse=0.021254	lr=0.357007
[1,0]<stderr>:INFO:root:Epoch[61] Batch[600] Loss[3.433]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[600]	rmse=0.021308	lr=0.356890
[1,0]<stderr>:INFO:root:Epoch[61] Batch[700] Loss[2.916]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[700]	rmse=0.021348	lr=0.356772
[1,0]<stderr>:INFO:root:Epoch[61] Batch[800] Loss[4.002]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[800]	rmse=0.021361	lr=0.356655
[1,0]<stderr>:INFO:root:Epoch[61] Batch[900] Loss[2.850]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[900]	rmse=0.021375	lr=0.356537
[1,0]<stderr>:INFO:root:Epoch[61] Batch[1000] Loss[3.160]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[1000]	rmse=0.021372	lr=0.356419
[1,0]<stderr>:INFO:root:Epoch[61] Batch[1100] Loss[3.405]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[1100]	rmse=0.021357	lr=0.356301
[1,0]<stderr>:INFO:root:Epoch[61] Batch[1200] Loss[3.861]
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[1200]	rmse=0.021380	lr=0.356183
[1,0]<stderr>:INFO:root:Epoch[61] Rank[0] Batch[1251]	Time cost=399.14	Train-metric=0.021382
[1,0]<stderr>:INFO:root:Epoch[61]	Speed: 3209.44 samples/sec
[1,0]<stderr>:INFO:root:Epoch[62] Batch[100] Loss[2.955]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[100]	rmse=0.021346	lr=0.356004
[1,0]<stderr>:INFO:root:Epoch[62] Batch[200] Loss[3.860]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[200]	rmse=0.021232	lr=0.355885
[1,0]<stderr>:INFO:root:Epoch[62] Batch[300] Loss[4.532]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[300]	rmse=0.021270	lr=0.355767
[1,0]<stderr>:INFO:root:Epoch[62] Batch[400] Loss[3.216]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[400]	rmse=0.021282	lr=0.355648
[1,0]<stderr>:INFO:root:Epoch[62] Batch[500] Loss[5.358]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[500]	rmse=0.021279	lr=0.355529
[1,0]<stderr>:INFO:root:Epoch[62] Batch[600] Loss[3.165]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[600]	rmse=0.021279	lr=0.355410
[1,0]<stderr>:INFO:root:Epoch[62] Batch[700] Loss[3.186]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[700]	rmse=0.021305	lr=0.355290
[1,0]<stderr>:INFO:root:Epoch[62] Batch[800] Loss[3.294]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[800]	rmse=0.021346	lr=0.355171
[1,0]<stderr>:INFO:root:Epoch[62] Batch[900] Loss[4.868]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[900]	rmse=0.021350	lr=0.355051
[1,0]<stderr>:INFO:root:Epoch[62] Batch[1000] Loss[3.553]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[1000]	rmse=0.021311	lr=0.354932
[1,0]<stderr>:INFO:root:Epoch[62] Batch[1100] Loss[3.015]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[1100]	rmse=0.021329	lr=0.354812
[1,0]<stderr>:INFO:root:Epoch[62] Batch[1200] Loss[2.970]
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[1200]	rmse=0.021333	lr=0.354692
[1,0]<stderr>:INFO:root:Epoch[62] Rank[0] Batch[1251]	Time cost=399.11	Train-metric=0.021349
[1,0]<stderr>:INFO:root:Epoch[62]	Speed: 3209.71 samples/sec
[1,0]<stderr>:INFO:root:Epoch[63] Batch[100] Loss[2.987]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[100]	rmse=0.021450	lr=0.354511
[1,0]<stderr>:INFO:root:Epoch[63] Batch[200] Loss[3.088]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[200]	rmse=0.021398	lr=0.354390
[1,0]<stderr>:INFO:root:Epoch[63] Batch[300] Loss[3.077]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[300]	rmse=0.021382	lr=0.354270
[1,0]<stderr>:INFO:root:Epoch[63] Batch[400] Loss[3.089]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[400]	rmse=0.021372	lr=0.354149
[1,0]<stderr>:INFO:root:Epoch[63] Batch[500] Loss[3.293]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[500]	rmse=0.021340	lr=0.354029
[1,0]<stderr>:INFO:root:Epoch[63] Batch[600] Loss[5.331]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[600]	rmse=0.021351	lr=0.353908
[1,0]<stderr>:INFO:root:Epoch[63] Batch[700] Loss[4.569]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[700]	rmse=0.021356	lr=0.353787
[1,0]<stderr>:INFO:root:Epoch[63] Batch[800] Loss[2.890]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[800]	rmse=0.021367	lr=0.353665
[1,0]<stderr>:INFO:root:Epoch[63] Batch[900] Loss[3.320]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[900]	rmse=0.021340	lr=0.353544
[1,0]<stderr>:INFO:root:Epoch[63] Batch[1000] Loss[5.166]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[1000]	rmse=0.021334	lr=0.353423
[1,0]<stderr>:INFO:root:Epoch[63] Batch[1100] Loss[2.758]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[1100]	rmse=0.021327	lr=0.353301
[1,0]<stderr>:INFO:root:Epoch[63] Batch[1200] Loss[4.872]
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[1200]	rmse=0.021337	lr=0.353180
[1,0]<stderr>:INFO:root:Epoch[63] Rank[0] Batch[1251]	Time cost=398.61	Train-metric=0.021334
[1,0]<stderr>:INFO:root:Epoch[63]	Speed: 3213.71 samples/sec
[1,0]<stderr>:INFO:root:Epoch[64] Batch[100] Loss[2.982]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[100]	rmse=0.021226	lr=0.352996
[1,0]<stderr>:INFO:root:Epoch[64] Batch[200] Loss[5.034]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[200]	rmse=0.021215	lr=0.352873
[1,0]<stderr>:INFO:root:Epoch[64] Batch[300] Loss[3.123]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[300]	rmse=0.021238	lr=0.352751
[1,0]<stderr>:INFO:root:Epoch[64] Batch[400] Loss[5.342]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[400]	rmse=0.021219	lr=0.352629
[1,0]<stderr>:INFO:root:Epoch[64] Batch[500] Loss[4.914]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[500]	rmse=0.021255	lr=0.352507
[1,0]<stderr>:INFO:root:Epoch[64] Batch[600] Loss[2.806]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[600]	rmse=0.021281	lr=0.352384
[1,0]<stderr>:INFO:root:Epoch[64] Batch[700] Loss[2.832]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[700]	rmse=0.021264	lr=0.352261
[1,0]<stderr>:INFO:root:Epoch[64] Batch[800] Loss[4.591]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[800]	rmse=0.021287	lr=0.352138
[1,0]<stderr>:INFO:root:Epoch[64] Batch[900] Loss[3.012]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[900]	rmse=0.021315	lr=0.352015
[1,0]<stderr>:INFO:root:Epoch[64] Batch[1000] Loss[3.834]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[1000]	rmse=0.021329	lr=0.351892
[1,0]<stderr>:INFO:root:Epoch[64] Batch[1100] Loss[4.093]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[1100]	rmse=0.021333	lr=0.351769
[1,0]<stderr>:INFO:root:Epoch[64] Batch[1200] Loss[5.034]
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[1200]	rmse=0.021346	lr=0.351646
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0] Batch[1251]	Time cost=398.93	Train-metric=0.021346
[1,0]<stderr>:INFO:root:Epoch[64]	Speed: 3211.17 samples/sec
[1,0]<stderr>:INFO:root:Epoch[64] Rank[0]	Validation-accuracy=0.587840	Validation-top_k_accuracy_5=0.823020
[1,0]<stderr>:INFO:root:Epoch[65] Batch[100] Loss[3.222]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[100]	rmse=0.021190	lr=0.351459
[1,0]<stderr>:INFO:root:Epoch[65] Batch[200] Loss[2.901]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[200]	rmse=0.021161	lr=0.351335
[1,0]<stderr>:INFO:root:Epoch[65] Batch[300] Loss[4.694]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[300]	rmse=0.021233	lr=0.351211
[1,0]<stderr>:INFO:root:Epoch[65] Batch[400] Loss[3.470]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[400]	rmse=0.021277	lr=0.351087
[1,0]<stderr>:INFO:root:Epoch[65] Batch[500] Loss[4.302]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[500]	rmse=0.021274	lr=0.350963
[1,0]<stderr>:INFO:root:Epoch[65] Batch[600] Loss[3.953]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[600]	rmse=0.021304	lr=0.350839
[1,0]<stderr>:INFO:root:Epoch[65] Batch[700] Loss[2.940]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[700]	rmse=0.021320	lr=0.350714
[1,0]<stderr>:INFO:root:Epoch[65] Batch[800] Loss[3.611]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[800]	rmse=0.021320	lr=0.350590
[1,0]<stderr>:INFO:root:Epoch[65] Batch[900] Loss[4.573]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[900]	rmse=0.021313	lr=0.350465
[1,0]<stderr>:INFO:root:Epoch[65] Batch[1000] Loss[4.632]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[1000]	rmse=0.021309	lr=0.350340
[1,0]<stderr>:INFO:root:Epoch[65] Batch[1100] Loss[5.208]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[1100]	rmse=0.021330	lr=0.350215
[1,0]<stderr>:INFO:root:Epoch[65] Batch[1200] Loss[2.995]
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[1200]	rmse=0.021331	lr=0.350090
[1,0]<stderr>:INFO:root:Epoch[65] Rank[0] Batch[1251]	Time cost=399.33	Train-metric=0.021329
[1,0]<stderr>:INFO:root:Epoch[65]	Speed: 3207.94 samples/sec
[1,0]<stderr>:INFO:root:Epoch[66] Batch[100] Loss[3.528]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[100]	rmse=0.021123	lr=0.349901
[1,0]<stderr>:INFO:root:Epoch[66] Batch[200] Loss[5.436]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[200]	rmse=0.021089	lr=0.349776
[1,0]<stderr>:INFO:root:Epoch[66] Batch[300] Loss[3.007]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[300]	rmse=0.021116	lr=0.349650
[1,0]<stderr>:INFO:root:Epoch[66] Batch[400] Loss[5.195]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[400]	rmse=0.021180	lr=0.349524
[1,0]<stderr>:INFO:root:Epoch[66] Batch[500] Loss[3.628]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[500]	rmse=0.021173	lr=0.349399
[1,0]<stderr>:INFO:root:Epoch[66] Batch[600] Loss[3.214]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[600]	rmse=0.021191	lr=0.349273
[1,0]<stderr>:INFO:root:Epoch[66] Batch[700] Loss[4.199]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[700]	rmse=0.021228	lr=0.349147
[1,0]<stderr>:INFO:root:Epoch[66] Batch[800] Loss[4.516]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[800]	rmse=0.021232	lr=0.349020
[1,0]<stderr>:INFO:root:Epoch[66] Batch[900] Loss[4.603]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[900]	rmse=0.021223	lr=0.348894
[1,0]<stderr>:INFO:root:Epoch[66] Batch[1000] Loss[3.082]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[1000]	rmse=0.021220	lr=0.348767
[1,0]<stderr>:INFO:root:Epoch[66] Batch[1100] Loss[5.140]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[1100]	rmse=0.021232	lr=0.348641
[1,0]<stderr>:INFO:root:Epoch[66] Batch[1200] Loss[4.200]
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[1200]	rmse=0.021255	lr=0.348514
[1,0]<stderr>:INFO:root:Epoch[66] Rank[0] Batch[1251]	Time cost=398.66	Train-metric=0.021269
[1,0]<stderr>:INFO:root:Epoch[66]	Speed: 3213.30 samples/sec
[1,0]<stderr>:INFO:root:Epoch[67] Batch[100] Loss[3.276]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[100]	rmse=0.021279	lr=0.348322
[1,0]<stderr>:INFO:root:Epoch[67] Batch[200] Loss[4.205]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[200]	rmse=0.021301	lr=0.348195
[1,0]<stderr>:INFO:root:Epoch[67] Batch[300] Loss[3.440]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[300]	rmse=0.021264	lr=0.348068
[1,0]<stderr>:INFO:root:Epoch[67] Batch[400] Loss[3.153]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[400]	rmse=0.021274	lr=0.347941
[1,0]<stderr>:INFO:root:Epoch[67] Batch[500] Loss[4.841]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[500]	rmse=0.021271	lr=0.347813
[1,0]<stderr>:INFO:root:Epoch[67] Batch[600] Loss[2.965]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[600]	rmse=0.021285	lr=0.347685
[1,0]<stderr>:INFO:root:Epoch[67] Batch[700] Loss[5.435]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[700]	rmse=0.021289	lr=0.347558
[1,0]<stderr>:INFO:root:Epoch[67] Batch[800] Loss[3.079]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[800]	rmse=0.021292	lr=0.347430
[1,0]<stderr>:INFO:root:Epoch[67] Batch[900] Loss[3.201]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[900]	rmse=0.021311	lr=0.347302
[1,0]<stderr>:INFO:root:Epoch[67] Batch[1000] Loss[5.390]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[1000]	rmse=0.021309	lr=0.347174
[1,0]<stderr>:INFO:root:Epoch[67] Batch[1100] Loss[4.223]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[1100]	rmse=0.021308	lr=0.347045
[1,0]<stderr>:INFO:root:Epoch[67] Batch[1200] Loss[3.049]
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[1200]	rmse=0.021320	lr=0.346917
[1,0]<stderr>:INFO:root:Epoch[67] Rank[0] Batch[1251]	Time cost=399.24	Train-metric=0.021314
[1,0]<stderr>:INFO:root:Epoch[67]	Speed: 3208.64 samples/sec
[1,0]<stderr>:INFO:root:Epoch[68] Batch[100] Loss[3.093]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[100]	rmse=0.021109	lr=0.346723
[1,0]<stderr>:INFO:root:Epoch[68] Batch[200] Loss[3.422]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[200]	rmse=0.021265	lr=0.346594
[1,0]<stderr>:INFO:root:Epoch[68] Batch[300] Loss[4.928]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[300]	rmse=0.021353	lr=0.346465
[1,0]<stderr>:INFO:root:Epoch[68] Batch[400] Loss[2.947]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[400]	rmse=0.021342	lr=0.346336
[1,0]<stderr>:INFO:root:Epoch[68] Batch[500] Loss[5.116]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[500]	rmse=0.021278	lr=0.346207
[1,0]<stderr>:INFO:root:Epoch[68] Batch[600] Loss[2.848]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[600]	rmse=0.021275	lr=0.346078
[1,0]<stderr>:INFO:root:Epoch[68] Batch[700] Loss[2.993]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[700]	rmse=0.021263	lr=0.345948
[1,0]<stderr>:INFO:root:Epoch[68] Batch[800] Loss[3.255]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[800]	rmse=0.021274	lr=0.345819
[1,0]<stderr>:INFO:root:Epoch[68] Batch[900] Loss[3.690]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[900]	rmse=0.021303	lr=0.345689
[1,0]<stderr>:INFO:root:Epoch[68] Batch[1000] Loss[4.824]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[1000]	rmse=0.021289	lr=0.345559
[1,0]<stderr>:INFO:root:Epoch[68] Batch[1100] Loss[2.946]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[1100]	rmse=0.021280	lr=0.345429
[1,0]<stderr>:INFO:root:Epoch[68] Batch[1200] Loss[3.037]
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[1200]	rmse=0.021265	lr=0.345299
[1,0]<stderr>:INFO:root:Epoch[68] Rank[0] Batch[1251]	Time cost=400.88	Train-metric=0.021266
[1,0]<stderr>:INFO:root:Epoch[68]	Speed: 3195.50 samples/sec
[1,0]<stderr>:INFO:root:Epoch[69] Batch[100] Loss[3.196]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[100]	rmse=0.021378	lr=0.345102
[1,0]<stderr>:INFO:root:Epoch[69] Batch[200] Loss[3.895]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[200]	rmse=0.021290	lr=0.344972
[1,0]<stderr>:INFO:root:Epoch[69] Batch[300] Loss[3.527]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[300]	rmse=0.021307	lr=0.344842
[1,0]<stderr>:INFO:root:Epoch[69] Batch[400] Loss[3.099]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[400]	rmse=0.021281	lr=0.344711
[1,0]<stderr>:INFO:root:Epoch[69] Batch[500] Loss[3.071]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[500]	rmse=0.021276	lr=0.344580
[1,0]<stderr>:INFO:root:Epoch[69] Batch[600] Loss[5.183]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[600]	rmse=0.021281	lr=0.344449
[1,0]<stderr>:INFO:root:Epoch[69] Batch[700] Loss[5.385]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[700]	rmse=0.021276	lr=0.344318
[1,0]<stderr>:INFO:root:Epoch[69] Batch[800] Loss[5.155]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[800]	rmse=0.021296	lr=0.344187
[1,0]<stderr>:INFO:root:Epoch[69] Batch[900] Loss[3.066]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[900]	rmse=0.021322	lr=0.344056
[1,0]<stderr>:INFO:root:Epoch[69] Batch[1000] Loss[4.413]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[1000]	rmse=0.021325	lr=0.343924
[1,0]<stderr>:INFO:root:Epoch[69] Batch[1100] Loss[2.971]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[1100]	rmse=0.021327	lr=0.343793
[1,0]<stderr>:INFO:root:Epoch[69] Batch[1200] Loss[3.079]
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[1200]	rmse=0.021323	lr=0.343661
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0] Batch[1251]	Time cost=402.02	Train-metric=0.021322
[1,0]<stderr>:INFO:root:Epoch[69]	Speed: 3186.47 samples/sec
[1,0]<stderr>:INFO:root:Epoch[69] Rank[0]	Validation-accuracy=0.591940	Validation-top_k_accuracy_5=0.828400
[1,0]<stderr>:INFO:root:Epoch[70] Batch[100] Loss[2.697]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[100]	rmse=0.021141	lr=0.343462
[1,0]<stderr>:INFO:root:Epoch[70] Batch[200] Loss[3.745]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[200]	rmse=0.021229	lr=0.343330
[1,0]<stderr>:INFO:root:Epoch[70] Batch[300] Loss[5.397]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[300]	rmse=0.021204	lr=0.343198
[1,0]<stderr>:INFO:root:Epoch[70] Batch[400] Loss[3.114]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[400]	rmse=0.021178	lr=0.343065
[1,0]<stderr>:INFO:root:Epoch[70] Batch[500] Loss[3.229]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[500]	rmse=0.021205	lr=0.342933
[1,0]<stderr>:INFO:root:Epoch[70] Batch[600] Loss[2.935]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[600]	rmse=0.021179	lr=0.342801
[1,0]<stderr>:INFO:root:Epoch[70] Batch[700] Loss[3.349]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[700]	rmse=0.021188	lr=0.342668
[1,0]<stderr>:INFO:root:Epoch[70] Batch[800] Loss[3.097]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[800]	rmse=0.021206	lr=0.342535
[1,0]<stderr>:INFO:root:Epoch[70] Batch[900] Loss[3.039]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[900]	rmse=0.021219	lr=0.342402
[1,0]<stderr>:INFO:root:Epoch[70] Batch[1000] Loss[2.870]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[1000]	rmse=0.021213	lr=0.342269
[1,0]<stderr>:INFO:root:Epoch[70] Batch[1100] Loss[3.391]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[1100]	rmse=0.021221	lr=0.342136
[1,0]<stderr>:INFO:root:Epoch[70] Batch[1200] Loss[3.243]
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[1200]	rmse=0.021224	lr=0.342003
[1,0]<stderr>:INFO:root:Epoch[70] Rank[0] Batch[1251]	Time cost=400.80	Train-metric=0.021229
[1,0]<stderr>:INFO:root:Epoch[70]	Speed: 3196.20 samples/sec
[1,0]<stderr>:INFO:root:Epoch[71] Batch[100] Loss[2.844]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[100]	rmse=0.021253	lr=0.341801
[1,0]<stderr>:INFO:root:Epoch[71] Batch[200] Loss[5.391]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[200]	rmse=0.021241	lr=0.341668
[1,0]<stderr>:INFO:root:Epoch[71] Batch[300] Loss[3.576]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[300]	rmse=0.021199	lr=0.341534
[1,0]<stderr>:INFO:root:Epoch[71] Batch[400] Loss[3.064]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[400]	rmse=0.021229	lr=0.341400
[1,0]<stderr>:INFO:root:Epoch[71] Batch[500] Loss[4.663]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[500]	rmse=0.021236	lr=0.341266
[1,0]<stderr>:INFO:root:Epoch[71] Batch[600] Loss[3.194]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[600]	rmse=0.021245	lr=0.341132
[1,0]<stderr>:INFO:root:Epoch[71] Batch[700] Loss[2.812]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[700]	rmse=0.021237	lr=0.340998
[1,0]<stderr>:INFO:root:Epoch[71] Batch[800] Loss[5.257]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[800]	rmse=0.021252	lr=0.340863
[1,0]<stderr>:INFO:root:Epoch[71] Batch[900] Loss[2.988]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[900]	rmse=0.021270	lr=0.340729
[1,0]<stderr>:INFO:root:Epoch[71] Batch[1000] Loss[3.046]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[1000]	rmse=0.021283	lr=0.340594
[1,0]<stderr>:INFO:root:Epoch[71] Batch[1100] Loss[3.510]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[1100]	rmse=0.021281	lr=0.340459
[1,0]<stderr>:INFO:root:Epoch[71] Batch[1200] Loss[2.969]
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[1200]	rmse=0.021292	lr=0.340324
[1,0]<stderr>:INFO:root:Epoch[71] Rank[0] Batch[1251]	Time cost=405.20	Train-metric=0.021295
[1,0]<stderr>:INFO:root:Epoch[71]	Speed: 3161.46 samples/sec
[1,0]<stderr>:INFO:root:Epoch[72] Batch[100] Loss[3.098]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[100]	rmse=0.021329	lr=0.340121
[1,0]<stderr>:INFO:root:Epoch[72] Batch[200] Loss[2.871]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[200]	rmse=0.021146	lr=0.339985
[1,0]<stderr>:INFO:root:Epoch[72] Batch[300] Loss[5.307]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[300]	rmse=0.021173	lr=0.339850
[1,0]<stderr>:INFO:root:Epoch[72] Batch[400] Loss[2.870]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[400]	rmse=0.021171	lr=0.339715
[1,0]<stderr>:INFO:root:Epoch[72] Batch[500] Loss[4.046]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[500]	rmse=0.021166	lr=0.339579
[1,0]<stderr>:INFO:root:Epoch[72] Batch[600] Loss[3.337]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[600]	rmse=0.021182	lr=0.339443
[1,0]<stderr>:INFO:root:Epoch[72] Batch[700] Loss[3.216]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[700]	rmse=0.021196	lr=0.339307
[1,0]<stderr>:INFO:root:Epoch[72] Batch[800] Loss[3.160]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[800]	rmse=0.021190	lr=0.339172
[1,0]<stderr>:INFO:root:Epoch[72] Batch[900] Loss[2.911]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[900]	rmse=0.021172	lr=0.339035
[1,0]<stderr>:INFO:root:Epoch[72] Batch[1000] Loss[3.244]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[1000]	rmse=0.021176	lr=0.338899
[1,0]<stderr>:INFO:root:Epoch[72] Batch[1100] Loss[3.153]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[1100]	rmse=0.021182	lr=0.338763
[1,0]<stderr>:INFO:root:Epoch[72] Batch[1200] Loss[3.076]
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[1200]	rmse=0.021200	lr=0.338627
[1,0]<stderr>:INFO:root:Epoch[72] Rank[0] Batch[1251]	Time cost=400.68	Train-metric=0.021204
[1,0]<stderr>:INFO:root:Epoch[72]	Speed: 3197.15 samples/sec
[1,0]<stderr>:INFO:root:Epoch[73] Batch[100] Loss[3.026]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[100]	rmse=0.021346	lr=0.338420
[1,0]<stderr>:INFO:root:Epoch[73] Batch[200] Loss[5.237]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[200]	rmse=0.021390	lr=0.338283
[1,0]<stderr>:INFO:root:Epoch[73] Batch[300] Loss[4.965]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[300]	rmse=0.021267	lr=0.338147
[1,0]<stderr>:INFO:root:Epoch[73] Batch[400] Loss[2.948]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[400]	rmse=0.021285	lr=0.338010
[1,0]<stderr>:INFO:root:Epoch[73] Batch[500] Loss[3.165]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[500]	rmse=0.021254	lr=0.337872
[1,0]<stderr>:INFO:root:Epoch[73] Batch[600] Loss[3.591]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[600]	rmse=0.021269	lr=0.337735
[1,0]<stderr>:INFO:root:Epoch[73] Batch[700] Loss[5.228]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[700]	rmse=0.021284	lr=0.337598
[1,0]<stderr>:INFO:root:Epoch[73] Batch[800] Loss[5.454]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[800]	rmse=0.021309	lr=0.337460
[1,0]<stderr>:INFO:root:Epoch[73] Batch[900] Loss[4.237]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[900]	rmse=0.021316	lr=0.337323
[1,0]<stderr>:INFO:root:Epoch[73] Batch[1000] Loss[2.674]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[1000]	rmse=0.021301	lr=0.337185
[1,0]<stderr>:INFO:root:Epoch[73] Batch[1100] Loss[4.007]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[1100]	rmse=0.021310	lr=0.337047
[1,0]<stderr>:INFO:root:Epoch[73] Batch[1200] Loss[2.827]
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[1200]	rmse=0.021311	lr=0.336909
[1,0]<stderr>:INFO:root:Epoch[73] Rank[0] Batch[1251]	Time cost=400.25	Train-metric=0.021305
[1,0]<stderr>:INFO:root:Epoch[73]	Speed: 3200.54 samples/sec
[1,0]<stderr>:INFO:root:Epoch[74] Batch[100] Loss[2.757]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[100]	rmse=0.021104	lr=0.336701
[1,0]<stderr>:INFO:root:Epoch[74] Batch[200] Loss[4.952]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[200]	rmse=0.021145	lr=0.336562
[1,0]<stderr>:INFO:root:Epoch[74] Batch[300] Loss[3.083]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[300]	rmse=0.021183	lr=0.336424
[1,0]<stderr>:INFO:root:Epoch[74] Batch[400] Loss[2.756]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[400]	rmse=0.021252	lr=0.336285
[1,0]<stderr>:INFO:root:Epoch[74] Batch[500] Loss[2.939]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[500]	rmse=0.021242	lr=0.336147
[1,0]<stderr>:INFO:root:Epoch[74] Batch[600] Loss[3.494]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[600]	rmse=0.021248	lr=0.336008
[1,0]<stderr>:INFO:root:Epoch[74] Batch[700] Loss[2.777]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[700]	rmse=0.021216	lr=0.335869
[1,0]<stderr>:INFO:root:Epoch[74] Batch[800] Loss[4.254]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[800]	rmse=0.021236	lr=0.335730
[1,0]<stderr>:INFO:root:Epoch[74] Batch[900] Loss[3.102]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[900]	rmse=0.021231	lr=0.335591
[1,0]<stderr>:INFO:root:Epoch[74] Batch[1000] Loss[2.863]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[1000]	rmse=0.021218	lr=0.335451
[1,0]<stderr>:INFO:root:Epoch[74] Batch[1100] Loss[3.175]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[1100]	rmse=0.021231	lr=0.335312
[1,0]<stderr>:INFO:root:Epoch[74] Batch[1200] Loss[5.522]
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[1200]	rmse=0.021241	lr=0.335173
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0] Batch[1251]	Time cost=399.05	Train-metric=0.021237
[1,0]<stderr>:INFO:root:Epoch[74]	Speed: 3210.22 samples/sec
[1,0]<stderr>:INFO:root:Epoch[74] Rank[0]	Validation-accuracy=0.585600	Validation-top_k_accuracy_5=0.821700
[1,0]<stderr>:INFO:root:Epoch[75] Batch[100] Loss[2.875]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[100]	rmse=0.021281	lr=0.334962
[1,0]<stderr>:INFO:root:Epoch[75] Batch[200] Loss[3.123]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[200]	rmse=0.021141	lr=0.334822
[1,0]<stderr>:INFO:root:Epoch[75] Batch[300] Loss[2.793]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[300]	rmse=0.021136	lr=0.334682
[1,0]<stderr>:INFO:root:Epoch[75] Batch[400] Loss[2.995]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[400]	rmse=0.021057	lr=0.334542
[1,0]<stderr>:INFO:root:Epoch[75] Batch[500] Loss[3.086]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[500]	rmse=0.021140	lr=0.334402
[1,0]<stderr>:INFO:root:Epoch[75] Batch[600] Loss[3.471]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[600]	rmse=0.021142	lr=0.334261
[1,0]<stderr>:INFO:root:Epoch[75] Batch[700] Loss[3.012]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[700]	rmse=0.021153	lr=0.334121
[1,0]<stderr>:INFO:root:Epoch[75] Batch[800] Loss[5.481]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[800]	rmse=0.021148	lr=0.333980
[1,0]<stderr>:INFO:root:Epoch[75] Batch[900] Loss[3.524]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[900]	rmse=0.021164	lr=0.333840
[1,0]<stderr>:INFO:root:Epoch[75] Batch[1000] Loss[3.039]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[1000]	rmse=0.021184	lr=0.333699
[1,0]<stderr>:INFO:root:Epoch[75] Batch[1100] Loss[5.142]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[1100]	rmse=0.021169	lr=0.333558
[1,0]<stderr>:INFO:root:Epoch[75] Batch[1200] Loss[3.558]
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[1200]	rmse=0.021163	lr=0.333417
[1,0]<stderr>:INFO:root:Epoch[75] Rank[0] Batch[1251]	Time cost=398.86	Train-metric=0.021176
[1,0]<stderr>:INFO:root:Epoch[75]	Speed: 3211.74 samples/sec
[1,0]<stderr>:INFO:root:Epoch[76] Batch[100] Loss[3.200]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[100]	rmse=0.021257	lr=0.333204
[1,0]<stderr>:INFO:root:Epoch[76] Batch[200] Loss[3.357]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[200]	rmse=0.021359	lr=0.333063
[1,0]<stderr>:INFO:root:Epoch[76] Batch[300] Loss[4.021]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[300]	rmse=0.021267	lr=0.332921
[1,0]<stderr>:INFO:root:Epoch[76] Batch[400] Loss[2.919]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[400]	rmse=0.021288	lr=0.332780
[1,0]<stderr>:INFO:root:Epoch[76] Batch[500] Loss[5.363]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[500]	rmse=0.021243	lr=0.332638
[1,0]<stderr>:INFO:root:Epoch[76] Batch[600] Loss[5.322]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[600]	rmse=0.021244	lr=0.332496
[1,0]<stderr>:INFO:root:Epoch[76] Batch[700] Loss[3.313]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[700]	rmse=0.021243	lr=0.332354
[1,0]<stderr>:INFO:root:Epoch[76] Batch[800] Loss[5.376]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[800]	rmse=0.021232	lr=0.332212
[1,0]<stderr>:INFO:root:Epoch[76] Batch[900] Loss[3.013]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[900]	rmse=0.021232	lr=0.332070
[1,0]<stderr>:INFO:root:Epoch[76] Batch[1000] Loss[5.460]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[1000]	rmse=0.021251	lr=0.331928
[1,0]<stderr>:INFO:root:Epoch[76] Batch[1100] Loss[5.197]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[1100]	rmse=0.021244	lr=0.331785
[1,0]<stderr>:INFO:root:Epoch[76] Batch[1200] Loss[5.002]
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[1200]	rmse=0.021256	lr=0.331643
[1,0]<stderr>:INFO:root:Epoch[76] Rank[0] Batch[1251]	Time cost=398.67	Train-metric=0.021262
[1,0]<stderr>:INFO:root:Epoch[76]	Speed: 3213.22 samples/sec
[1,0]<stderr>:INFO:root:Epoch[77] Batch[100] Loss[2.908]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[100]	rmse=0.021166	lr=0.331427
[1,0]<stderr>:INFO:root:Epoch[77] Batch[200] Loss[5.103]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[200]	rmse=0.021201	lr=0.331285
[1,0]<stderr>:INFO:root:Epoch[77] Batch[300] Loss[3.030]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[300]	rmse=0.021170	lr=0.331142
[1,0]<stderr>:INFO:root:Epoch[77] Batch[400] Loss[2.869]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[400]	rmse=0.021171	lr=0.330999
[1,0]<stderr>:INFO:root:Epoch[77] Batch[500] Loss[2.973]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[500]	rmse=0.021144	lr=0.330855
[1,0]<stderr>:INFO:root:Epoch[77] Batch[600] Loss[3.242]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[600]	rmse=0.021138	lr=0.330712
[1,0]<stderr>:INFO:root:Epoch[77] Batch[700] Loss[5.130]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[700]	rmse=0.021168	lr=0.330569
[1,0]<stderr>:INFO:root:Epoch[77] Batch[800] Loss[3.109]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[800]	rmse=0.021173	lr=0.330425
[1,0]<stderr>:INFO:root:Epoch[77] Batch[900] Loss[4.981]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[900]	rmse=0.021187	lr=0.330282
[1,0]<stderr>:INFO:root:Epoch[77] Batch[1000] Loss[5.067]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[1000]	rmse=0.021217	lr=0.330138
[1,0]<stderr>:INFO:root:Epoch[77] Batch[1100] Loss[4.083]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[1100]	rmse=0.021198	lr=0.329994
[1,0]<stderr>:INFO:root:Epoch[77] Batch[1200] Loss[3.672]
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[1200]	rmse=0.021196	lr=0.329850
[1,0]<stderr>:INFO:root:Epoch[77] Rank[0] Batch[1251]	Time cost=399.88	Train-metric=0.021192
[1,0]<stderr>:INFO:root:Epoch[77]	Speed: 3203.48 samples/sec
[1,0]<stderr>:INFO:root:Epoch[78] Batch[100] Loss[2.922]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[100]	rmse=0.021180	lr=0.329632
[1,0]<stderr>:INFO:root:Epoch[78] Batch[200] Loss[2.858]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[200]	rmse=0.021156	lr=0.329488
[1,0]<stderr>:INFO:root:Epoch[78] Batch[300] Loss[3.763]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[300]	rmse=0.021203	lr=0.329344
[1,0]<stderr>:INFO:root:Epoch[78] Batch[400] Loss[2.992]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[400]	rmse=0.021215	lr=0.329199
[1,0]<stderr>:INFO:root:Epoch[78] Batch[500] Loss[2.976]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[500]	rmse=0.021255	lr=0.329055
[1,0]<stderr>:INFO:root:Epoch[78] Batch[600] Loss[3.264]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[600]	rmse=0.021249	lr=0.328910
[1,0]<stderr>:INFO:root:Epoch[78] Batch[700] Loss[3.068]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[700]	rmse=0.021248	lr=0.328765
[1,0]<stderr>:INFO:root:Epoch[78] Batch[800] Loss[2.909]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[800]	rmse=0.021260	lr=0.328620
[1,0]<stderr>:INFO:root:Epoch[78] Batch[900] Loss[3.494]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[900]	rmse=0.021247	lr=0.328475
[1,0]<stderr>:INFO:root:Epoch[78] Batch[1000] Loss[4.751]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[1000]	rmse=0.021256	lr=0.328330
[1,0]<stderr>:INFO:root:Epoch[78] Batch[1100] Loss[2.974]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[1100]	rmse=0.021255	lr=0.328184
[1,0]<stderr>:INFO:root:Epoch[78] Batch[1200] Loss[2.889]
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[1200]	rmse=0.021259	lr=0.328039
[1,0]<stderr>:INFO:root:Epoch[78] Rank[0] Batch[1251]	Time cost=399.02	Train-metric=0.021267
[1,0]<stderr>:INFO:root:Epoch[78]	Speed: 3210.41 samples/sec
[1,0]<stderr>:INFO:root:Epoch[79] Batch[100] Loss[2.936]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[100]	rmse=0.021078	lr=0.327819
[1,0]<stderr>:INFO:root:Epoch[79] Batch[200] Loss[3.049]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[200]	rmse=0.021160	lr=0.327674
[1,0]<stderr>:INFO:root:Epoch[79] Batch[300] Loss[3.479]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[300]	rmse=0.021093	lr=0.327528
[1,0]<stderr>:INFO:root:Epoch[79] Batch[400] Loss[4.938]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[400]	rmse=0.021120	lr=0.327382
[1,0]<stderr>:INFO:root:Epoch[79] Batch[500] Loss[3.743]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[500]	rmse=0.021177	lr=0.327236
[1,0]<stderr>:INFO:root:Epoch[79] Batch[600] Loss[4.719]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[600]	rmse=0.021139	lr=0.327090
[1,0]<stderr>:INFO:root:Epoch[79] Batch[700] Loss[3.353]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[700]	rmse=0.021146	lr=0.326943
[1,0]<stderr>:INFO:root:Epoch[79] Batch[800] Loss[5.276]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[800]	rmse=0.021179	lr=0.326797
[1,0]<stderr>:INFO:root:Epoch[79] Batch[900] Loss[3.222]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[900]	rmse=0.021170	lr=0.326650
[1,0]<stderr>:INFO:root:Epoch[79] Batch[1000] Loss[3.096]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[1000]	rmse=0.021182	lr=0.326504
[1,0]<stderr>:INFO:root:Epoch[79] Batch[1100] Loss[5.058]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[1100]	rmse=0.021200	lr=0.326357
[1,0]<stderr>:INFO:root:Epoch[79] Batch[1200] Loss[3.500]
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[1200]	rmse=0.021190	lr=0.326210
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0] Batch[1251]	Time cost=399.34	Train-metric=0.021195
[1,0]<stderr>:INFO:root:Epoch[79]	Speed: 3207.89 samples/sec
[1,0]<stderr>:INFO:root:Epoch[79] Rank[0]	Validation-accuracy=0.598280	Validation-top_k_accuracy_5=0.833720
[1,0]<stderr>:INFO:root:Epoch[80] Batch[100] Loss[4.197]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[100]	rmse=0.021067	lr=0.325988
[1,0]<stderr>:INFO:root:Epoch[80] Batch[200] Loss[3.863]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[200]	rmse=0.021048	lr=0.325841
[1,0]<stderr>:INFO:root:Epoch[80] Batch[300] Loss[2.943]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[300]	rmse=0.020991	lr=0.325694
[1,0]<stderr>:INFO:root:Epoch[80] Batch[400] Loss[2.894]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[400]	rmse=0.021035	lr=0.325546
[1,0]<stderr>:INFO:root:Epoch[80] Batch[500] Loss[5.414]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[500]	rmse=0.021080	lr=0.325399
[1,0]<stderr>:INFO:root:Epoch[80] Batch[600] Loss[2.842]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[600]	rmse=0.021117	lr=0.325251
[1,0]<stderr>:INFO:root:Epoch[80] Batch[700] Loss[3.288]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[700]	rmse=0.021099	lr=0.325104
[1,0]<stderr>:INFO:root:Epoch[80] Batch[800] Loss[2.980]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[800]	rmse=0.021108	lr=0.324956
[1,0]<stderr>:INFO:root:Epoch[80] Batch[900] Loss[3.093]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[900]	rmse=0.021132	lr=0.324808
[1,0]<stderr>:INFO:root:Epoch[80] Batch[1000] Loss[5.229]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[1000]	rmse=0.021170	lr=0.324660
[1,0]<stderr>:INFO:root:Epoch[80] Batch[1100] Loss[2.964]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[1100]	rmse=0.021163	lr=0.324512
[1,0]<stderr>:INFO:root:Epoch[80] Batch[1200] Loss[3.075]
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[1200]	rmse=0.021178	lr=0.324363
[1,0]<stderr>:INFO:root:Epoch[80] Rank[0] Batch[1251]	Time cost=398.42	Train-metric=0.021183
[1,0]<stderr>:INFO:root:Epoch[80]	Speed: 3215.30 samples/sec
[1,0]<stderr>:INFO:root:Epoch[81] Batch[100] Loss[2.797]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[100]	rmse=0.021185	lr=0.324139
[1,0]<stderr>:INFO:root:Epoch[81] Batch[200] Loss[2.879]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[200]	rmse=0.021193	lr=0.323991
[1,0]<stderr>:INFO:root:Epoch[81] Batch[300] Loss[5.346]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[300]	rmse=0.021122	lr=0.323842
[1,0]<stderr>:INFO:root:Epoch[81] Batch[400] Loss[3.350]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[400]	rmse=0.021106	lr=0.323693
[1,0]<stderr>:INFO:root:Epoch[81] Batch[500] Loss[3.180]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[500]	rmse=0.021121	lr=0.323545
[1,0]<stderr>:INFO:root:Epoch[81] Batch[600] Loss[3.052]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[600]	rmse=0.021110	lr=0.323396
[1,0]<stderr>:INFO:root:Epoch[81] Batch[700] Loss[4.462]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[700]	rmse=0.021128	lr=0.323247
[1,0]<stderr>:INFO:root:Epoch[81] Batch[800] Loss[3.103]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[800]	rmse=0.021142	lr=0.323097
[1,0]<stderr>:INFO:root:Epoch[81] Batch[900] Loss[4.484]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[900]	rmse=0.021144	lr=0.322948
[1,0]<stderr>:INFO:root:Epoch[81] Batch[1000] Loss[3.138]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[1000]	rmse=0.021140	lr=0.322799
[1,0]<stderr>:INFO:root:Epoch[81] Batch[1100] Loss[2.672]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[1100]	rmse=0.021130	lr=0.322649
[1,0]<stderr>:INFO:root:Epoch[81] Batch[1200] Loss[5.064]
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[1200]	rmse=0.021145	lr=0.322499
[1,0]<stderr>:INFO:root:Epoch[81] Rank[0] Batch[1251]	Time cost=399.23	Train-metric=0.021153
[1,0]<stderr>:INFO:root:Epoch[81]	Speed: 3208.72 samples/sec
[1,0]<stderr>:INFO:root:Epoch[82] Batch[100] Loss[4.081]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[100]	rmse=0.021187	lr=0.322273
[1,0]<stderr>:INFO:root:Epoch[82] Batch[200] Loss[3.920]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[200]	rmse=0.021152	lr=0.322123
[1,0]<stderr>:INFO:root:Epoch[82] Batch[300] Loss[3.294]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[300]	rmse=0.021132	lr=0.321973
[1,0]<stderr>:INFO:root:Epoch[82] Batch[400] Loss[3.689]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[400]	rmse=0.021157	lr=0.321823
[1,0]<stderr>:INFO:root:Epoch[82] Batch[500] Loss[2.954]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[500]	rmse=0.021180	lr=0.321673
[1,0]<stderr>:INFO:root:Epoch[82] Batch[600] Loss[5.294]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[600]	rmse=0.021217	lr=0.321523
[1,0]<stderr>:INFO:root:Epoch[82] Batch[700] Loss[2.972]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[700]	rmse=0.021237	lr=0.321372
[1,0]<stderr>:INFO:root:Epoch[82] Batch[800] Loss[2.834]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[800]	rmse=0.021238	lr=0.321221
[1,0]<stderr>:INFO:root:Epoch[82] Batch[900] Loss[3.453]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[900]	rmse=0.021245	lr=0.321071
[1,0]<stderr>:INFO:root:Epoch[82] Batch[1000] Loss[3.437]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[1000]	rmse=0.021245	lr=0.320920
[1,0]<stderr>:INFO:root:Epoch[82] Batch[1100] Loss[3.439]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[1100]	rmse=0.021233	lr=0.320769
[1,0]<stderr>:INFO:root:Epoch[82] Batch[1200] Loss[3.055]
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[1200]	rmse=0.021236	lr=0.320618
[1,0]<stderr>:INFO:root:Epoch[82] Rank[0] Batch[1251]	Time cost=399.47	Train-metric=0.021230
[1,0]<stderr>:INFO:root:Epoch[82]	Speed: 3206.77 samples/sec
[1,0]<stderr>:INFO:root:Epoch[83] Batch[100] Loss[2.815]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[100]	rmse=0.021168	lr=0.320390
[1,0]<stderr>:INFO:root:Epoch[83] Batch[200] Loss[3.359]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[200]	rmse=0.021143	lr=0.320239
[1,0]<stderr>:INFO:root:Epoch[83] Batch[300] Loss[3.578]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[300]	rmse=0.021122	lr=0.320087
[1,0]<stderr>:INFO:root:Epoch[83] Batch[400] Loss[2.720]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[400]	rmse=0.021152	lr=0.319936
[1,0]<stderr>:INFO:root:Epoch[83] Batch[500] Loss[2.616]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[500]	rmse=0.021107	lr=0.319784
[1,0]<stderr>:INFO:root:Epoch[83] Batch[600] Loss[4.774]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[600]	rmse=0.021123	lr=0.319632
[1,0]<stderr>:INFO:root:Epoch[83] Batch[700] Loss[4.192]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[700]	rmse=0.021113	lr=0.319481
[1,0]<stderr>:INFO:root:Epoch[83] Batch[800] Loss[2.774]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[800]	rmse=0.021117	lr=0.319329
[1,0]<stderr>:INFO:root:Epoch[83] Batch[900] Loss[2.820]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[900]	rmse=0.021137	lr=0.319177
[1,0]<stderr>:INFO:root:Epoch[83] Batch[1000] Loss[3.955]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[1000]	rmse=0.021143	lr=0.319025
[1,0]<stderr>:INFO:root:Epoch[83] Batch[1100] Loss[2.662]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[1100]	rmse=0.021131	lr=0.318872
[1,0]<stderr>:INFO:root:Epoch[83] Batch[1200] Loss[3.917]
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[1200]	rmse=0.021122	lr=0.318720
[1,0]<stderr>:INFO:root:Epoch[83] Rank[0] Batch[1251]	Time cost=398.98	Train-metric=0.021127
[1,0]<stderr>:INFO:root:Epoch[83]	Speed: 3210.76 samples/sec
[1,0]<stderr>:INFO:root:Epoch[84] Batch[100] Loss[3.633]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[100]	rmse=0.021126	lr=0.318490
[1,0]<stderr>:INFO:root:Epoch[84] Batch[200] Loss[3.567]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[200]	rmse=0.021088	lr=0.318337
[1,0]<stderr>:INFO:root:Epoch[84] Batch[300] Loss[2.859]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[300]	rmse=0.021114	lr=0.318184
[1,0]<stderr>:INFO:root:Epoch[84] Batch[400] Loss[2.840]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[400]	rmse=0.021110	lr=0.318031
[1,0]<stderr>:INFO:root:Epoch[84] Batch[500] Loss[2.905]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[500]	rmse=0.021117	lr=0.317879
[1,0]<stderr>:INFO:root:Epoch[84] Batch[600] Loss[4.889]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[600]	rmse=0.021101	lr=0.317726
[1,0]<stderr>:INFO:root:Epoch[84] Batch[700] Loss[3.141]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[700]	rmse=0.021116	lr=0.317572
[1,0]<stderr>:INFO:root:Epoch[84] Batch[800] Loss[5.414]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[800]	rmse=0.021109	lr=0.317419
[1,0]<stderr>:INFO:root:Epoch[84] Batch[900] Loss[2.854]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[900]	rmse=0.021110	lr=0.317266
[1,0]<stderr>:INFO:root:Epoch[84] Batch[1000] Loss[3.486]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[1000]	rmse=0.021119	lr=0.317112
[1,0]<stderr>:INFO:root:Epoch[84] Batch[1100] Loss[4.454]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[1100]	rmse=0.021114	lr=0.316959
[1,0]<stderr>:INFO:root:Epoch[84] Batch[1200] Loss[3.208]
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[1200]	rmse=0.021132	lr=0.316805
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0] Batch[1251]	Time cost=398.99	Train-metric=0.021144
[1,0]<stderr>:INFO:root:Epoch[84]	Speed: 3210.63 samples/sec
[1,0]<stderr>:INFO:root:Epoch[84] Rank[0]	Validation-accuracy=0.599040	Validation-top_k_accuracy_5=0.832480
[1,0]<stderr>:INFO:root:Epoch[85] Batch[100] Loss[3.041]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[100]	rmse=0.021189	lr=0.316573
[1,0]<stderr>:INFO:root:Epoch[85] Batch[200] Loss[2.737]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[200]	rmse=0.021054	lr=0.316419
[1,0]<stderr>:INFO:root:Epoch[85] Batch[300] Loss[5.308]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[300]	rmse=0.021059	lr=0.316265
[1,0]<stderr>:INFO:root:Epoch[85] Batch[400] Loss[3.082]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[400]	rmse=0.021097	lr=0.316111
[1,0]<stderr>:INFO:root:Epoch[85] Batch[500] Loss[3.265]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[500]	rmse=0.021089	lr=0.315956
[1,0]<stderr>:INFO:root:Epoch[85] Batch[600] Loss[4.776]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[600]	rmse=0.021088	lr=0.315802
[1,0]<stderr>:INFO:root:Epoch[85] Batch[700] Loss[3.337]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[700]	rmse=0.021093	lr=0.315648
[1,0]<stderr>:INFO:root:Epoch[85] Batch[800] Loss[4.677]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[800]	rmse=0.021109	lr=0.315493
[1,0]<stderr>:INFO:root:Epoch[85] Batch[900] Loss[3.128]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[900]	rmse=0.021064	lr=0.315338
[1,0]<stderr>:INFO:root:Epoch[85] Batch[1000] Loss[2.987]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[1000]	rmse=0.021099	lr=0.315184
[1,0]<stderr>:INFO:root:Epoch[85] Batch[1100] Loss[3.727]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[1100]	rmse=0.021108	lr=0.315029
[1,0]<stderr>:INFO:root:Epoch[85] Batch[1200] Loss[4.450]
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[1200]	rmse=0.021108	lr=0.314874
[1,0]<stderr>:INFO:root:Epoch[85] Rank[0] Batch[1251]	Time cost=399.15	Train-metric=0.021105
[1,0]<stderr>:INFO:root:Epoch[85]	Speed: 3209.41 samples/sec
[1,0]<stderr>:INFO:root:Epoch[86] Batch[100] Loss[3.561]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[100]	rmse=0.021343	lr=0.314640
[1,0]<stderr>:INFO:root:Epoch[86] Batch[200] Loss[2.786]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[200]	rmse=0.021185	lr=0.314484
[1,0]<stderr>:INFO:root:Epoch[86] Batch[300] Loss[2.870]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[300]	rmse=0.021168	lr=0.314329
[1,0]<stderr>:INFO:root:Epoch[86] Batch[400] Loss[5.382]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[400]	rmse=0.021106	lr=0.314174
[1,0]<stderr>:INFO:root:Epoch[86] Batch[500] Loss[2.655]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[500]	rmse=0.021116	lr=0.314018
[1,0]<stderr>:INFO:root:Epoch[86] Batch[600] Loss[3.329]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[600]	rmse=0.021153	lr=0.313862
[1,0]<stderr>:INFO:root:Epoch[86] Batch[700] Loss[2.816]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[700]	rmse=0.021142	lr=0.313707
[1,0]<stderr>:INFO:root:Epoch[86] Batch[800] Loss[2.830]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[800]	rmse=0.021108	lr=0.313551
[1,0]<stderr>:INFO:root:Epoch[86] Batch[900] Loss[3.127]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[900]	rmse=0.021127	lr=0.313395
[1,0]<stderr>:INFO:root:Epoch[86] Batch[1000] Loss[4.704]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[1000]	rmse=0.021137	lr=0.313239
[1,0]<stderr>:INFO:root:Epoch[86] Batch[1100] Loss[3.115]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[1100]	rmse=0.021122	lr=0.313083
[1,0]<stderr>:INFO:root:Epoch[86] Batch[1200] Loss[2.784]
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[1200]	rmse=0.021115	lr=0.312926
[1,0]<stderr>:INFO:root:Epoch[86] Rank[0] Batch[1251]	Time cost=399.45	Train-metric=0.021104
[1,0]<stderr>:INFO:root:Epoch[86]	Speed: 3206.97 samples/sec
[1,0]<stderr>:INFO:root:Epoch[87] Batch[100] Loss[3.440]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[100]	rmse=0.020928	lr=0.312690
[1,0]<stderr>:INFO:root:Epoch[87] Batch[200] Loss[3.104]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[200]	rmse=0.020948	lr=0.312534
[1,0]<stderr>:INFO:root:Epoch[87] Batch[300] Loss[2.993]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[300]	rmse=0.021046	lr=0.312377
[1,0]<stderr>:INFO:root:Epoch[87] Batch[400] Loss[3.035]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[400]	rmse=0.021118	lr=0.312220
[1,0]<stderr>:INFO:root:Epoch[87] Batch[500] Loss[3.513]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[500]	rmse=0.021149	lr=0.312064
[1,0]<stderr>:INFO:root:Epoch[87] Batch[600] Loss[2.867]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[600]	rmse=0.021207	lr=0.311907
[1,0]<stderr>:INFO:root:Epoch[87] Batch[700] Loss[3.032]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[700]	rmse=0.021212	lr=0.311750
[1,0]<stderr>:INFO:root:Epoch[87] Batch[800] Loss[2.909]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[800]	rmse=0.021186	lr=0.311593
[1,0]<stderr>:INFO:root:Epoch[87] Batch[900] Loss[5.385]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[900]	rmse=0.021146	lr=0.311435
[1,0]<stderr>:INFO:root:Epoch[87] Batch[1000] Loss[4.886]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[1000]	rmse=0.021137	lr=0.311278
[1,0]<stderr>:INFO:root:Epoch[87] Batch[1100] Loss[4.888]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[1100]	rmse=0.021135	lr=0.311121
[1,0]<stderr>:INFO:root:Epoch[87] Batch[1200] Loss[3.017]
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[1200]	rmse=0.021142	lr=0.310963
[1,0]<stderr>:INFO:root:Epoch[87] Rank[0] Batch[1251]	Time cost=399.40	Train-metric=0.021155
[1,0]<stderr>:INFO:root:Epoch[87]	Speed: 3207.39 samples/sec
[1,0]<stderr>:INFO:root:Epoch[88] Batch[100] Loss[3.124]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[100]	rmse=0.021127	lr=0.310725
[1,0]<stderr>:INFO:root:Epoch[88] Batch[200] Loss[3.254]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[200]	rmse=0.021205	lr=0.310567
[1,0]<stderr>:INFO:root:Epoch[88] Batch[300] Loss[5.539]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[300]	rmse=0.021169	lr=0.310410
[1,0]<stderr>:INFO:root:Epoch[88] Batch[400] Loss[3.037]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[400]	rmse=0.021135	lr=0.310252
[1,0]<stderr>:INFO:root:Epoch[88] Batch[500] Loss[3.252]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[500]	rmse=0.021118	lr=0.310094
[1,0]<stderr>:INFO:root:Epoch[88] Batch[600] Loss[3.407]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[600]	rmse=0.021088	lr=0.309935
[1,0]<stderr>:INFO:root:Epoch[88] Batch[700] Loss[4.120]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[700]	rmse=0.021089	lr=0.309777
[1,0]<stderr>:INFO:root:Epoch[88] Batch[800] Loss[5.405]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[800]	rmse=0.021091	lr=0.309619
[1,0]<stderr>:INFO:root:Epoch[88] Batch[900] Loss[2.732]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[900]	rmse=0.021085	lr=0.309460
[1,0]<stderr>:INFO:root:Epoch[88] Batch[1000] Loss[3.361]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[1000]	rmse=0.021095	lr=0.309302
[1,0]<stderr>:INFO:root:Epoch[88] Batch[1100] Loss[3.453]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[1100]	rmse=0.021096	lr=0.309143
[1,0]<stderr>:INFO:root:Epoch[88] Batch[1200] Loss[2.680]
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[1200]	rmse=0.021111	lr=0.308984
[1,0]<stderr>:INFO:root:Epoch[88] Rank[0] Batch[1251]	Time cost=399.46	Train-metric=0.021113
[1,0]<stderr>:INFO:root:Epoch[88]	Speed: 3206.92 samples/sec
[1,0]<stderr>:INFO:root:Epoch[89] Batch[100] Loss[5.354]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[100]	rmse=0.021053	lr=0.308745
[1,0]<stderr>:INFO:root:Epoch[89] Batch[200] Loss[2.837]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[200]	rmse=0.021075	lr=0.308586
[1,0]<stderr>:INFO:root:Epoch[89] Batch[300] Loss[2.869]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[300]	rmse=0.021129	lr=0.308426
[1,0]<stderr>:INFO:root:Epoch[89] Batch[400] Loss[3.302]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[400]	rmse=0.021091	lr=0.308267
[1,0]<stderr>:INFO:root:Epoch[89] Batch[500] Loss[3.008]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[500]	rmse=0.021068	lr=0.308108
[1,0]<stderr>:INFO:root:Epoch[89] Batch[600] Loss[4.943]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[600]	rmse=0.021073	lr=0.307949
[1,0]<stderr>:INFO:root:Epoch[89] Batch[700] Loss[4.848]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[700]	rmse=0.021105	lr=0.307789
[1,0]<stderr>:INFO:root:Epoch[89] Batch[800] Loss[3.206]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[800]	rmse=0.021089	lr=0.307630
[1,0]<stderr>:INFO:root:Epoch[89] Batch[900] Loss[3.593]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[900]	rmse=0.021099	lr=0.307470
[1,0]<stderr>:INFO:root:Epoch[89] Batch[1000] Loss[3.011]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[1000]	rmse=0.021102	lr=0.307310
[1,0]<stderr>:INFO:root:Epoch[89] Batch[1100] Loss[2.955]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[1100]	rmse=0.021106	lr=0.307150
[1,0]<stderr>:INFO:root:Epoch[89] Batch[1200] Loss[3.199]
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[1200]	rmse=0.021128	lr=0.306990
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0] Batch[1251]	Time cost=399.32	Train-metric=0.021124
[1,0]<stderr>:INFO:root:Epoch[89]	Speed: 3208.03 samples/sec
[1,0]<stderr>:INFO:root:Epoch[89] Rank[0]	Validation-accuracy=0.588660	Validation-top_k_accuracy_5=0.822760
[1,0]<stderr>:INFO:root:Epoch[90] Batch[100] Loss[3.013]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[100]	rmse=0.020665	lr=0.306749
[1,0]<stderr>:INFO:root:Epoch[90] Batch[200] Loss[3.856]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[200]	rmse=0.020732	lr=0.306588
[1,0]<stderr>:INFO:root:Epoch[90] Batch[300] Loss[4.015]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[300]	rmse=0.020917	lr=0.306428
[1,0]<stderr>:INFO:root:Epoch[90] Batch[400] Loss[4.100]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[400]	rmse=0.020943	lr=0.306268
[1,0]<stderr>:INFO:root:Epoch[90] Batch[500] Loss[2.759]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[500]	rmse=0.021011	lr=0.306107
[1,0]<stderr>:INFO:root:Epoch[90] Batch[600] Loss[5.037]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[600]	rmse=0.021022	lr=0.305947
[1,0]<stderr>:INFO:root:Epoch[90] Batch[700] Loss[3.168]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[700]	rmse=0.021026	lr=0.305786
[1,0]<stderr>:INFO:root:Epoch[90] Batch[800] Loss[2.803]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[800]	rmse=0.021058	lr=0.305625
[1,0]<stderr>:INFO:root:Epoch[90] Batch[900] Loss[5.361]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[900]	rmse=0.021074	lr=0.305464
[1,0]<stderr>:INFO:root:Epoch[90] Batch[1000] Loss[2.880]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[1000]	rmse=0.021083	lr=0.305303
[1,0]<stderr>:INFO:root:Epoch[90] Batch[1100] Loss[4.901]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[1100]	rmse=0.021079	lr=0.305142
[1,0]<stderr>:INFO:root:Epoch[90] Batch[1200] Loss[5.329]
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[1200]	rmse=0.021100	lr=0.304981
[1,0]<stderr>:INFO:root:Epoch[90] Rank[0] Batch[1251]	Time cost=398.93	Train-metric=0.021105
[1,0]<stderr>:INFO:root:Epoch[90]	Speed: 3211.12 samples/sec
[1,0]<stderr>:INFO:root:Epoch[91] Batch[100] Loss[2.812]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[100]	rmse=0.021068	lr=0.304738
[1,0]<stderr>:INFO:root:Epoch[91] Batch[200] Loss[5.519]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[200]	rmse=0.020965	lr=0.304576
[1,0]<stderr>:INFO:root:Epoch[91] Batch[300] Loss[3.159]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[300]	rmse=0.021001	lr=0.304415
[1,0]<stderr>:INFO:root:Epoch[91] Batch[400] Loss[3.083]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[400]	rmse=0.020997	lr=0.304253
[1,0]<stderr>:INFO:root:Epoch[91] Batch[500] Loss[3.119]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[500]	rmse=0.021041	lr=0.304092
[1,0]<stderr>:INFO:root:Epoch[91] Batch[600] Loss[3.022]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[600]	rmse=0.021061	lr=0.303930
[1,0]<stderr>:INFO:root:Epoch[91] Batch[700] Loss[2.874]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[700]	rmse=0.021086	lr=0.303768
[1,0]<stderr>:INFO:root:Epoch[91] Batch[800] Loss[4.648]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[800]	rmse=0.021103	lr=0.303606
[1,0]<stderr>:INFO:root:Epoch[91] Batch[900] Loss[3.020]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[900]	rmse=0.021104	lr=0.303444
[1,0]<stderr>:INFO:root:Epoch[91] Batch[1000] Loss[5.217]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[1000]	rmse=0.021101	lr=0.303282
[1,0]<stderr>:INFO:root:Epoch[91] Batch[1100] Loss[4.042]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[1100]	rmse=0.021112	lr=0.303120
[1,0]<stderr>:INFO:root:Epoch[91] Batch[1200] Loss[2.755]
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[1200]	rmse=0.021115	lr=0.302957
[1,0]<stderr>:INFO:root:Epoch[91] Rank[0] Batch[1251]	Time cost=399.44	Train-metric=0.021116
[1,0]<stderr>:INFO:root:Epoch[91]	Speed: 3207.06 samples/sec
[1,0]<stderr>:INFO:root:Epoch[92] Batch[100] Loss[4.780]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[100]	rmse=0.020955	lr=0.302712
[1,0]<stderr>:INFO:root:Epoch[92] Batch[200] Loss[5.159]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[200]	rmse=0.021030	lr=0.302550
[1,0]<stderr>:INFO:root:Epoch[92] Batch[300] Loss[2.868]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[300]	rmse=0.021046	lr=0.302387
[1,0]<stderr>:INFO:root:Epoch[92] Batch[400] Loss[2.700]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[400]	rmse=0.021053	lr=0.302224
[1,0]<stderr>:INFO:root:Epoch[92] Batch[500] Loss[3.045]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[500]	rmse=0.021066	lr=0.302061
[1,0]<stderr>:INFO:root:Epoch[92] Batch[600] Loss[4.004]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[600]	rmse=0.021088	lr=0.301899
[1,0]<stderr>:INFO:root:Epoch[92] Batch[700] Loss[5.272]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[700]	rmse=0.021082	lr=0.301736
[1,0]<stderr>:INFO:root:Epoch[92] Batch[800] Loss[5.314]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[800]	rmse=0.021088	lr=0.301572
[1,0]<stderr>:INFO:root:Epoch[92] Batch[900] Loss[3.158]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[900]	rmse=0.021107	lr=0.301409
[1,0]<stderr>:INFO:root:Epoch[92] Batch[1000] Loss[3.115]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[1000]	rmse=0.021106	lr=0.301246
[1,0]<stderr>:INFO:root:Epoch[92] Batch[1100] Loss[5.352]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[1100]	rmse=0.021111	lr=0.301083
[1,0]<stderr>:INFO:root:Epoch[92] Batch[1200] Loss[2.966]
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[1200]	rmse=0.021107	lr=0.300919
[1,0]<stderr>:INFO:root:Epoch[92] Rank[0] Batch[1251]	Time cost=398.48	Train-metric=0.021107
[1,0]<stderr>:INFO:root:Epoch[92]	Speed: 3214.81 samples/sec
[1,0]<stderr>:INFO:root:Epoch[93] Batch[100] Loss[4.892]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[100]	rmse=0.020893	lr=0.300672
[1,0]<stderr>:INFO:root:Epoch[93] Batch[200] Loss[2.842]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[200]	rmse=0.020826	lr=0.300509
[1,0]<stderr>:INFO:root:Epoch[93] Batch[300] Loss[3.177]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[300]	rmse=0.020920	lr=0.300345
[1,0]<stderr>:INFO:root:Epoch[93] Batch[400] Loss[3.029]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[400]	rmse=0.020930	lr=0.300181
[1,0]<stderr>:INFO:root:Epoch[93] Batch[500] Loss[2.920]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[500]	rmse=0.020927	lr=0.300017
[1,0]<stderr>:INFO:root:Epoch[93] Batch[600] Loss[3.756]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[600]	rmse=0.020962	lr=0.299853
[1,0]<stderr>:INFO:root:Epoch[93] Batch[700] Loss[3.454]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[700]	rmse=0.020983	lr=0.299689
[1,0]<stderr>:INFO:root:Epoch[93] Batch[800] Loss[2.715]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[800]	rmse=0.021001	lr=0.299525
[1,0]<stderr>:INFO:root:Epoch[93] Batch[900] Loss[3.110]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[900]	rmse=0.021040	lr=0.299360
[1,0]<stderr>:INFO:root:Epoch[93] Batch[1000] Loss[3.830]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[1000]	rmse=0.021056	lr=0.299196
[1,0]<stderr>:INFO:root:Epoch[93] Batch[1100] Loss[3.052]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[1100]	rmse=0.021071	lr=0.299031
[1,0]<stderr>:INFO:root:Epoch[93] Batch[1200] Loss[4.080]
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[1200]	rmse=0.021080	lr=0.298867
[1,0]<stderr>:INFO:root:Epoch[93] Rank[0] Batch[1251]	Time cost=399.19	Train-metric=0.021077
[1,0]<stderr>:INFO:root:Epoch[93]	Speed: 3209.09 samples/sec
[1,0]<stderr>:INFO:root:Epoch[94] Batch[100] Loss[3.244]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[100]	rmse=0.021015	lr=0.298618
[1,0]<stderr>:INFO:root:Epoch[94] Batch[200] Loss[3.181]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[200]	rmse=0.020889	lr=0.298453
[1,0]<stderr>:INFO:root:Epoch[94] Batch[300] Loss[2.910]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[300]	rmse=0.021017	lr=0.298288
[1,0]<stderr>:INFO:root:Epoch[94] Batch[400] Loss[2.681]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[400]	rmse=0.021002	lr=0.298123
[1,0]<stderr>:INFO:root:Epoch[94] Batch[500] Loss[2.796]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[500]	rmse=0.021075	lr=0.297958
[1,0]<stderr>:INFO:root:Epoch[94] Batch[600] Loss[3.095]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[600]	rmse=0.021061	lr=0.297793
[1,0]<stderr>:INFO:root:Epoch[94] Batch[700] Loss[3.247]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[700]	rmse=0.021052	lr=0.297628
[1,0]<stderr>:INFO:root:Epoch[94] Batch[800] Loss[2.811]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[800]	rmse=0.021055	lr=0.297463
[1,0]<stderr>:INFO:root:Epoch[94] Batch[900] Loss[2.657]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[900]	rmse=0.021072	lr=0.297297
[1,0]<stderr>:INFO:root:Epoch[94] Batch[1000] Loss[3.087]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[1000]	rmse=0.021072	lr=0.297132
[1,0]<stderr>:INFO:root:Epoch[94] Batch[1100] Loss[2.945]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[1100]	rmse=0.021064	lr=0.296966
[1,0]<stderr>:INFO:root:Epoch[94] Batch[1200] Loss[2.811]
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[1200]	rmse=0.021061	lr=0.296800
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0] Batch[1251]	Time cost=400.41	Train-metric=0.021064
[1,0]<stderr>:INFO:root:Epoch[94]	Speed: 3199.28 samples/sec
[1,0]<stderr>:INFO:root:Epoch[94] Rank[0]	Validation-accuracy=0.604720	Validation-top_k_accuracy_5=0.834480
[1,0]<stderr>:INFO:root:Epoch[95] Batch[100] Loss[2.679]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[100]	rmse=0.020916	lr=0.296550
[1,0]<stderr>:INFO:root:Epoch[95] Batch[200] Loss[2.686]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[200]	rmse=0.020914	lr=0.296384
[1,0]<stderr>:INFO:root:Epoch[95] Batch[300] Loss[3.007]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[300]	rmse=0.020937	lr=0.296218
[1,0]<stderr>:INFO:root:Epoch[95] Batch[400] Loss[3.187]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[400]	rmse=0.020978	lr=0.296052
[1,0]<stderr>:INFO:root:Epoch[95] Batch[500] Loss[5.196]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[500]	rmse=0.021044	lr=0.295886
[1,0]<stderr>:INFO:root:Epoch[95] Batch[600] Loss[3.143]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[600]	rmse=0.021035	lr=0.295720
[1,0]<stderr>:INFO:root:Epoch[95] Batch[700] Loss[4.565]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[700]	rmse=0.021003	lr=0.295553
[1,0]<stderr>:INFO:root:Epoch[95] Batch[800] Loss[4.031]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[800]	rmse=0.021002	lr=0.295387
[1,0]<stderr>:INFO:root:Epoch[95] Batch[900] Loss[4.993]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[900]	rmse=0.020987	lr=0.295221
[1,0]<stderr>:INFO:root:Epoch[95] Batch[1000] Loss[4.458]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[1000]	rmse=0.020991	lr=0.295054
[1,0]<stderr>:INFO:root:Epoch[95] Batch[1100] Loss[3.731]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[1100]	rmse=0.021022	lr=0.294887
[1,0]<stderr>:INFO:root:Epoch[95] Batch[1200] Loss[2.723]
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[1200]	rmse=0.021026	lr=0.294721
[1,0]<stderr>:INFO:root:Epoch[95] Rank[0] Batch[1251]	Time cost=402.17	Train-metric=0.021035
[1,0]<stderr>:INFO:root:Epoch[95]	Speed: 3185.28 samples/sec
[1,0]<stderr>:INFO:root:Epoch[96] Batch[100] Loss[3.084]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[100]	rmse=0.021150	lr=0.294469
[1,0]<stderr>:INFO:root:Epoch[96] Batch[200] Loss[2.669]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[200]	rmse=0.021011	lr=0.294302
[1,0]<stderr>:INFO:root:Epoch[96] Batch[300] Loss[3.207]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[300]	rmse=0.021060	lr=0.294135
[1,0]<stderr>:INFO:root:Epoch[96] Batch[400] Loss[2.556]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[400]	rmse=0.021059	lr=0.293968
[1,0]<stderr>:INFO:root:Epoch[96] Batch[500] Loss[3.080]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[500]	rmse=0.021052	lr=0.293800
[1,0]<stderr>:INFO:root:Epoch[96] Batch[600] Loss[2.804]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[600]	rmse=0.021035	lr=0.293633
[1,0]<stderr>:INFO:root:Epoch[96] Batch[700] Loss[2.929]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[700]	rmse=0.021046	lr=0.293466
[1,0]<stderr>:INFO:root:Epoch[96] Batch[800] Loss[3.270]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[800]	rmse=0.021050	lr=0.293298
[1,0]<stderr>:INFO:root:Epoch[96] Batch[900] Loss[5.085]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[900]	rmse=0.021050	lr=0.293131
[1,0]<stderr>:INFO:root:Epoch[96] Batch[1000] Loss[4.671]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[1000]	rmse=0.021051	lr=0.292963
[1,0]<stderr>:INFO:root:Epoch[96] Batch[1100] Loss[2.905]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[1100]	rmse=0.021056	lr=0.292795
[1,0]<stderr>:INFO:root:Epoch[96] Batch[1200] Loss[2.945]
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[1200]	rmse=0.021064	lr=0.292627
[1,0]<stderr>:INFO:root:Epoch[96] Rank[0] Batch[1251]	Time cost=403.18	Train-metric=0.021070
[1,0]<stderr>:INFO:root:Epoch[96]	Speed: 3177.31 samples/sec
[1,0]<stderr>:INFO:root:Epoch[97] Batch[100] Loss[4.244]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[100]	rmse=0.020851	lr=0.292374
[1,0]<stderr>:INFO:root:Epoch[97] Batch[200] Loss[2.789]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[200]	rmse=0.020953	lr=0.292206
[1,0]<stderr>:INFO:root:Epoch[97] Batch[300] Loss[4.082]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[300]	rmse=0.020928	lr=0.292038
[1,0]<stderr>:INFO:root:Epoch[97] Batch[400] Loss[3.284]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[400]	rmse=0.020955	lr=0.291870
[1,0]<stderr>:INFO:root:Epoch[97] Batch[500] Loss[4.269]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[500]	rmse=0.020956	lr=0.291701
[1,0]<stderr>:INFO:root:Epoch[97] Batch[600] Loss[4.670]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[600]	rmse=0.020965	lr=0.291533
[1,0]<stderr>:INFO:root:Epoch[97] Batch[700] Loss[4.368]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[700]	rmse=0.021020	lr=0.291365
[1,0]<stderr>:INFO:root:Epoch[97] Batch[800] Loss[2.771]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[800]	rmse=0.021025	lr=0.291196
[1,0]<stderr>:INFO:root:Epoch[97] Batch[900] Loss[2.747]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[900]	rmse=0.021022	lr=0.291028
[1,0]<stderr>:INFO:root:Epoch[97] Batch[1000] Loss[4.692]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[1000]	rmse=0.021026	lr=0.290859
[1,0]<stderr>:INFO:root:Epoch[97] Batch[1100] Loss[3.475]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[1100]	rmse=0.021029	lr=0.290690
[1,0]<stderr>:INFO:root:Epoch[97] Batch[1200] Loss[2.836]
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[1200]	rmse=0.021025	lr=0.290521
[1,0]<stderr>:INFO:root:Epoch[97] Rank[0] Batch[1251]	Time cost=399.52	Train-metric=0.021020
[1,0]<stderr>:INFO:root:Epoch[97]	Speed: 3206.39 samples/sec
[1,0]<stderr>:INFO:root:Epoch[98] Batch[100] Loss[4.449]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[100]	rmse=0.020840	lr=0.290266
[1,0]<stderr>:INFO:root:Epoch[98] Batch[200] Loss[3.144]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[200]	rmse=0.020870	lr=0.290097
[1,0]<stderr>:INFO:root:Epoch[98] Batch[300] Loss[4.637]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[300]	rmse=0.020904	lr=0.289928
[1,0]<stderr>:INFO:root:Epoch[98] Batch[400] Loss[2.930]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[400]	rmse=0.020956	lr=0.289759
[1,0]<stderr>:INFO:root:Epoch[98] Batch[500] Loss[3.002]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[500]	rmse=0.020987	lr=0.289590
[1,0]<stderr>:INFO:root:Epoch[98] Batch[600] Loss[4.981]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[600]	rmse=0.020920	lr=0.289420
[1,0]<stderr>:INFO:root:Epoch[98] Batch[700] Loss[2.995]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[700]	rmse=0.020944	lr=0.289251
[1,0]<stderr>:INFO:root:Epoch[98] Batch[800] Loss[2.990]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[800]	rmse=0.020951	lr=0.289081
[1,0]<stderr>:INFO:root:Epoch[98] Batch[900] Loss[3.081]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[900]	rmse=0.020945	lr=0.288912
[1,0]<stderr>:INFO:root:Epoch[98] Batch[1000] Loss[3.144]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[1000]	rmse=0.020959	lr=0.288742
[1,0]<stderr>:INFO:root:Epoch[98] Batch[1100] Loss[4.498]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[1100]	rmse=0.020943	lr=0.288572
[1,0]<stderr>:INFO:root:Epoch[98] Batch[1200] Loss[3.273]
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[1200]	rmse=0.020949	lr=0.288402
[1,0]<stderr>:INFO:root:Epoch[98] Rank[0] Batch[1251]	Time cost=399.16	Train-metric=0.020960
[1,0]<stderr>:INFO:root:Epoch[98]	Speed: 3209.32 samples/sec
[1,0]<stderr>:INFO:root:Epoch[99] Batch[100] Loss[3.178]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[100]	rmse=0.021020	lr=0.288146
[1,0]<stderr>:INFO:root:Epoch[99] Batch[200] Loss[3.648]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[200]	rmse=0.020929	lr=0.287976
[1,0]<stderr>:INFO:root:Epoch[99] Batch[300] Loss[3.922]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[300]	rmse=0.020977	lr=0.287806
[1,0]<stderr>:INFO:root:Epoch[99] Batch[400] Loss[3.536]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[400]	rmse=0.020973	lr=0.287635
[1,0]<stderr>:INFO:root:Epoch[99] Batch[500] Loss[3.056]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[500]	rmse=0.020948	lr=0.287465
[1,0]<stderr>:INFO:root:Epoch[99] Batch[600] Loss[3.231]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[600]	rmse=0.020975	lr=0.287295
[1,0]<stderr>:INFO:root:Epoch[99] Batch[700] Loss[3.732]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[700]	rmse=0.021010	lr=0.287124
[1,0]<stderr>:INFO:root:Epoch[99] Batch[800] Loss[3.444]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[800]	rmse=0.021002	lr=0.286954
[1,0]<stderr>:INFO:root:Epoch[99] Batch[900] Loss[2.791]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[900]	rmse=0.020971	lr=0.286783
[1,0]<stderr>:INFO:root:Epoch[99] Batch[1000] Loss[2.778]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[1000]	rmse=0.020957	lr=0.286613
[1,0]<stderr>:INFO:root:Epoch[99] Batch[1100] Loss[2.823]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[1100]	rmse=0.020964	lr=0.286442
[1,0]<stderr>:INFO:root:Epoch[99] Batch[1200] Loss[2.649]
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[1200]	rmse=0.020971	lr=0.286271
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0] Batch[1251]	Time cost=399.53	Train-metric=0.020990
[1,0]<stderr>:INFO:root:Epoch[99]	Speed: 3206.36 samples/sec
[1,0]<stderr>:INFO:root:Epoch[99] Rank[0]	Validation-accuracy=0.596420	Validation-top_k_accuracy_5=0.831900
[1,0]<stderr>:INFO:root:Epoch[100] Batch[100] Loss[2.886]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[100]	rmse=0.020695	lr=0.286013
[1,0]<stderr>:INFO:root:Epoch[100] Batch[200] Loss[5.084]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[200]	rmse=0.020820	lr=0.285842
[1,0]<stderr>:INFO:root:Epoch[100] Batch[300] Loss[2.883]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[300]	rmse=0.020855	lr=0.285671
[1,0]<stderr>:INFO:root:Epoch[100] Batch[400] Loss[4.013]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[400]	rmse=0.020888	lr=0.285500
[1,0]<stderr>:INFO:root:Epoch[100] Batch[500] Loss[3.710]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[500]	rmse=0.020904	lr=0.285329
[1,0]<stderr>:INFO:root:Epoch[100] Batch[600] Loss[5.536]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[600]	rmse=0.020917	lr=0.285157
[1,0]<stderr>:INFO:root:Epoch[100] Batch[700] Loss[2.918]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[700]	rmse=0.020951	lr=0.284986
[1,0]<stderr>:INFO:root:Epoch[100] Batch[800] Loss[2.845]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[800]	rmse=0.020945	lr=0.284814
[1,0]<stderr>:INFO:root:Epoch[100] Batch[900] Loss[2.910]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[900]	rmse=0.020947	lr=0.284643
[1,0]<stderr>:INFO:root:Epoch[100] Batch[1000] Loss[3.176]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[1000]	rmse=0.020962	lr=0.284471
[1,0]<stderr>:INFO:root:Epoch[100] Batch[1100] Loss[3.299]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[1100]	rmse=0.020975	lr=0.284300
[1,0]<stderr>:INFO:root:Epoch[100] Batch[1200] Loss[4.983]
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[1200]	rmse=0.020991	lr=0.284128
[1,0]<stderr>:INFO:root:Epoch[100] Rank[0] Batch[1251]	Time cost=398.57	Train-metric=0.021000
[1,0]<stderr>:INFO:root:Epoch[100]	Speed: 3214.05 samples/sec
[1,0]<stderr>:INFO:root:Epoch[101] Batch[100] Loss[2.880]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[100]	rmse=0.021006	lr=0.283868
[1,0]<stderr>:INFO:root:Epoch[101] Batch[200] Loss[4.172]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[200]	rmse=0.020976	lr=0.283696
[1,0]<stderr>:INFO:root:Epoch[101] Batch[300] Loss[3.504]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[300]	rmse=0.020961	lr=0.283524
[1,0]<stderr>:INFO:root:Epoch[101] Batch[400] Loss[3.017]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[400]	rmse=0.020953	lr=0.283352
[1,0]<stderr>:INFO:root:Epoch[101] Batch[500] Loss[2.911]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[500]	rmse=0.020976	lr=0.283180
[1,0]<stderr>:INFO:root:Epoch[101] Batch[600] Loss[2.971]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[600]	rmse=0.020963	lr=0.283008
[1,0]<stderr>:INFO:root:Epoch[101] Batch[700] Loss[2.657]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[700]	rmse=0.020997	lr=0.282835
[1,0]<stderr>:INFO:root:Epoch[101] Batch[800] Loss[3.145]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[800]	rmse=0.020976	lr=0.282663
[1,0]<stderr>:INFO:root:Epoch[101] Batch[900] Loss[5.344]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[900]	rmse=0.020969	lr=0.282490
[1,0]<stderr>:INFO:root:Epoch[101] Batch[1000] Loss[3.438]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[1000]	rmse=0.020994	lr=0.282318
[1,0]<stderr>:INFO:root:Epoch[101] Batch[1100] Loss[3.078]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[1100]	rmse=0.021004	lr=0.282145
[1,0]<stderr>:INFO:root:Epoch[101] Batch[1200] Loss[2.899]
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[1200]	rmse=0.020997	lr=0.281973
[1,0]<stderr>:INFO:root:Epoch[101] Rank[0] Batch[1251]	Time cost=399.42	Train-metric=0.021008
[1,0]<stderr>:INFO:root:Epoch[101]	Speed: 3207.21 samples/sec
[1,0]<stderr>:INFO:root:Epoch[102] Batch[100] Loss[2.841]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[100]	rmse=0.020832	lr=0.281712
[1,0]<stderr>:INFO:root:Epoch[102] Batch[200] Loss[2.723]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[200]	rmse=0.020812	lr=0.281539
[1,0]<stderr>:INFO:root:Epoch[102] Batch[300] Loss[2.819]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[300]	rmse=0.020830	lr=0.281366
[1,0]<stderr>:INFO:root:Epoch[102] Batch[400] Loss[3.402]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[400]	rmse=0.020814	lr=0.281193
[1,0]<stderr>:INFO:root:Epoch[102] Batch[500] Loss[5.270]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[500]	rmse=0.020861	lr=0.281020
[1,0]<stderr>:INFO:root:Epoch[102] Batch[600] Loss[5.123]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[600]	rmse=0.020886	lr=0.280846
[1,0]<stderr>:INFO:root:Epoch[102] Batch[700] Loss[3.029]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[700]	rmse=0.020922	lr=0.280673
[1,0]<stderr>:INFO:root:Epoch[102] Batch[800] Loss[5.052]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[800]	rmse=0.020913	lr=0.280500
[1,0]<stderr>:INFO:root:Epoch[102] Batch[900] Loss[3.040]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[900]	rmse=0.020919	lr=0.280326
[1,0]<stderr>:INFO:root:Epoch[102] Batch[1000] Loss[4.214]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[1000]	rmse=0.020926	lr=0.280153
[1,0]<stderr>:INFO:root:Epoch[102] Batch[1100] Loss[3.164]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[1100]	rmse=0.020925	lr=0.279979
[1,0]<stderr>:INFO:root:Epoch[102] Batch[1200] Loss[3.626]
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[1200]	rmse=0.020941	lr=0.279806
[1,0]<stderr>:INFO:root:Epoch[102] Rank[0] Batch[1251]	Time cost=399.27	Train-metric=0.020940
[1,0]<stderr>:INFO:root:Epoch[102]	Speed: 3208.41 samples/sec
[1,0]<stderr>:INFO:root:Epoch[103] Batch[100] Loss[3.356]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[100]	rmse=0.020750	lr=0.279544
[1,0]<stderr>:INFO:root:Epoch[103] Batch[200] Loss[3.771]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[200]	rmse=0.020871	lr=0.279370
[1,0]<stderr>:INFO:root:Epoch[103] Batch[300] Loss[5.503]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[300]	rmse=0.020862	lr=0.279196
[1,0]<stderr>:INFO:root:Epoch[103] Batch[400] Loss[2.885]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[400]	rmse=0.020905	lr=0.279022
[1,0]<stderr>:INFO:root:Epoch[103] Batch[500] Loss[3.020]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[500]	rmse=0.020967	lr=0.278848
[1,0]<stderr>:INFO:root:Epoch[103] Batch[600] Loss[2.981]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[600]	rmse=0.020948	lr=0.278674
[1,0]<stderr>:INFO:root:Epoch[103] Batch[700] Loss[4.708]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[700]	rmse=0.020955	lr=0.278500
[1,0]<stderr>:INFO:root:Epoch[103] Batch[800] Loss[3.072]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[800]	rmse=0.020963	lr=0.278326
[1,0]<stderr>:INFO:root:Epoch[103] Batch[900] Loss[2.666]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[900]	rmse=0.020981	lr=0.278151
[1,0]<stderr>:INFO:root:Epoch[103] Batch[1000] Loss[3.237]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[1000]	rmse=0.020968	lr=0.277977
[1,0]<stderr>:INFO:root:Epoch[103] Batch[1100] Loss[2.737]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[1100]	rmse=0.020961	lr=0.277802
[1,0]<stderr>:INFO:root:Epoch[103] Batch[1200] Loss[5.054]
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[1200]	rmse=0.020961	lr=0.277628
[1,0]<stderr>:INFO:root:Epoch[103] Rank[0] Batch[1251]	Time cost=399.03	Train-metric=0.020969
[1,0]<stderr>:INFO:root:Epoch[103]	Speed: 3210.32 samples/sec
[1,0]<stderr>:INFO:root:Epoch[104] Batch[100] Loss[2.758]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[100]	rmse=0.020884	lr=0.277364
[1,0]<stderr>:INFO:root:Epoch[104] Batch[200] Loss[4.810]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[200]	rmse=0.020908	lr=0.277190
[1,0]<stderr>:INFO:root:Epoch[104] Batch[300] Loss[2.968]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[300]	rmse=0.020816	lr=0.277015
[1,0]<stderr>:INFO:root:Epoch[104] Batch[400] Loss[4.690]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[400]	rmse=0.020827	lr=0.276840
[1,0]<stderr>:INFO:root:Epoch[104] Batch[500] Loss[3.161]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[500]	rmse=0.020838	lr=0.276665
[1,0]<stderr>:INFO:root:Epoch[104] Batch[600] Loss[2.757]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[600]	rmse=0.020863	lr=0.276490
[1,0]<stderr>:INFO:root:Epoch[104] Batch[700] Loss[2.974]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[700]	rmse=0.020912	lr=0.276315
[1,0]<stderr>:INFO:root:Epoch[104] Batch[800] Loss[3.143]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[800]	rmse=0.020935	lr=0.276140
[1,0]<stderr>:INFO:root:Epoch[104] Batch[900] Loss[4.622]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[900]	rmse=0.020952	lr=0.275965
[1,0]<stderr>:INFO:root:Epoch[104] Batch[1000] Loss[2.995]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[1000]	rmse=0.020966	lr=0.275790
[1,0]<stderr>:INFO:root:Epoch[104] Batch[1100] Loss[3.637]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[1100]	rmse=0.020974	lr=0.275615
[1,0]<stderr>:INFO:root:Epoch[104] Batch[1200] Loss[3.114]
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[1200]	rmse=0.020981	lr=0.275439
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0] Batch[1251]	Time cost=399.46	Train-metric=0.020978
[1,0]<stderr>:INFO:root:Epoch[104]	Speed: 3206.88 samples/sec
[1,0]<stderr>:INFO:root:Epoch[104] Rank[0]	Validation-accuracy=0.607660	Validation-top_k_accuracy_5=0.837200
[1,0]<stderr>:INFO:root:Epoch[105] Batch[100] Loss[2.737]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[100]	rmse=0.020970	lr=0.275174
[1,0]<stderr>:INFO:root:Epoch[105] Batch[200] Loss[5.301]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[200]	rmse=0.020889	lr=0.274999
[1,0]<stderr>:INFO:root:Epoch[105] Batch[300] Loss[2.895]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[300]	rmse=0.020918	lr=0.274823
[1,0]<stderr>:INFO:root:Epoch[105] Batch[400] Loss[2.651]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[400]	rmse=0.020912	lr=0.274648
[1,0]<stderr>:INFO:root:Epoch[105] Batch[500] Loss[2.999]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[500]	rmse=0.020928	lr=0.274472
[1,0]<stderr>:INFO:root:Epoch[105] Batch[600] Loss[2.941]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[600]	rmse=0.020917	lr=0.274296
[1,0]<stderr>:INFO:root:Epoch[105] Batch[700] Loss[5.005]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[700]	rmse=0.020938	lr=0.274120
[1,0]<stderr>:INFO:root:Epoch[105] Batch[800] Loss[3.179]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[800]	rmse=0.020942	lr=0.273944
[1,0]<stderr>:INFO:root:Epoch[105] Batch[900] Loss[2.827]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[900]	rmse=0.020963	lr=0.273768
[1,0]<stderr>:INFO:root:Epoch[105] Batch[1000] Loss[3.255]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[1000]	rmse=0.020948	lr=0.273592
[1,0]<stderr>:INFO:root:Epoch[105] Batch[1100] Loss[4.536]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[1100]	rmse=0.020929	lr=0.273416
[1,0]<stderr>:INFO:root:Epoch[105] Batch[1200] Loss[4.820]
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[1200]	rmse=0.020910	lr=0.273240
[1,0]<stderr>:INFO:root:Epoch[105] Rank[0] Batch[1251]	Time cost=398.34	Train-metric=0.020915
[1,0]<stderr>:INFO:root:Epoch[105]	Speed: 3215.90 samples/sec
[1,0]<stderr>:INFO:root:Epoch[106] Batch[100] Loss[2.868]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[100]	rmse=0.020839	lr=0.272974
[1,0]<stderr>:INFO:root:Epoch[106] Batch[200] Loss[3.422]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[200]	rmse=0.020855	lr=0.272797
[1,0]<stderr>:INFO:root:Epoch[106] Batch[300] Loss[3.211]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[300]	rmse=0.020928	lr=0.272621
[1,0]<stderr>:INFO:root:Epoch[106] Batch[400] Loss[2.939]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[400]	rmse=0.020935	lr=0.272444
[1,0]<stderr>:INFO:root:Epoch[106] Batch[500] Loss[3.621]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[500]	rmse=0.020931	lr=0.272268
[1,0]<stderr>:INFO:root:Epoch[106] Batch[600] Loss[2.964]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[600]	rmse=0.020929	lr=0.272091
[1,0]<stderr>:INFO:root:Epoch[106] Batch[700] Loss[3.489]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[700]	rmse=0.020912	lr=0.271915
[1,0]<stderr>:INFO:root:Epoch[106] Batch[800] Loss[3.013]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[800]	rmse=0.020917	lr=0.271738
[1,0]<stderr>:INFO:root:Epoch[106] Batch[900] Loss[3.290]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[900]	rmse=0.020888	lr=0.271561
[1,0]<stderr>:INFO:root:Epoch[106] Batch[1000] Loss[3.129]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[1000]	rmse=0.020896	lr=0.271384
[1,0]<stderr>:INFO:root:Epoch[106] Batch[1100] Loss[2.896]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[1100]	rmse=0.020886	lr=0.271207
[1,0]<stderr>:INFO:root:Epoch[106] Batch[1200] Loss[4.302]
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[1200]	rmse=0.020877	lr=0.271030
[1,0]<stderr>:INFO:root:Epoch[106] Rank[0] Batch[1251]	Time cost=399.39	Train-metric=0.020874
[1,0]<stderr>:INFO:root:Epoch[106]	Speed: 3207.42 samples/sec
[1,0]<stderr>:INFO:root:Epoch[107] Batch[100] Loss[4.217]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[100]	rmse=0.020913	lr=0.270763
[1,0]<stderr>:INFO:root:Epoch[107] Batch[200] Loss[4.126]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[200]	rmse=0.020893	lr=0.270586
[1,0]<stderr>:INFO:root:Epoch[107] Batch[300] Loss[4.230]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[300]	rmse=0.020946	lr=0.270408
[1,0]<stderr>:INFO:root:Epoch[107] Batch[400] Loss[2.770]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[400]	rmse=0.020945	lr=0.270231
[1,0]<stderr>:INFO:root:Epoch[107] Batch[500] Loss[3.030]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[500]	rmse=0.020930	lr=0.270054
[1,0]<stderr>:INFO:root:Epoch[107] Batch[600] Loss[2.834]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[600]	rmse=0.020917	lr=0.269876
[1,0]<stderr>:INFO:root:Epoch[107] Batch[700] Loss[2.855]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[700]	rmse=0.020945	lr=0.269699
[1,0]<stderr>:INFO:root:Epoch[107] Batch[800] Loss[2.800]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[800]	rmse=0.020941	lr=0.269521
[1,0]<stderr>:INFO:root:Epoch[107] Batch[900] Loss[3.170]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[900]	rmse=0.020949	lr=0.269344
[1,0]<stderr>:INFO:root:Epoch[107] Batch[1000] Loss[3.057]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[1000]	rmse=0.020940	lr=0.269166
[1,0]<stderr>:INFO:root:Epoch[107] Batch[1100] Loss[4.697]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[1100]	rmse=0.020946	lr=0.268988
[1,0]<stderr>:INFO:root:Epoch[107] Batch[1200] Loss[3.001]
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[1200]	rmse=0.020948	lr=0.268811
[1,0]<stderr>:INFO:root:Epoch[107] Rank[0] Batch[1251]	Time cost=399.24	Train-metric=0.020949
[1,0]<stderr>:INFO:root:Epoch[107]	Speed: 3208.64 samples/sec
[1,0]<stderr>:INFO:root:Epoch[108] Batch[100] Loss[3.020]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[100]	rmse=0.020682	lr=0.268542
[1,0]<stderr>:INFO:root:Epoch[108] Batch[200] Loss[3.139]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[200]	rmse=0.020704	lr=0.268364
[1,0]<stderr>:INFO:root:Epoch[108] Batch[300] Loss[3.942]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[300]	rmse=0.020676	lr=0.268186
[1,0]<stderr>:INFO:root:Epoch[108] Batch[400] Loss[5.282]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[400]	rmse=0.020768	lr=0.268008
[1,0]<stderr>:INFO:root:Epoch[108] Batch[500] Loss[2.933]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[500]	rmse=0.020809	lr=0.267830
[1,0]<stderr>:INFO:root:Epoch[108] Batch[600] Loss[2.788]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[600]	rmse=0.020781	lr=0.267652
[1,0]<stderr>:INFO:root:Epoch[108] Batch[700] Loss[4.344]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[700]	rmse=0.020782	lr=0.267473
[1,0]<stderr>:INFO:root:Epoch[108] Batch[800] Loss[2.943]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[800]	rmse=0.020788	lr=0.267295
[1,0]<stderr>:INFO:root:Epoch[108] Batch[900] Loss[2.928]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[900]	rmse=0.020801	lr=0.267117
[1,0]<stderr>:INFO:root:Epoch[108] Batch[1000] Loss[4.360]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[1000]	rmse=0.020817	lr=0.266938
[1,0]<stderr>:INFO:root:Epoch[108] Batch[1100] Loss[2.971]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[1100]	rmse=0.020826	lr=0.266760
[1,0]<stderr>:INFO:root:Epoch[108] Batch[1200] Loss[3.347]
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[1200]	rmse=0.020817	lr=0.266581
[1,0]<stderr>:INFO:root:Epoch[108] Rank[0] Batch[1251]	Time cost=399.25	Train-metric=0.020826
[1,0]<stderr>:INFO:root:Epoch[108]	Speed: 3208.59 samples/sec
[1,0]<stderr>:INFO:root:Epoch[109] Batch[100] Loss[2.965]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[100]	rmse=0.020751	lr=0.266312
[1,0]<stderr>:INFO:root:Epoch[109] Batch[200] Loss[3.710]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[200]	rmse=0.020818	lr=0.266133
[1,0]<stderr>:INFO:root:Epoch[109] Batch[300] Loss[5.219]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[300]	rmse=0.020828	lr=0.265954
[1,0]<stderr>:INFO:root:Epoch[109] Batch[400] Loss[3.075]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[400]	rmse=0.020847	lr=0.265775
[1,0]<stderr>:INFO:root:Epoch[109] Batch[500] Loss[2.708]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[500]	rmse=0.020846	lr=0.265596
[1,0]<stderr>:INFO:root:Epoch[109] Batch[600] Loss[2.738]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[600]	rmse=0.020854	lr=0.265418
[1,0]<stderr>:INFO:root:Epoch[109] Batch[700] Loss[3.198]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[700]	rmse=0.020856	lr=0.265239
[1,0]<stderr>:INFO:root:Epoch[109] Batch[800] Loss[2.911]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[800]	rmse=0.020876	lr=0.265060
[1,0]<stderr>:INFO:root:Epoch[109] Batch[900] Loss[3.237]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[900]	rmse=0.020858	lr=0.264880
[1,0]<stderr>:INFO:root:Epoch[109] Batch[1000] Loss[3.660]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[1000]	rmse=0.020877	lr=0.264701
[1,0]<stderr>:INFO:root:Epoch[109] Batch[1100] Loss[2.752]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[1100]	rmse=0.020892	lr=0.264522
[1,0]<stderr>:INFO:root:Epoch[109] Batch[1200] Loss[2.625]
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[1200]	rmse=0.020899	lr=0.264343
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0] Batch[1251]	Time cost=399.56	Train-metric=0.020897
[1,0]<stderr>:INFO:root:Epoch[109]	Speed: 3206.05 samples/sec
[1,0]<stderr>:INFO:root:Epoch[109] Rank[0]	Validation-accuracy=0.618080	Validation-top_k_accuracy_5=0.847000
[1,0]<stderr>:INFO:root:Epoch[110] Batch[100] Loss[2.565]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[100]	rmse=0.020692	lr=0.264072
[1,0]<stderr>:INFO:root:Epoch[110] Batch[200] Loss[3.428]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[200]	rmse=0.020770	lr=0.263892
[1,0]<stderr>:INFO:root:Epoch[110] Batch[300] Loss[4.750]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[300]	rmse=0.020745	lr=0.263713
[1,0]<stderr>:INFO:root:Epoch[110] Batch[400] Loss[2.616]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[400]	rmse=0.020785	lr=0.263533
[1,0]<stderr>:INFO:root:Epoch[110] Batch[500] Loss[3.704]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[500]	rmse=0.020741	lr=0.263354
[1,0]<stderr>:INFO:root:Epoch[110] Batch[600] Loss[4.241]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[600]	rmse=0.020769	lr=0.263174
[1,0]<stderr>:INFO:root:Epoch[110] Batch[700] Loss[2.704]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[700]	rmse=0.020789	lr=0.262995
[1,0]<stderr>:INFO:root:Epoch[110] Batch[800] Loss[2.825]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[800]	rmse=0.020805	lr=0.262815
[1,0]<stderr>:INFO:root:Epoch[110] Batch[900] Loss[3.651]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[900]	rmse=0.020823	lr=0.262635
[1,0]<stderr>:INFO:root:Epoch[110] Batch[1000] Loss[4.786]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[1000]	rmse=0.020840	lr=0.262455
[1,0]<stderr>:INFO:root:Epoch[110] Batch[1100] Loss[3.201]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[1100]	rmse=0.020833	lr=0.262275
[1,0]<stderr>:INFO:root:Epoch[110] Batch[1200] Loss[2.827]
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[1200]	rmse=0.020825	lr=0.262095
[1,0]<stderr>:INFO:root:Epoch[110] Rank[0] Batch[1251]	Time cost=399.61	Train-metric=0.020833
[1,0]<stderr>:INFO:root:Epoch[110]	Speed: 3205.71 samples/sec
[1,0]<stderr>:INFO:root:Epoch[111] Batch[100] Loss[4.803]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[100]	rmse=0.021003	lr=0.261823
[1,0]<stderr>:INFO:root:Epoch[111] Batch[200] Loss[5.304]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[200]	rmse=0.020892	lr=0.261643
[1,0]<stderr>:INFO:root:Epoch[111] Batch[300] Loss[2.768]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[300]	rmse=0.020908	lr=0.261463
[1,0]<stderr>:INFO:root:Epoch[111] Batch[400] Loss[2.994]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[400]	rmse=0.020852	lr=0.261283
[1,0]<stderr>:INFO:root:Epoch[111] Batch[500] Loss[2.842]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[500]	rmse=0.020849	lr=0.261102
[1,0]<stderr>:INFO:root:Epoch[111] Batch[600] Loss[2.706]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[600]	rmse=0.020830	lr=0.260922
[1,0]<stderr>:INFO:root:Epoch[111] Batch[700] Loss[3.082]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[700]	rmse=0.020860	lr=0.260742
[1,0]<stderr>:INFO:root:Epoch[111] Batch[800] Loss[2.899]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[800]	rmse=0.020853	lr=0.260561
[1,0]<stderr>:INFO:root:Epoch[111] Batch[900] Loss[2.955]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[900]	rmse=0.020873	lr=0.260381
[1,0]<stderr>:INFO:root:Epoch[111] Batch[1000] Loss[2.910]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[1000]	rmse=0.020897	lr=0.260200
[1,0]<stderr>:INFO:root:Epoch[111] Batch[1100] Loss[3.670]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[1100]	rmse=0.020904	lr=0.260019
[1,0]<stderr>:INFO:root:Epoch[111] Batch[1200] Loss[4.618]
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[1200]	rmse=0.020891	lr=0.259839
[1,0]<stderr>:INFO:root:Epoch[111] Rank[0] Batch[1251]	Time cost=398.96	Train-metric=0.020887
[1,0]<stderr>:INFO:root:Epoch[111]	Speed: 3210.89 samples/sec
[1,0]<stderr>:INFO:root:Epoch[112] Batch[100] Loss[3.929]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[100]	rmse=0.020796	lr=0.259566
[1,0]<stderr>:INFO:root:Epoch[112] Batch[200] Loss[3.736]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[200]	rmse=0.020701	lr=0.259385
[1,0]<stderr>:INFO:root:Epoch[112] Batch[300] Loss[3.272]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[300]	rmse=0.020721	lr=0.259204
[1,0]<stderr>:INFO:root:Epoch[112] Batch[400] Loss[4.531]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[400]	rmse=0.020772	lr=0.259023
[1,0]<stderr>:INFO:root:Epoch[112] Batch[500] Loss[2.746]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[500]	rmse=0.020800	lr=0.258842
[1,0]<stderr>:INFO:root:Epoch[112] Batch[600] Loss[3.205]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[600]	rmse=0.020831	lr=0.258661
[1,0]<stderr>:INFO:root:Epoch[112] Batch[700] Loss[4.996]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[700]	rmse=0.020790	lr=0.258480
[1,0]<stderr>:INFO:root:Epoch[112] Batch[800] Loss[2.911]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[800]	rmse=0.020799	lr=0.258299
[1,0]<stderr>:INFO:root:Epoch[112] Batch[900] Loss[3.100]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[900]	rmse=0.020811	lr=0.258118
[1,0]<stderr>:INFO:root:Epoch[112] Batch[1000] Loss[3.040]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[1000]	rmse=0.020808	lr=0.257937
[1,0]<stderr>:INFO:root:Epoch[112] Batch[1100] Loss[2.850]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[1100]	rmse=0.020807	lr=0.257755
[1,0]<stderr>:INFO:root:Epoch[112] Batch[1200] Loss[4.056]
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[1200]	rmse=0.020809	lr=0.257574
[1,0]<stderr>:INFO:root:Epoch[112] Rank[0] Batch[1251]	Time cost=398.79	Train-metric=0.020810
[1,0]<stderr>:INFO:root:Epoch[112]	Speed: 3212.24 samples/sec
[1,0]<stderr>:INFO:root:Epoch[113] Batch[100] Loss[3.305]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[100]	rmse=0.020723	lr=0.257300
[1,0]<stderr>:INFO:root:Epoch[113] Batch[200] Loss[2.815]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[200]	rmse=0.020729	lr=0.257119
[1,0]<stderr>:INFO:root:Epoch[113] Batch[300] Loss[2.791]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[300]	rmse=0.020782	lr=0.256937
[1,0]<stderr>:INFO:root:Epoch[113] Batch[400] Loss[2.757]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[400]	rmse=0.020805	lr=0.256756
[1,0]<stderr>:INFO:root:Epoch[113] Batch[500] Loss[3.653]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[500]	rmse=0.020835	lr=0.256574
[1,0]<stderr>:INFO:root:Epoch[113] Batch[600] Loss[3.735]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[600]	rmse=0.020843	lr=0.256392
[1,0]<stderr>:INFO:root:Epoch[113] Batch[700] Loss[2.936]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[700]	rmse=0.020849	lr=0.256211
[1,0]<stderr>:INFO:root:Epoch[113] Batch[800] Loss[2.698]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[800]	rmse=0.020841	lr=0.256029
[1,0]<stderr>:INFO:root:Epoch[113] Batch[900] Loss[4.029]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[900]	rmse=0.020828	lr=0.255847
[1,0]<stderr>:INFO:root:Epoch[113] Batch[1000] Loss[4.879]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[1000]	rmse=0.020814	lr=0.255665
[1,0]<stderr>:INFO:root:Epoch[113] Batch[1100] Loss[2.891]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[1100]	rmse=0.020821	lr=0.255483
[1,0]<stderr>:INFO:root:Epoch[113] Batch[1200] Loss[2.945]
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[1200]	rmse=0.020818	lr=0.255301
[1,0]<stderr>:INFO:root:Epoch[113] Rank[0] Batch[1251]	Time cost=398.74	Train-metric=0.020808
[1,0]<stderr>:INFO:root:Epoch[113]	Speed: 3212.70 samples/sec
[1,0]<stderr>:INFO:root:Epoch[114] Batch[100] Loss[2.558]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[100]	rmse=0.020579	lr=0.255026
[1,0]<stderr>:INFO:root:Epoch[114] Batch[200] Loss[2.705]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[200]	rmse=0.020711	lr=0.254844
[1,0]<stderr>:INFO:root:Epoch[114] Batch[300] Loss[3.305]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[300]	rmse=0.020686	lr=0.254662
[1,0]<stderr>:INFO:root:Epoch[114] Batch[400] Loss[4.736]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[400]	rmse=0.020718	lr=0.254480
[1,0]<stderr>:INFO:root:Epoch[114] Batch[500] Loss[3.198]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[500]	rmse=0.020705	lr=0.254298
[1,0]<stderr>:INFO:root:Epoch[114] Batch[600] Loss[5.161]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[600]	rmse=0.020684	lr=0.254115
[1,0]<stderr>:INFO:root:Epoch[114] Batch[700] Loss[4.087]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[700]	rmse=0.020717	lr=0.253933
[1,0]<stderr>:INFO:root:Epoch[114] Batch[800] Loss[5.397]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[800]	rmse=0.020736	lr=0.253751
[1,0]<stderr>:INFO:root:Epoch[114] Batch[900] Loss[2.981]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[900]	rmse=0.020737	lr=0.253568
[1,0]<stderr>:INFO:root:Epoch[114] Batch[1000] Loss[2.868]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[1000]	rmse=0.020744	lr=0.253386
[1,0]<stderr>:INFO:root:Epoch[114] Batch[1100] Loss[4.421]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[1100]	rmse=0.020748	lr=0.253203
[1,0]<stderr>:INFO:root:Epoch[114] Batch[1200] Loss[2.962]
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[1200]	rmse=0.020745	lr=0.253021
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0] Batch[1251]	Time cost=399.05	Train-metric=0.020742
[1,0]<stderr>:INFO:root:Epoch[114]	Speed: 3210.20 samples/sec
[1,0]<stderr>:INFO:root:Epoch[114] Rank[0]	Validation-accuracy=0.615780	Validation-top_k_accuracy_5=0.845040
[1,0]<stderr>:INFO:root:Epoch[115] Batch[100] Loss[2.896]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[100]	rmse=0.020637	lr=0.252745
[1,0]<stderr>:INFO:root:Epoch[115] Batch[200] Loss[2.534]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[200]	rmse=0.020676	lr=0.252562
[1,0]<stderr>:INFO:root:Epoch[115] Batch[300] Loss[2.667]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[300]	rmse=0.020775	lr=0.252379
[1,0]<stderr>:INFO:root:Epoch[115] Batch[400] Loss[3.105]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[400]	rmse=0.020784	lr=0.252197
[1,0]<stderr>:INFO:root:Epoch[115] Batch[500] Loss[3.191]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[500]	rmse=0.020764	lr=0.252014
[1,0]<stderr>:INFO:root:Epoch[115] Batch[600] Loss[2.726]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[600]	rmse=0.020791	lr=0.251831
[1,0]<stderr>:INFO:root:Epoch[115] Batch[700] Loss[4.576]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[700]	rmse=0.020790	lr=0.251648
[1,0]<stderr>:INFO:root:Epoch[115] Batch[800] Loss[2.931]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[800]	rmse=0.020797	lr=0.251465
[1,0]<stderr>:INFO:root:Epoch[115] Batch[900] Loss[3.381]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[900]	rmse=0.020822	lr=0.251282
[1,0]<stderr>:INFO:root:Epoch[115] Batch[1000] Loss[2.839]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[1000]	rmse=0.020817	lr=0.251099
[1,0]<stderr>:INFO:root:Epoch[115] Batch[1100] Loss[4.743]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[1100]	rmse=0.020832	lr=0.250916
[1,0]<stderr>:INFO:root:Epoch[115] Batch[1200] Loss[3.049]
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[1200]	rmse=0.020836	lr=0.250733
[1,0]<stderr>:INFO:root:Epoch[115] Rank[0] Batch[1251]	Time cost=398.37	Train-metric=0.020836
[1,0]<stderr>:INFO:root:Epoch[115]	Speed: 3215.65 samples/sec
[1,0]<stderr>:INFO:root:Epoch[116] Batch[100] Loss[2.728]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[100]	rmse=0.020723	lr=0.250456
[1,0]<stderr>:INFO:root:Epoch[116] Batch[200] Loss[4.684]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[200]	rmse=0.020723	lr=0.250273
[1,0]<stderr>:INFO:root:Epoch[116] Batch[300] Loss[5.020]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[300]	rmse=0.020709	lr=0.250089
[1,0]<stderr>:INFO:root:Epoch[116] Batch[400] Loss[3.801]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[400]	rmse=0.020749	lr=0.249906
[1,0]<stderr>:INFO:root:Epoch[116] Batch[500] Loss[2.893]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[500]	rmse=0.020762	lr=0.249723
[1,0]<stderr>:INFO:root:Epoch[116] Batch[600] Loss[2.690]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[600]	rmse=0.020769	lr=0.249539
[1,0]<stderr>:INFO:root:Epoch[116] Batch[700] Loss[2.774]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[700]	rmse=0.020755	lr=0.249356
[1,0]<stderr>:INFO:root:Epoch[116] Batch[800] Loss[4.044]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[800]	rmse=0.020756	lr=0.249172
[1,0]<stderr>:INFO:root:Epoch[116] Batch[900] Loss[2.756]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[900]	rmse=0.020770	lr=0.248988
[1,0]<stderr>:INFO:root:Epoch[116] Batch[1000] Loss[4.817]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[1000]	rmse=0.020770	lr=0.248805
[1,0]<stderr>:INFO:root:Epoch[116] Batch[1100] Loss[2.858]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[1100]	rmse=0.020798	lr=0.248621
[1,0]<stderr>:INFO:root:Epoch[116] Batch[1200] Loss[5.266]
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[1200]	rmse=0.020804	lr=0.248437
[1,0]<stderr>:INFO:root:Epoch[116] Rank[0] Batch[1251]	Time cost=399.82	Train-metric=0.020806
[1,0]<stderr>:INFO:root:Epoch[116]	Speed: 3204.00 samples/sec
[1,0]<stderr>:INFO:root:Epoch[117] Batch[100] Loss[2.884]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[100]	rmse=0.020768	lr=0.248160
[1,0]<stderr>:INFO:root:Epoch[117] Batch[200] Loss[4.880]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[200]	rmse=0.020747	lr=0.247976
[1,0]<stderr>:INFO:root:Epoch[117] Batch[300] Loss[3.722]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[300]	rmse=0.020801	lr=0.247792
[1,0]<stderr>:INFO:root:Epoch[117] Batch[400] Loss[2.711]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[400]	rmse=0.020776	lr=0.247608
[1,0]<stderr>:INFO:root:Epoch[117] Batch[500] Loss[3.696]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[500]	rmse=0.020799	lr=0.247424
[1,0]<stderr>:INFO:root:Epoch[117] Batch[600] Loss[3.060]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[600]	rmse=0.020763	lr=0.247240
[1,0]<stderr>:INFO:root:Epoch[117] Batch[700] Loss[4.458]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[700]	rmse=0.020780	lr=0.247056
[1,0]<stderr>:INFO:root:Epoch[117] Batch[800] Loss[5.156]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[800]	rmse=0.020782	lr=0.246872
[1,0]<stderr>:INFO:root:Epoch[117] Batch[900] Loss[2.957]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[900]	rmse=0.020808	lr=0.246688
[1,0]<stderr>:INFO:root:Epoch[117] Batch[1000] Loss[4.821]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[1000]	rmse=0.020792	lr=0.246504
[1,0]<stderr>:INFO:root:Epoch[117] Batch[1100] Loss[3.195]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[1100]	rmse=0.020789	lr=0.246320
[1,0]<stderr>:INFO:root:Epoch[117] Batch[1200] Loss[2.641]
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[1200]	rmse=0.020803	lr=0.246136
[1,0]<stderr>:INFO:root:Epoch[117] Rank[0] Batch[1251]	Time cost=399.75	Train-metric=0.020820
[1,0]<stderr>:INFO:root:Epoch[117]	Speed: 3204.57 samples/sec
[1,0]<stderr>:INFO:root:Epoch[118] Batch[100] Loss[2.710]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[100]	rmse=0.020881	lr=0.245857
[1,0]<stderr>:INFO:root:Epoch[118] Batch[200] Loss[3.210]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[200]	rmse=0.020797	lr=0.245673
[1,0]<stderr>:INFO:root:Epoch[118] Batch[300] Loss[4.836]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[300]	rmse=0.020731	lr=0.245488
[1,0]<stderr>:INFO:root:Epoch[118] Batch[400] Loss[2.739]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[400]	rmse=0.020721	lr=0.245304
[1,0]<stderr>:INFO:root:Epoch[118] Batch[500] Loss[2.836]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[500]	rmse=0.020731	lr=0.245120
[1,0]<stderr>:INFO:root:Epoch[118] Batch[600] Loss[3.016]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[600]	rmse=0.020758	lr=0.244935
[1,0]<stderr>:INFO:root:Epoch[118] Batch[700] Loss[3.161]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[700]	rmse=0.020749	lr=0.244751
[1,0]<stderr>:INFO:root:Epoch[118] Batch[800] Loss[2.973]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[800]	rmse=0.020741	lr=0.244566
[1,0]<stderr>:INFO:root:Epoch[118] Batch[900] Loss[3.143]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[900]	rmse=0.020765	lr=0.244381
[1,0]<stderr>:INFO:root:Epoch[118] Batch[1000] Loss[3.102]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[1000]	rmse=0.020764	lr=0.244197
[1,0]<stderr>:INFO:root:Epoch[118] Batch[1100] Loss[2.981]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[1100]	rmse=0.020785	lr=0.244012
[1,0]<stderr>:INFO:root:Epoch[118] Batch[1200] Loss[4.288]
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[1200]	rmse=0.020787	lr=0.243827
[1,0]<stderr>:INFO:root:Epoch[118] Rank[0] Batch[1251]	Time cost=398.96	Train-metric=0.020789
[1,0]<stderr>:INFO:root:Epoch[118]	Speed: 3210.88 samples/sec
[1,0]<stderr>:INFO:root:Epoch[119] Batch[100] Loss[2.782]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[100]	rmse=0.020557	lr=0.243548
[1,0]<stderr>:INFO:root:Epoch[119] Batch[200] Loss[2.671]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[200]	rmse=0.020632	lr=0.243363
[1,0]<stderr>:INFO:root:Epoch[119] Batch[300] Loss[3.082]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[300]	rmse=0.020698	lr=0.243178
[1,0]<stderr>:INFO:root:Epoch[119] Batch[400] Loss[5.082]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[400]	rmse=0.020691	lr=0.242993
[1,0]<stderr>:INFO:root:Epoch[119] Batch[500] Loss[3.179]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[500]	rmse=0.020717	lr=0.242808
[1,0]<stderr>:INFO:root:Epoch[119] Batch[600] Loss[2.786]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[600]	rmse=0.020704	lr=0.242623
[1,0]<stderr>:INFO:root:Epoch[119] Batch[700] Loss[2.836]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[700]	rmse=0.020705	lr=0.242438
[1,0]<stderr>:INFO:root:Epoch[119] Batch[800] Loss[2.692]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[800]	rmse=0.020684	lr=0.242253
[1,0]<stderr>:INFO:root:Epoch[119] Batch[900] Loss[2.704]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[900]	rmse=0.020691	lr=0.242068
[1,0]<stderr>:INFO:root:Epoch[119] Batch[1000] Loss[2.789]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[1000]	rmse=0.020710	lr=0.241883
[1,0]<stderr>:INFO:root:Epoch[119] Batch[1100] Loss[3.921]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[1100]	rmse=0.020740	lr=0.241698
[1,0]<stderr>:INFO:root:Epoch[119] Batch[1200] Loss[2.745]
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[1200]	rmse=0.020741	lr=0.241513
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0] Batch[1251]	Time cost=398.93	Train-metric=0.020749
[1,0]<stderr>:INFO:root:Epoch[119]	Speed: 3211.12 samples/sec
[1,0]<stderr>:INFO:root:Epoch[119] Rank[0]	Validation-accuracy=0.625240	Validation-top_k_accuracy_5=0.851320
[1,0]<stderr>:INFO:root:Epoch[120] Batch[100] Loss[3.525]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[100]	rmse=0.020804	lr=0.241233
[1,0]<stderr>:INFO:root:Epoch[120] Batch[200] Loss[3.359]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[200]	rmse=0.020859	lr=0.241047
[1,0]<stderr>:INFO:root:Epoch[120] Batch[300] Loss[5.126]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[300]	rmse=0.020857	lr=0.240862
[1,0]<stderr>:INFO:root:Epoch[120] Batch[400] Loss[3.322]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[400]	rmse=0.020807	lr=0.240677
[1,0]<stderr>:INFO:root:Epoch[120] Batch[500] Loss[4.615]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[500]	rmse=0.020773	lr=0.240491
[1,0]<stderr>:INFO:root:Epoch[120] Batch[600] Loss[2.887]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[600]	rmse=0.020764	lr=0.240306
[1,0]<stderr>:INFO:root:Epoch[120] Batch[700] Loss[5.239]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[700]	rmse=0.020733	lr=0.240120
[1,0]<stderr>:INFO:root:Epoch[120] Batch[800] Loss[2.835]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[800]	rmse=0.020745	lr=0.239935
[1,0]<stderr>:INFO:root:Epoch[120] Batch[900] Loss[3.168]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[900]	rmse=0.020739	lr=0.239749
[1,0]<stderr>:INFO:root:Epoch[120] Batch[1000] Loss[3.679]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[1000]	rmse=0.020733	lr=0.239564
[1,0]<stderr>:INFO:root:Epoch[120] Batch[1100] Loss[2.856]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[1100]	rmse=0.020725	lr=0.239378
[1,0]<stderr>:INFO:root:Epoch[120] Batch[1200] Loss[3.223]
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[1200]	rmse=0.020727	lr=0.239192
[1,0]<stderr>:INFO:root:Epoch[120] Rank[0] Batch[1251]	Time cost=399.33	Train-metric=0.020726
[1,0]<stderr>:INFO:root:Epoch[120]	Speed: 3207.90 samples/sec
[1,0]<stderr>:INFO:root:Epoch[121] Batch[100] Loss[2.973]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[100]	rmse=0.020737	lr=0.238912
[1,0]<stderr>:INFO:root:Epoch[121] Batch[200] Loss[2.796]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[200]	rmse=0.020774	lr=0.238726
[1,0]<stderr>:INFO:root:Epoch[121] Batch[300] Loss[2.802]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[300]	rmse=0.020705	lr=0.238540
[1,0]<stderr>:INFO:root:Epoch[121] Batch[400] Loss[5.272]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[400]	rmse=0.020665	lr=0.238354
[1,0]<stderr>:INFO:root:Epoch[121] Batch[500] Loss[3.437]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[500]	rmse=0.020655	lr=0.238168
[1,0]<stderr>:INFO:root:Epoch[121] Batch[600] Loss[2.865]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[600]	rmse=0.020644	lr=0.237982
[1,0]<stderr>:INFO:root:Epoch[121] Batch[700] Loss[4.026]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[700]	rmse=0.020641	lr=0.237797
[1,0]<stderr>:INFO:root:Epoch[121] Batch[800] Loss[4.513]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[800]	rmse=0.020677	lr=0.237611
[1,0]<stderr>:INFO:root:Epoch[121] Batch[900] Loss[2.855]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[900]	rmse=0.020686	lr=0.237425
[1,0]<stderr>:INFO:root:Epoch[121] Batch[1000] Loss[3.302]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[1000]	rmse=0.020684	lr=0.237238
[1,0]<stderr>:INFO:root:Epoch[121] Batch[1100] Loss[2.821]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[1100]	rmse=0.020685	lr=0.237052
[1,0]<stderr>:INFO:root:Epoch[121] Batch[1200] Loss[5.108]
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[1200]	rmse=0.020692	lr=0.236866
[1,0]<stderr>:INFO:root:Epoch[121] Rank[0] Batch[1251]	Time cost=402.32	Train-metric=0.020700
[1,0]<stderr>:INFO:root:Epoch[121]	Speed: 3184.12 samples/sec
[1,0]<stderr>:INFO:root:Epoch[122] Batch[100] Loss[2.950]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[100]	rmse=0.020475	lr=0.236585
[1,0]<stderr>:INFO:root:Epoch[122] Batch[200] Loss[2.562]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[200]	rmse=0.020519	lr=0.236399
[1,0]<stderr>:INFO:root:Epoch[122] Batch[300] Loss[2.711]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[300]	rmse=0.020577	lr=0.236213
[1,0]<stderr>:INFO:root:Epoch[122] Batch[400] Loss[2.927]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[400]	rmse=0.020601	lr=0.236027
[1,0]<stderr>:INFO:root:Epoch[122] Batch[500] Loss[2.771]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[500]	rmse=0.020654	lr=0.235840
[1,0]<stderr>:INFO:root:Epoch[122] Batch[600] Loss[2.768]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[600]	rmse=0.020652	lr=0.235654
[1,0]<stderr>:INFO:root:Epoch[122] Batch[700] Loss[5.108]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[700]	rmse=0.020667	lr=0.235468
[1,0]<stderr>:INFO:root:Epoch[122] Batch[800] Loss[2.812]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[800]	rmse=0.020651	lr=0.235281
[1,0]<stderr>:INFO:root:Epoch[122] Batch[900] Loss[2.712]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[900]	rmse=0.020657	lr=0.235095
[1,0]<stderr>:INFO:root:Epoch[122] Batch[1000] Loss[5.167]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[1000]	rmse=0.020649	lr=0.234908
[1,0]<stderr>:INFO:root:Epoch[122] Batch[1100] Loss[3.191]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[1100]	rmse=0.020658	lr=0.234722
[1,0]<stderr>:INFO:root:Epoch[122] Batch[1200] Loss[3.367]
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[1200]	rmse=0.020673	lr=0.234535
[1,0]<stderr>:INFO:root:Epoch[122] Rank[0] Batch[1251]	Time cost=402.27	Train-metric=0.020667
[1,0]<stderr>:INFO:root:Epoch[122]	Speed: 3184.52 samples/sec
[1,0]<stderr>:INFO:root:Epoch[123] Batch[100] Loss[2.923]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[100]	rmse=0.020532	lr=0.234254
[1,0]<stderr>:INFO:root:Epoch[123] Batch[200] Loss[2.651]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[200]	rmse=0.020560	lr=0.234067
[1,0]<stderr>:INFO:root:Epoch[123] Batch[300] Loss[2.852]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[300]	rmse=0.020538	lr=0.233880
[1,0]<stderr>:INFO:root:Epoch[123] Batch[400] Loss[2.815]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[400]	rmse=0.020601	lr=0.233694
[1,0]<stderr>:INFO:root:Epoch[123] Batch[500] Loss[2.828]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[500]	rmse=0.020624	lr=0.233507
[1,0]<stderr>:INFO:root:Epoch[123] Batch[600] Loss[5.290]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[600]	rmse=0.020605	lr=0.233320
[1,0]<stderr>:INFO:root:Epoch[123] Batch[700] Loss[2.948]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[700]	rmse=0.020618	lr=0.233134
[1,0]<stderr>:INFO:root:Epoch[123] Batch[800] Loss[3.394]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[800]	rmse=0.020611	lr=0.232947
[1,0]<stderr>:INFO:root:Epoch[123] Batch[900] Loss[4.895]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[900]	rmse=0.020652	lr=0.232760
[1,0]<stderr>:INFO:root:Epoch[123] Batch[1000] Loss[4.918]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[1000]	rmse=0.020675	lr=0.232573
[1,0]<stderr>:INFO:root:Epoch[123] Batch[1100] Loss[4.947]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[1100]	rmse=0.020662	lr=0.232386
[1,0]<stderr>:INFO:root:Epoch[123] Batch[1200] Loss[3.793]
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[1200]	rmse=0.020679	lr=0.232199
[1,0]<stderr>:INFO:root:Epoch[123] Rank[0] Batch[1251]	Time cost=399.56	Train-metric=0.020695
[1,0]<stderr>:INFO:root:Epoch[123]	Speed: 3206.12 samples/sec
[1,0]<stderr>:INFO:root:Epoch[124] Batch[100] Loss[3.993]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[100]	rmse=0.020621	lr=0.231917
[1,0]<stderr>:INFO:root:Epoch[124] Batch[200] Loss[2.720]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[200]	rmse=0.020564	lr=0.231730
[1,0]<stderr>:INFO:root:Epoch[124] Batch[300] Loss[2.915]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[300]	rmse=0.020637	lr=0.231543
[1,0]<stderr>:INFO:root:Epoch[124] Batch[400] Loss[2.749]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[400]	rmse=0.020632	lr=0.231356
[1,0]<stderr>:INFO:root:Epoch[124] Batch[500] Loss[2.815]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[500]	rmse=0.020644	lr=0.231169
[1,0]<stderr>:INFO:root:Epoch[124] Batch[600] Loss[2.857]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[600]	rmse=0.020648	lr=0.230982
[1,0]<stderr>:INFO:root:Epoch[124] Batch[700] Loss[3.067]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[700]	rmse=0.020656	lr=0.230795
[1,0]<stderr>:INFO:root:Epoch[124] Batch[800] Loss[3.161]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[800]	rmse=0.020645	lr=0.230608
[1,0]<stderr>:INFO:root:Epoch[124] Batch[900] Loss[3.416]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[900]	rmse=0.020648	lr=0.230421
[1,0]<stderr>:INFO:root:Epoch[124] Batch[1000] Loss[3.957]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[1000]	rmse=0.020651	lr=0.230233
[1,0]<stderr>:INFO:root:Epoch[124] Batch[1100] Loss[4.886]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[1100]	rmse=0.020642	lr=0.230046
[1,0]<stderr>:INFO:root:Epoch[124] Batch[1200] Loss[2.501]
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[1200]	rmse=0.020661	lr=0.229859
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0] Batch[1251]	Time cost=400.19	Train-metric=0.020665
[1,0]<stderr>:INFO:root:Epoch[124]	Speed: 3201.08 samples/sec
[1,0]<stderr>:INFO:root:Epoch[124] Rank[0]	Validation-accuracy=0.621100	Validation-top_k_accuracy_5=0.848380
[1,0]<stderr>:INFO:root:Epoch[125] Batch[100] Loss[2.947]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[100]	rmse=0.020766	lr=0.229576
[1,0]<stderr>:INFO:root:Epoch[125] Batch[200] Loss[3.553]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[200]	rmse=0.020650	lr=0.229389
[1,0]<stderr>:INFO:root:Epoch[125] Batch[300] Loss[3.032]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[300]	rmse=0.020644	lr=0.229201
[1,0]<stderr>:INFO:root:Epoch[125] Batch[400] Loss[3.044]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[400]	rmse=0.020653	lr=0.229014
[1,0]<stderr>:INFO:root:Epoch[125] Batch[500] Loss[3.154]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[500]	rmse=0.020620	lr=0.228827
[1,0]<stderr>:INFO:root:Epoch[125] Batch[600] Loss[4.434]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[600]	rmse=0.020644	lr=0.228639
[1,0]<stderr>:INFO:root:Epoch[125] Batch[700] Loss[2.596]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[700]	rmse=0.020674	lr=0.228452
[1,0]<stderr>:INFO:root:Epoch[125] Batch[800] Loss[4.445]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[800]	rmse=0.020672	lr=0.228264
[1,0]<stderr>:INFO:root:Epoch[125] Batch[900] Loss[3.486]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[900]	rmse=0.020682	lr=0.228077
[1,0]<stderr>:INFO:root:Epoch[125] Batch[1000] Loss[2.853]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[1000]	rmse=0.020678	lr=0.227889
[1,0]<stderr>:INFO:root:Epoch[125] Batch[1100] Loss[3.769]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[1100]	rmse=0.020695	lr=0.227702
[1,0]<stderr>:INFO:root:Epoch[125] Batch[1200] Loss[4.128]
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[1200]	rmse=0.020700	lr=0.227514
[1,0]<stderr>:INFO:root:Epoch[125] Rank[0] Batch[1251]	Time cost=399.05	Train-metric=0.020696
[1,0]<stderr>:INFO:root:Epoch[125]	Speed: 3210.16 samples/sec
[1,0]<stderr>:INFO:root:Epoch[126] Batch[100] Loss[3.064]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[100]	rmse=0.020212	lr=0.227231
[1,0]<stderr>:INFO:root:Epoch[126] Batch[200] Loss[4.627]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[200]	rmse=0.020406	lr=0.227043
[1,0]<stderr>:INFO:root:Epoch[126] Batch[300] Loss[2.569]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[300]	rmse=0.020445	lr=0.226856
[1,0]<stderr>:INFO:root:Epoch[126] Batch[400] Loss[3.429]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[400]	rmse=0.020471	lr=0.226668
[1,0]<stderr>:INFO:root:Epoch[126] Batch[500] Loss[2.748]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[500]	rmse=0.020519	lr=0.226480
[1,0]<stderr>:INFO:root:Epoch[126] Batch[600] Loss[2.785]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[600]	rmse=0.020539	lr=0.226293
[1,0]<stderr>:INFO:root:Epoch[126] Batch[700] Loss[4.884]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[700]	rmse=0.020544	lr=0.226105
[1,0]<stderr>:INFO:root:Epoch[126] Batch[800] Loss[5.023]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[800]	rmse=0.020553	lr=0.225917
[1,0]<stderr>:INFO:root:Epoch[126] Batch[900] Loss[3.068]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[900]	rmse=0.020587	lr=0.225729
[1,0]<stderr>:INFO:root:Epoch[126] Batch[1000] Loss[5.411]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[1000]	rmse=0.020620	lr=0.225542
[1,0]<stderr>:INFO:root:Epoch[126] Batch[1100] Loss[4.314]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[1100]	rmse=0.020632	lr=0.225354
[1,0]<stderr>:INFO:root:Epoch[126] Batch[1200] Loss[5.297]
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[1200]	rmse=0.020635	lr=0.225166
[1,0]<stderr>:INFO:root:Epoch[126] Rank[0] Batch[1251]	Time cost=398.59	Train-metric=0.020641
[1,0]<stderr>:INFO:root:Epoch[126]	Speed: 3213.91 samples/sec
[1,0]<stderr>:INFO:root:Epoch[127] Batch[100] Loss[2.885]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[100]	rmse=0.020411	lr=0.224882
[1,0]<stderr>:INFO:root:Epoch[127] Batch[200] Loss[3.019]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[200]	rmse=0.020532	lr=0.224694
[1,0]<stderr>:INFO:root:Epoch[127] Batch[300] Loss[4.956]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[300]	rmse=0.020539	lr=0.224506
[1,0]<stderr>:INFO:root:Epoch[127] Batch[400] Loss[3.798]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[400]	rmse=0.020592	lr=0.224318
[1,0]<stderr>:INFO:root:Epoch[127] Batch[500] Loss[4.537]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[500]	rmse=0.020569	lr=0.224130
[1,0]<stderr>:INFO:root:Epoch[127] Batch[600] Loss[4.720]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[600]	rmse=0.020574	lr=0.223942
[1,0]<stderr>:INFO:root:Epoch[127] Batch[700] Loss[3.216]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[700]	rmse=0.020589	lr=0.223754
[1,0]<stderr>:INFO:root:Epoch[127] Batch[800] Loss[3.120]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[800]	rmse=0.020616	lr=0.223566
[1,0]<stderr>:INFO:root:Epoch[127] Batch[900] Loss[2.716]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[900]	rmse=0.020625	lr=0.223378
[1,0]<stderr>:INFO:root:Epoch[127] Batch[1000] Loss[5.124]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[1000]	rmse=0.020632	lr=0.223190
[1,0]<stderr>:INFO:root:Epoch[127] Batch[1100] Loss[3.152]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[1100]	rmse=0.020645	lr=0.223002
[1,0]<stderr>:INFO:root:Epoch[127] Batch[1200] Loss[2.823]
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[1200]	rmse=0.020643	lr=0.222814
[1,0]<stderr>:INFO:root:Epoch[127] Rank[0] Batch[1251]	Time cost=399.06	Train-metric=0.020642
[1,0]<stderr>:INFO:root:Epoch[127]	Speed: 3210.14 samples/sec
[1,0]<stderr>:INFO:root:Epoch[128] Batch[100] Loss[4.015]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[100]	rmse=0.020733	lr=0.222530
[1,0]<stderr>:INFO:root:Epoch[128] Batch[200] Loss[2.924]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[200]	rmse=0.020638	lr=0.222342
[1,0]<stderr>:INFO:root:Epoch[128] Batch[300] Loss[3.182]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[300]	rmse=0.020578	lr=0.222153
[1,0]<stderr>:INFO:root:Epoch[128] Batch[400] Loss[3.418]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[400]	rmse=0.020628	lr=0.221965
[1,0]<stderr>:INFO:root:Epoch[128] Batch[500] Loss[4.656]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[500]	rmse=0.020553	lr=0.221777
[1,0]<stderr>:INFO:root:Epoch[128] Batch[600] Loss[3.384]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[600]	rmse=0.020524	lr=0.221589
[1,0]<stderr>:INFO:root:Epoch[128] Batch[700] Loss[2.936]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[700]	rmse=0.020548	lr=0.221400
[1,0]<stderr>:INFO:root:Epoch[128] Batch[800] Loss[2.570]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[800]	rmse=0.020564	lr=0.221212
[1,0]<stderr>:INFO:root:Epoch[128] Batch[900] Loss[3.013]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[900]	rmse=0.020580	lr=0.221024
[1,0]<stderr>:INFO:root:Epoch[128] Batch[1000] Loss[3.570]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[1000]	rmse=0.020602	lr=0.220835
[1,0]<stderr>:INFO:root:Epoch[128] Batch[1100] Loss[2.744]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[1100]	rmse=0.020605	lr=0.220647
[1,0]<stderr>:INFO:root:Epoch[128] Batch[1200] Loss[2.803]
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[1200]	rmse=0.020608	lr=0.220459
[1,0]<stderr>:INFO:root:Epoch[128] Rank[0] Batch[1251]	Time cost=398.84	Train-metric=0.020613
[1,0]<stderr>:INFO:root:Epoch[128]	Speed: 3211.84 samples/sec
[1,0]<stderr>:INFO:root:Epoch[129] Batch[100] Loss[5.083]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[100]	rmse=0.020456	lr=0.220174
[1,0]<stderr>:INFO:root:Epoch[129] Batch[200] Loss[3.125]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[200]	rmse=0.020497	lr=0.219986
[1,0]<stderr>:INFO:root:Epoch[129] Batch[300] Loss[4.845]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[300]	rmse=0.020569	lr=0.219797
[1,0]<stderr>:INFO:root:Epoch[129] Batch[400] Loss[2.807]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[400]	rmse=0.020590	lr=0.219609
[1,0]<stderr>:INFO:root:Epoch[129] Batch[500] Loss[3.176]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[500]	rmse=0.020585	lr=0.219420
[1,0]<stderr>:INFO:root:Epoch[129] Batch[600] Loss[3.009]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[600]	rmse=0.020610	lr=0.219232
[1,0]<stderr>:INFO:root:Epoch[129] Batch[700] Loss[2.891]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[700]	rmse=0.020615	lr=0.219043
[1,0]<stderr>:INFO:root:Epoch[129] Batch[800] Loss[3.220]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[800]	rmse=0.020629	lr=0.218855
[1,0]<stderr>:INFO:root:Epoch[129] Batch[900] Loss[3.927]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[900]	rmse=0.020638	lr=0.218666
[1,0]<stderr>:INFO:root:Epoch[129] Batch[1000] Loss[3.477]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[1000]	rmse=0.020635	lr=0.218478
[1,0]<stderr>:INFO:root:Epoch[129] Batch[1100] Loss[2.991]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[1100]	rmse=0.020640	lr=0.218289
[1,0]<stderr>:INFO:root:Epoch[129] Batch[1200] Loss[2.818]
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[1200]	rmse=0.020633	lr=0.218101
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0] Batch[1251]	Time cost=399.58	Train-metric=0.020636
[1,0]<stderr>:INFO:root:Epoch[129]	Speed: 3205.91 samples/sec
[1,0]<stderr>:INFO:root:Epoch[129] Rank[0]	Validation-accuracy=0.626560	Validation-top_k_accuracy_5=0.855000
[1,0]<stderr>:INFO:root:Epoch[130] Batch[100] Loss[2.752]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[100]	rmse=0.020431	lr=0.217816
[1,0]<stderr>:INFO:root:Epoch[130] Batch[200] Loss[4.392]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[200]	rmse=0.020531	lr=0.217627
[1,0]<stderr>:INFO:root:Epoch[130] Batch[300] Loss[4.735]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[300]	rmse=0.020550	lr=0.217438
[1,0]<stderr>:INFO:root:Epoch[130] Batch[400] Loss[2.998]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[400]	rmse=0.020563	lr=0.217250
[1,0]<stderr>:INFO:root:Epoch[130] Batch[500] Loss[3.088]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[500]	rmse=0.020578	lr=0.217061
[1,0]<stderr>:INFO:root:Epoch[130] Batch[600] Loss[2.999]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[600]	rmse=0.020596	lr=0.216872
[1,0]<stderr>:INFO:root:Epoch[130] Batch[700] Loss[2.308]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[700]	rmse=0.020613	lr=0.216684
[1,0]<stderr>:INFO:root:Epoch[130] Batch[800] Loss[3.819]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[800]	rmse=0.020664	lr=0.216495
[1,0]<stderr>:INFO:root:Epoch[130] Batch[900] Loss[4.128]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[900]	rmse=0.020650	lr=0.216306
[1,0]<stderr>:INFO:root:Epoch[130] Batch[1000] Loss[2.948]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[1000]	rmse=0.020640	lr=0.216118
[1,0]<stderr>:INFO:root:Epoch[130] Batch[1100] Loss[4.915]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[1100]	rmse=0.020642	lr=0.215929
[1,0]<stderr>:INFO:root:Epoch[130] Batch[1200] Loss[2.979]
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[1200]	rmse=0.020650	lr=0.215740
[1,0]<stderr>:INFO:root:Epoch[130] Rank[0] Batch[1251]	Time cost=398.55	Train-metric=0.020646
[1,0]<stderr>:INFO:root:Epoch[130]	Speed: 3214.24 samples/sec
[1,0]<stderr>:INFO:root:Epoch[131] Batch[100] Loss[2.979]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[100]	rmse=0.020655	lr=0.215455
[1,0]<stderr>:INFO:root:Epoch[131] Batch[200] Loss[2.870]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[200]	rmse=0.020476	lr=0.215266
[1,0]<stderr>:INFO:root:Epoch[131] Batch[300] Loss[2.849]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[300]	rmse=0.020570	lr=0.215077
[1,0]<stderr>:INFO:root:Epoch[131] Batch[400] Loss[5.382]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[400]	rmse=0.020543	lr=0.214888
[1,0]<stderr>:INFO:root:Epoch[131] Batch[500] Loss[2.620]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[500]	rmse=0.020536	lr=0.214700
[1,0]<stderr>:INFO:root:Epoch[131] Batch[600] Loss[2.815]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[600]	rmse=0.020531	lr=0.214511
[1,0]<stderr>:INFO:root:Epoch[131] Batch[700] Loss[2.549]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[700]	rmse=0.020530	lr=0.214322
[1,0]<stderr>:INFO:root:Epoch[131] Batch[800] Loss[2.884]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[800]	rmse=0.020533	lr=0.214133
[1,0]<stderr>:INFO:root:Epoch[131] Batch[900] Loss[5.264]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[900]	rmse=0.020551	lr=0.213944
[1,0]<stderr>:INFO:root:Epoch[131] Batch[1000] Loss[4.917]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[1000]	rmse=0.020562	lr=0.213755
[1,0]<stderr>:INFO:root:Epoch[131] Batch[1100] Loss[3.244]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[1100]	rmse=0.020552	lr=0.213566
[1,0]<stderr>:INFO:root:Epoch[131] Batch[1200] Loss[3.049]
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[1200]	rmse=0.020572	lr=0.213377
[1,0]<stderr>:INFO:root:Epoch[131] Rank[0] Batch[1251]	Time cost=398.97	Train-metric=0.020573
[1,0]<stderr>:INFO:root:Epoch[131]	Speed: 3210.85 samples/sec
[1,0]<stderr>:INFO:root:Epoch[132] Batch[100] Loss[4.977]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[100]	rmse=0.020711	lr=0.213092
[1,0]<stderr>:INFO:root:Epoch[132] Batch[200] Loss[4.915]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[200]	rmse=0.020656	lr=0.212903
[1,0]<stderr>:INFO:root:Epoch[132] Batch[300] Loss[3.104]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[300]	rmse=0.020629	lr=0.212714
[1,0]<stderr>:INFO:root:Epoch[132] Batch[400] Loss[3.623]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[400]	rmse=0.020616	lr=0.212525
[1,0]<stderr>:INFO:root:Epoch[132] Batch[500] Loss[3.995]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[500]	rmse=0.020541	lr=0.212336
[1,0]<stderr>:INFO:root:Epoch[132] Batch[600] Loss[2.541]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[600]	rmse=0.020556	lr=0.212147
[1,0]<stderr>:INFO:root:Epoch[132] Batch[700] Loss[2.840]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[700]	rmse=0.020547	lr=0.211958
[1,0]<stderr>:INFO:root:Epoch[132] Batch[800] Loss[4.673]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[800]	rmse=0.020560	lr=0.211769
[1,0]<stderr>:INFO:root:Epoch[132] Batch[900] Loss[2.954]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[900]	rmse=0.020581	lr=0.211580
[1,0]<stderr>:INFO:root:Epoch[132] Batch[1000] Loss[4.423]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[1000]	rmse=0.020578	lr=0.211391
[1,0]<stderr>:INFO:root:Epoch[132] Batch[1100] Loss[2.964]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[1100]	rmse=0.020599	lr=0.211202
[1,0]<stderr>:INFO:root:Epoch[132] Batch[1200] Loss[2.993]
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[1200]	rmse=0.020608	lr=0.211012
[1,0]<stderr>:INFO:root:Epoch[132] Rank[0] Batch[1251]	Time cost=399.24	Train-metric=0.020609
[1,0]<stderr>:INFO:root:Epoch[132]	Speed: 3208.69 samples/sec
[1,0]<stderr>:INFO:root:Epoch[133] Batch[100] Loss[3.051]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[100]	rmse=0.020667	lr=0.210727
[1,0]<stderr>:INFO:root:Epoch[133] Batch[200] Loss[4.816]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[200]	rmse=0.020556	lr=0.210538
[1,0]<stderr>:INFO:root:Epoch[133] Batch[300] Loss[5.240]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[300]	rmse=0.020536	lr=0.210349
[1,0]<stderr>:INFO:root:Epoch[133] Batch[400] Loss[3.667]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[400]	rmse=0.020587	lr=0.210160
[1,0]<stderr>:INFO:root:Epoch[133] Batch[500] Loss[2.564]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[500]	rmse=0.020576	lr=0.209970
[1,0]<stderr>:INFO:root:Epoch[133] Batch[600] Loss[2.997]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[600]	rmse=0.020563	lr=0.209781
[1,0]<stderr>:INFO:root:Epoch[133] Batch[700] Loss[5.187]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[700]	rmse=0.020567	lr=0.209592
[1,0]<stderr>:INFO:root:Epoch[133] Batch[800] Loss[3.047]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[800]	rmse=0.020582	lr=0.209403
[1,0]<stderr>:INFO:root:Epoch[133] Batch[900] Loss[5.049]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[900]	rmse=0.020558	lr=0.209214
[1,0]<stderr>:INFO:root:Epoch[133] Batch[1000] Loss[3.129]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[1000]	rmse=0.020548	lr=0.209025
[1,0]<stderr>:INFO:root:Epoch[133] Batch[1100] Loss[2.732]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[1100]	rmse=0.020567	lr=0.208835
[1,0]<stderr>:INFO:root:Epoch[133] Batch[1200] Loss[4.389]
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[1200]	rmse=0.020564	lr=0.208646
[1,0]<stderr>:INFO:root:Epoch[133] Rank[0] Batch[1251]	Time cost=399.61	Train-metric=0.020571
[1,0]<stderr>:INFO:root:Epoch[133]	Speed: 3205.66 samples/sec
[1,0]<stderr>:INFO:root:Epoch[134] Batch[100] Loss[5.255]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[100]	rmse=0.020327	lr=0.208360
[1,0]<stderr>:INFO:root:Epoch[134] Batch[200] Loss[2.776]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[200]	rmse=0.020426	lr=0.208171
[1,0]<stderr>:INFO:root:Epoch[134] Batch[300] Loss[2.604]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[300]	rmse=0.020471	lr=0.207982
[1,0]<stderr>:INFO:root:Epoch[134] Batch[400] Loss[3.835]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[400]	rmse=0.020469	lr=0.207793
[1,0]<stderr>:INFO:root:Epoch[134] Batch[500] Loss[5.063]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[500]	rmse=0.020498	lr=0.207604
[1,0]<stderr>:INFO:root:Epoch[134] Batch[600] Loss[3.410]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[600]	rmse=0.020496	lr=0.207414
[1,0]<stderr>:INFO:root:Epoch[134] Batch[700] Loss[2.710]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[700]	rmse=0.020506	lr=0.207225
[1,0]<stderr>:INFO:root:Epoch[134] Batch[800] Loss[2.623]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[800]	rmse=0.020520	lr=0.207036
[1,0]<stderr>:INFO:root:Epoch[134] Batch[900] Loss[2.856]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[900]	rmse=0.020529	lr=0.206847
[1,0]<stderr>:INFO:root:Epoch[134] Batch[1000] Loss[3.929]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[1000]	rmse=0.020526	lr=0.206657
[1,0]<stderr>:INFO:root:Epoch[134] Batch[1100] Loss[3.269]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[1100]	rmse=0.020522	lr=0.206468
[1,0]<stderr>:INFO:root:Epoch[134] Batch[1200] Loss[4.449]
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[1200]	rmse=0.020534	lr=0.206279
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0] Batch[1251]	Time cost=399.35	Train-metric=0.020532
[1,0]<stderr>:INFO:root:Epoch[134]	Speed: 3207.73 samples/sec
[1,0]<stderr>:INFO:root:Epoch[134] Rank[0]	Validation-accuracy=0.633520	Validation-top_k_accuracy_5=0.855260
[1,0]<stderr>:INFO:root:Epoch[135] Batch[100] Loss[2.670]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[100]	rmse=0.020461	lr=0.205993
[1,0]<stderr>:INFO:root:Epoch[135] Batch[200] Loss[2.979]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[200]	rmse=0.020396	lr=0.205804
[1,0]<stderr>:INFO:root:Epoch[135] Batch[300] Loss[2.865]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[300]	rmse=0.020417	lr=0.205614
[1,0]<stderr>:INFO:root:Epoch[135] Batch[400] Loss[2.918]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[400]	rmse=0.020394	lr=0.205425
[1,0]<stderr>:INFO:root:Epoch[135] Batch[500] Loss[2.808]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[500]	rmse=0.020397	lr=0.205236
[1,0]<stderr>:INFO:root:Epoch[135] Batch[600] Loss[2.772]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[600]	rmse=0.020377	lr=0.205046
[1,0]<stderr>:INFO:root:Epoch[135] Batch[700] Loss[2.565]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[700]	rmse=0.020400	lr=0.204857
[1,0]<stderr>:INFO:root:Epoch[135] Batch[800] Loss[3.879]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[800]	rmse=0.020400	lr=0.204668
[1,0]<stderr>:INFO:root:Epoch[135] Batch[900] Loss[5.385]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[900]	rmse=0.020425	lr=0.204478
[1,0]<stderr>:INFO:root:Epoch[135] Batch[1000] Loss[4.687]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[1000]	rmse=0.020440	lr=0.204289
[1,0]<stderr>:INFO:root:Epoch[135] Batch[1100] Loss[2.855]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[1100]	rmse=0.020451	lr=0.204100
[1,0]<stderr>:INFO:root:Epoch[135] Batch[1200] Loss[2.909]
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[1200]	rmse=0.020472	lr=0.203910
[1,0]<stderr>:INFO:root:Epoch[135] Rank[0] Batch[1251]	Time cost=399.71	Train-metric=0.020481
[1,0]<stderr>:INFO:root:Epoch[135]	Speed: 3204.87 samples/sec
[1,0]<stderr>:INFO:root:Epoch[136] Batch[100] Loss[3.021]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[100]	rmse=0.020312	lr=0.203624
[1,0]<stderr>:INFO:root:Epoch[136] Batch[200] Loss[2.918]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[200]	rmse=0.020279	lr=0.203435
[1,0]<stderr>:INFO:root:Epoch[136] Batch[300] Loss[4.671]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[300]	rmse=0.020321	lr=0.203246
[1,0]<stderr>:INFO:root:Epoch[136] Batch[400] Loss[2.968]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[400]	rmse=0.020411	lr=0.203056
[1,0]<stderr>:INFO:root:Epoch[136] Batch[500] Loss[2.746]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[500]	rmse=0.020443	lr=0.202867
[1,0]<stderr>:INFO:root:Epoch[136] Batch[600] Loss[4.421]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[600]	rmse=0.020453	lr=0.202678
[1,0]<stderr>:INFO:root:Epoch[136] Batch[700] Loss[4.731]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[700]	rmse=0.020489	lr=0.202488
[1,0]<stderr>:INFO:root:Epoch[136] Batch[800] Loss[3.606]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[800]	rmse=0.020497	lr=0.202299
[1,0]<stderr>:INFO:root:Epoch[136] Batch[900] Loss[5.157]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[900]	rmse=0.020527	lr=0.202110
[1,0]<stderr>:INFO:root:Epoch[136] Batch[1000] Loss[5.063]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[1000]	rmse=0.020532	lr=0.201920
[1,0]<stderr>:INFO:root:Epoch[136] Batch[1100] Loss[3.088]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[1100]	rmse=0.020538	lr=0.201731
[1,0]<stderr>:INFO:root:Epoch[136] Batch[1200] Loss[3.298]
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[1200]	rmse=0.020550	lr=0.201542
[1,0]<stderr>:INFO:root:Epoch[136] Rank[0] Batch[1251]	Time cost=399.11	Train-metric=0.020562
[1,0]<stderr>:INFO:root:Epoch[136]	Speed: 3209.70 samples/sec
[1,0]<stderr>:INFO:root:Epoch[137] Batch[100] Loss[5.215]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[100]	rmse=0.020477	lr=0.201256
[1,0]<stderr>:INFO:root:Epoch[137] Batch[200] Loss[3.150]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[200]	rmse=0.020490	lr=0.201066
[1,0]<stderr>:INFO:root:Epoch[137] Batch[300] Loss[3.194]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[300]	rmse=0.020554	lr=0.200877
[1,0]<stderr>:INFO:root:Epoch[137] Batch[400] Loss[3.180]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[400]	rmse=0.020540	lr=0.200687
[1,0]<stderr>:INFO:root:Epoch[137] Batch[500] Loss[3.499]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[500]	rmse=0.020538	lr=0.200498
[1,0]<stderr>:INFO:root:Epoch[137] Batch[600] Loss[3.063]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[600]	rmse=0.020530	lr=0.200309
[1,0]<stderr>:INFO:root:Epoch[137] Batch[700] Loss[2.600]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[700]	rmse=0.020517	lr=0.200119
[1,0]<stderr>:INFO:root:Epoch[137] Batch[800] Loss[3.072]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[800]	rmse=0.020546	lr=0.199930
[1,0]<stderr>:INFO:root:Epoch[137] Batch[900] Loss[3.025]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[900]	rmse=0.020558	lr=0.199741
[1,0]<stderr>:INFO:root:Epoch[137] Batch[1000] Loss[3.323]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[1000]	rmse=0.020542	lr=0.199551
[1,0]<stderr>:INFO:root:Epoch[137] Batch[1100] Loss[4.178]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[1100]	rmse=0.020534	lr=0.199362
[1,0]<stderr>:INFO:root:Epoch[137] Batch[1200] Loss[3.885]
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[1200]	rmse=0.020540	lr=0.199172
[1,0]<stderr>:INFO:root:Epoch[137] Rank[0] Batch[1251]	Time cost=398.92	Train-metric=0.020546
[1,0]<stderr>:INFO:root:Epoch[137]	Speed: 3211.23 samples/sec
[1,0]<stderr>:INFO:root:Epoch[138] Batch[100] Loss[2.632]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[100]	rmse=0.020267	lr=0.198886
[1,0]<stderr>:INFO:root:Epoch[138] Batch[200] Loss[2.925]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[200]	rmse=0.020418	lr=0.198697
[1,0]<stderr>:INFO:root:Epoch[138] Batch[300] Loss[4.241]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[300]	rmse=0.020504	lr=0.198508
[1,0]<stderr>:INFO:root:Epoch[138] Batch[400] Loss[3.333]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[400]	rmse=0.020486	lr=0.198318
[1,0]<stderr>:INFO:root:Epoch[138] Batch[500] Loss[4.565]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[500]	rmse=0.020492	lr=0.198129
[1,0]<stderr>:INFO:root:Epoch[138] Batch[600] Loss[3.253]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[600]	rmse=0.020487	lr=0.197940
[1,0]<stderr>:INFO:root:Epoch[138] Batch[700] Loss[3.604]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[700]	rmse=0.020491	lr=0.197750
[1,0]<stderr>:INFO:root:Epoch[138] Batch[800] Loss[2.763]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[800]	rmse=0.020478	lr=0.197561
[1,0]<stderr>:INFO:root:Epoch[138] Batch[900] Loss[2.710]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[900]	rmse=0.020493	lr=0.197372
[1,0]<stderr>:INFO:root:Epoch[138] Batch[1000] Loss[4.049]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[1000]	rmse=0.020481	lr=0.197182
[1,0]<stderr>:INFO:root:Epoch[138] Batch[1100] Loss[3.139]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[1100]	rmse=0.020495	lr=0.196993
[1,0]<stderr>:INFO:root:Epoch[138] Batch[1200] Loss[5.176]
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[1200]	rmse=0.020501	lr=0.196803
[1,0]<stderr>:INFO:root:Epoch[138] Rank[0] Batch[1251]	Time cost=398.74	Train-metric=0.020494
[1,0]<stderr>:INFO:root:Epoch[138]	Speed: 3212.66 samples/sec
[1,0]<stderr>:INFO:root:Epoch[139] Batch[100] Loss[3.681]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[100]	rmse=0.020201	lr=0.196518
[1,0]<stderr>:INFO:root:Epoch[139] Batch[200] Loss[2.600]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[200]	rmse=0.020277	lr=0.196328
[1,0]<stderr>:INFO:root:Epoch[139] Batch[300] Loss[2.930]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[300]	rmse=0.020381	lr=0.196139
[1,0]<stderr>:INFO:root:Epoch[139] Batch[400] Loss[2.912]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[400]	rmse=0.020360	lr=0.195949
[1,4]<stderr>:[ip-172-31-29-212][[55333,1],4][btl_tcp.c:559:mca_btl_tcp_recv_blocking] [1,4]<stderr>:recv(116) failed: Connection reset by peer (104)[1,4]<stderr>:
[1,0]<stderr>:INFO:root:Epoch[139] Batch[500] Loss[4.357]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[500]	rmse=0.020349	lr=0.195760
[1,0]<stderr>:INFO:root:Epoch[139] Batch[600] Loss[2.826]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[600]	rmse=0.020347	lr=0.195571
[1,0]<stderr>:INFO:root:Epoch[139] Batch[700] Loss[2.855]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[700]	rmse=0.020355	lr=0.195381
[1,0]<stderr>:INFO:root:Epoch[139] Batch[800] Loss[2.868]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[800]	rmse=0.020371	lr=0.195192
[1,0]<stderr>:INFO:root:Epoch[139] Batch[900] Loss[3.388]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[900]	rmse=0.020381	lr=0.195003
[1,0]<stderr>:INFO:root:Epoch[139] Batch[1000] Loss[2.868]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[1000]	rmse=0.020422	lr=0.194814
[1,0]<stderr>:INFO:root:Epoch[139] Batch[1100] Loss[3.929]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[1100]	rmse=0.020428	lr=0.194624
[1,0]<stderr>:INFO:root:Epoch[139] Batch[1200] Loss[3.846]
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[1200]	rmse=0.020438	lr=0.194435
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0] Batch[1251]	Time cost=399.53	Train-metric=0.020436
[1,0]<stderr>:INFO:root:Epoch[139]	Speed: 3206.35 samples/sec
[1,0]<stderr>:INFO:root:Epoch[139] Rank[0]	Validation-accuracy=0.639240	Validation-top_k_accuracy_5=0.859100
[1,0]<stderr>:INFO:root:Epoch[140] Batch[100] Loss[2.889]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[100]	rmse=0.020193	lr=0.194149
[1,0]<stderr>:INFO:root:Epoch[140] Batch[200] Loss[5.259]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[200]	rmse=0.020281	lr=0.193960
[1,0]<stderr>:INFO:root:Epoch[140] Batch[300] Loss[2.833]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[300]	rmse=0.020322	lr=0.193770
[1,0]<stderr>:INFO:root:Epoch[140] Batch[400] Loss[2.608]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[400]	rmse=0.020352	lr=0.193581
[1,0]<stderr>:INFO:root:Epoch[140] Batch[500] Loss[3.658]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[500]	rmse=0.020394	lr=0.193392
[1,0]<stderr>:INFO:root:Epoch[140] Batch[600] Loss[2.970]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[600]	rmse=0.020395	lr=0.193203
[1,0]<stderr>:INFO:root:Epoch[140] Batch[700] Loss[4.625]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[700]	rmse=0.020407	lr=0.193013
[1,0]<stderr>:INFO:root:Epoch[140] Batch[800] Loss[4.667]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[800]	rmse=0.020434	lr=0.192824
[1,0]<stderr>:INFO:root:Epoch[140] Batch[900] Loss[3.544]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[900]	rmse=0.020436	lr=0.192635
[1,0]<stderr>:INFO:root:Epoch[140] Batch[1000] Loss[2.840]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[1000]	rmse=0.020440	lr=0.192446
[1,0]<stderr>:INFO:root:Epoch[140] Batch[1100] Loss[3.120]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[1100]	rmse=0.020426	lr=0.192256
[1,0]<stderr>:INFO:root:Epoch[140] Batch[1200] Loss[2.979]
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[1200]	rmse=0.020444	lr=0.192067
[1,0]<stderr>:INFO:root:Epoch[140] Rank[0] Batch[1251]	Time cost=397.16	Train-metric=0.020455
[1,0]<stderr>:INFO:root:Epoch[140]	Speed: 3225.42 samples/sec
[1,0]<stderr>:INFO:root:Epoch[141] Batch[100] Loss[5.103]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[100]	rmse=0.020342	lr=0.191781
[1,0]<stderr>:INFO:root:Epoch[141] Batch[200] Loss[2.793]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[200]	rmse=0.020334	lr=0.191592
[1,0]<stderr>:INFO:root:Epoch[141] Batch[300] Loss[4.846]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[300]	rmse=0.020366	lr=0.191403
[1,0]<stderr>:INFO:root:Epoch[141] Batch[400] Loss[2.586]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[400]	rmse=0.020359	lr=0.191214
[1,0]<stderr>:INFO:root:Epoch[141] Batch[500] Loss[2.775]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[500]	rmse=0.020385	lr=0.191025
[1,0]<stderr>:INFO:root:Epoch[141] Batch[600] Loss[4.597]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[600]	rmse=0.020405	lr=0.190835
[1,0]<stderr>:INFO:root:Epoch[141] Batch[700] Loss[4.510]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[700]	rmse=0.020398	lr=0.190646
[1,0]<stderr>:INFO:root:Epoch[141] Batch[800] Loss[4.167]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[800]	rmse=0.020400	lr=0.190457
[1,0]<stderr>:INFO:root:Epoch[141] Batch[900] Loss[2.911]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[900]	rmse=0.020405	lr=0.190268
[1,0]<stderr>:INFO:root:Epoch[141] Batch[1000] Loss[4.285]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[1000]	rmse=0.020388	lr=0.190079
[1,0]<stderr>:INFO:root:Epoch[141] Batch[1100] Loss[4.014]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[1100]	rmse=0.020392	lr=0.189890
[1,0]<stderr>:INFO:root:Epoch[141] Batch[1200] Loss[4.957]
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[1200]	rmse=0.020400	lr=0.189700
[1,0]<stderr>:INFO:root:Epoch[141] Rank[0] Batch[1251]	Time cost=399.36	Train-metric=0.020412
[1,0]<stderr>:INFO:root:Epoch[141]	Speed: 3207.65 samples/sec
[1,0]<stderr>:INFO:root:Epoch[142] Batch[100] Loss[2.645]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[100]	rmse=0.020344	lr=0.189415
[1,0]<stderr>:INFO:root:Epoch[142] Batch[200] Loss[3.564]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[200]	rmse=0.020392	lr=0.189226
[1,0]<stderr>:INFO:root:Epoch[142] Batch[300] Loss[5.010]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[300]	rmse=0.020443	lr=0.189037
[1,0]<stderr>:INFO:root:Epoch[142] Batch[400] Loss[2.501]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[400]	rmse=0.020423	lr=0.188848
[1,0]<stderr>:INFO:root:Epoch[142] Batch[500] Loss[3.339]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[500]	rmse=0.020424	lr=0.188659
[1,0]<stderr>:INFO:root:Epoch[142] Batch[600] Loss[2.999]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[600]	rmse=0.020414	lr=0.188469
[1,0]<stderr>:INFO:root:Epoch[142] Batch[700] Loss[2.715]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[700]	rmse=0.020420	lr=0.188280
[1,0]<stderr>:INFO:root:Epoch[142] Batch[800] Loss[4.537]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[800]	rmse=0.020437	lr=0.188091
[1,0]<stderr>:INFO:root:Epoch[142] Batch[900] Loss[2.697]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[900]	rmse=0.020435	lr=0.187902
[1,0]<stderr>:INFO:root:Epoch[142] Batch[1000] Loss[4.291]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[1000]	rmse=0.020435	lr=0.187713
[1,0]<stderr>:INFO:root:Epoch[142] Batch[1100] Loss[4.882]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[1100]	rmse=0.020435	lr=0.187524
[1,0]<stderr>:INFO:root:Epoch[142] Batch[1200] Loss[2.440]
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[1200]	rmse=0.020444	lr=0.187335
[1,0]<stderr>:INFO:root:Epoch[142] Rank[0] Batch[1251]	Time cost=399.43	Train-metric=0.020452
[1,0]<stderr>:INFO:root:Epoch[142]	Speed: 3207.15 samples/sec
[1,0]<stderr>:INFO:root:Epoch[143] Batch[100] Loss[3.846]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[100]	rmse=0.020147	lr=0.187050
[1,0]<stderr>:INFO:root:Epoch[143] Batch[200] Loss[3.235]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[200]	rmse=0.020257	lr=0.186861
[1,0]<stderr>:INFO:root:Epoch[143] Batch[300] Loss[2.778]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[300]	rmse=0.020342	lr=0.186672
[1,0]<stderr>:INFO:root:Epoch[143] Batch[400] Loss[2.635]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[400]	rmse=0.020379	lr=0.186483
[1,0]<stderr>:INFO:root:Epoch[143] Batch[500] Loss[2.792]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[500]	rmse=0.020429	lr=0.186294
[1,0]<stderr>:INFO:root:Epoch[143] Batch[600] Loss[2.742]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[600]	rmse=0.020434	lr=0.186105
[1,0]<stderr>:INFO:root:Epoch[143] Batch[700] Loss[4.504]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[700]	rmse=0.020429	lr=0.185916
[1,0]<stderr>:INFO:root:Epoch[143] Batch[800] Loss[2.823]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[800]	rmse=0.020420	lr=0.185727
[1,0]<stderr>:INFO:root:Epoch[143] Batch[900] Loss[2.761]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[900]	rmse=0.020424	lr=0.185538
[1,0]<stderr>:INFO:root:Epoch[143] Batch[1000] Loss[3.934]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[1000]	rmse=0.020427	lr=0.185350
[1,0]<stderr>:INFO:root:Epoch[143] Batch[1100] Loss[2.908]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[1100]	rmse=0.020430	lr=0.185161
[1,0]<stderr>:INFO:root:Epoch[143] Batch[1200] Loss[2.965]
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[1200]	rmse=0.020432	lr=0.184972
[1,0]<stderr>:INFO:root:Epoch[143] Rank[0] Batch[1251]	Time cost=399.64	Train-metric=0.020433
[1,0]<stderr>:INFO:root:Epoch[143]	Speed: 3205.42 samples/sec
[1,0]<stderr>:INFO:root:Epoch[144] Batch[100] Loss[2.627]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[100]	rmse=0.020345	lr=0.184687
[1,0]<stderr>:INFO:root:Epoch[144] Batch[200] Loss[3.447]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[200]	rmse=0.020390	lr=0.184498
[1,0]<stderr>:INFO:root:Epoch[144] Batch[300] Loss[2.709]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[300]	rmse=0.020384	lr=0.184309
[1,0]<stderr>:INFO:root:Epoch[144] Batch[400] Loss[2.705]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[400]	rmse=0.020373	lr=0.184120
[1,0]<stderr>:INFO:root:Epoch[144] Batch[500] Loss[2.979]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[500]	rmse=0.020363	lr=0.183932
[1,0]<stderr>:INFO:root:Epoch[144] Batch[600] Loss[2.689]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[600]	rmse=0.020352	lr=0.183743
[1,0]<stderr>:INFO:root:Epoch[144] Batch[700] Loss[3.904]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[700]	rmse=0.020363	lr=0.183554
[1,0]<stderr>:INFO:root:Epoch[144] Batch[800] Loss[3.006]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[800]	rmse=0.020370	lr=0.183365
[1,0]<stderr>:INFO:root:Epoch[144] Batch[900] Loss[3.342]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[900]	rmse=0.020359	lr=0.183177
[1,0]<stderr>:INFO:root:Epoch[144] Batch[1000] Loss[4.887]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[1000]	rmse=0.020384	lr=0.182988
[1,0]<stderr>:INFO:root:Epoch[144] Batch[1100] Loss[2.629]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[1100]	rmse=0.020380	lr=0.182799
[1,0]<stderr>:INFO:root:Epoch[144] Batch[1200] Loss[4.004]
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[1200]	rmse=0.020351	lr=0.182611
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0] Batch[1251]	Time cost=399.77	Train-metric=0.020359
[1,0]<stderr>:INFO:root:Epoch[144]	Speed: 3204.37 samples/sec
[1,0]<stderr>:INFO:root:Epoch[144] Rank[0]	Validation-accuracy=0.647600	Validation-top_k_accuracy_5=0.863360
[1,0]<stderr>:INFO:root:Epoch[145] Batch[100] Loss[4.492]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[100]	rmse=0.020272	lr=0.182326
[1,0]<stderr>:INFO:root:Epoch[145] Batch[200] Loss[5.061]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[200]	rmse=0.020304	lr=0.182137
[1,0]<stderr>:INFO:root:Epoch[145] Batch[300] Loss[2.753]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[300]	rmse=0.020321	lr=0.181948
[1,0]<stderr>:INFO:root:Epoch[145] Batch[400] Loss[3.133]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[400]	rmse=0.020300	lr=0.181760
[1,0]<stderr>:INFO:root:Epoch[145] Batch[500] Loss[2.806]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[500]	rmse=0.020321	lr=0.181571
[1,0]<stderr>:INFO:root:Epoch[145] Batch[600] Loss[2.612]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[600]	rmse=0.020328	lr=0.181383
[1,0]<stderr>:INFO:root:Epoch[145] Batch[700] Loss[2.838]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[700]	rmse=0.020356	lr=0.181194
[1,0]<stderr>:INFO:root:Epoch[145] Batch[800] Loss[3.753]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[800]	rmse=0.020371	lr=0.181006
[1,0]<stderr>:INFO:root:Epoch[145] Batch[900] Loss[2.823]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[900]	rmse=0.020377	lr=0.180817
[1,0]<stderr>:INFO:root:Epoch[145] Batch[1000] Loss[4.226]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[1000]	rmse=0.020363	lr=0.180629
[1,0]<stderr>:INFO:root:Epoch[145] Batch[1100] Loss[5.143]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[1100]	rmse=0.020356	lr=0.180440
[1,0]<stderr>:INFO:root:Epoch[145] Batch[1200] Loss[4.959]
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[1200]	rmse=0.020355	lr=0.180252
[1,0]<stderr>:INFO:root:Epoch[145] Rank[0] Batch[1251]	Time cost=398.69	Train-metric=0.020353
[1,0]<stderr>:INFO:root:Epoch[145]	Speed: 3213.07 samples/sec
[1,0]<stderr>:INFO:root:Epoch[146] Batch[100] Loss[2.444]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[100]	rmse=0.020277	lr=0.179967
[1,0]<stderr>:INFO:root:Epoch[146] Batch[200] Loss[3.617]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[200]	rmse=0.020218	lr=0.179779
[1,0]<stderr>:INFO:root:Epoch[146] Batch[300] Loss[2.994]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[300]	rmse=0.020266	lr=0.179590
[1,0]<stderr>:INFO:root:Epoch[146] Batch[400] Loss[2.611]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[400]	rmse=0.020281	lr=0.179402
[1,0]<stderr>:INFO:root:Epoch[146] Batch[500] Loss[3.146]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[500]	rmse=0.020354	lr=0.179214
[1,0]<stderr>:INFO:root:Epoch[146] Batch[600] Loss[4.415]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[600]	rmse=0.020317	lr=0.179025
[1,0]<stderr>:INFO:root:Epoch[146] Batch[700] Loss[2.584]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[700]	rmse=0.020315	lr=0.178837
[1,0]<stderr>:INFO:root:Epoch[146] Batch[800] Loss[2.799]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[800]	rmse=0.020303	lr=0.178649
[1,0]<stderr>:INFO:root:Epoch[146] Batch[900] Loss[2.663]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[900]	rmse=0.020329	lr=0.178460
[1,0]<stderr>:INFO:root:Epoch[146] Batch[1000] Loss[2.642]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[1000]	rmse=0.020328	lr=0.178272
[1,0]<stderr>:INFO:root:Epoch[146] Batch[1100] Loss[2.790]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[1100]	rmse=0.020340	lr=0.178084
[1,0]<stderr>:INFO:root:Epoch[146] Batch[1200] Loss[2.553]
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[1200]	rmse=0.020353	lr=0.177896
[1,0]<stderr>:INFO:root:Epoch[146] Rank[0] Batch[1251]	Time cost=398.84	Train-metric=0.020360
[1,0]<stderr>:INFO:root:Epoch[146]	Speed: 3211.91 samples/sec
[1,0]<stderr>:INFO:root:Epoch[147] Batch[100] Loss[2.677]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[100]	rmse=0.020142	lr=0.177611
[1,0]<stderr>:INFO:root:Epoch[147] Batch[200] Loss[4.892]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[200]	rmse=0.020156	lr=0.177423
[1,0]<stderr>:INFO:root:Epoch[147] Batch[300] Loss[5.189]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[300]	rmse=0.020248	lr=0.177235
[1,0]<stderr>:INFO:root:Epoch[147] Batch[400] Loss[2.791]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[400]	rmse=0.020282	lr=0.177047
[1,0]<stderr>:INFO:root:Epoch[147] Batch[500] Loss[2.705]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[500]	rmse=0.020309	lr=0.176859
[1,0]<stderr>:INFO:root:Epoch[147] Batch[600] Loss[3.816]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[600]	rmse=0.020296	lr=0.176671
[1,0]<stderr>:INFO:root:Epoch[147] Batch[700] Loss[2.620]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[700]	rmse=0.020310	lr=0.176483
[1,0]<stderr>:INFO:root:Epoch[147] Batch[800] Loss[3.965]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[800]	rmse=0.020304	lr=0.176295
[1,0]<stderr>:INFO:root:Epoch[147] Batch[900] Loss[4.286]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[900]	rmse=0.020294	lr=0.176107
[1,0]<stderr>:INFO:root:Epoch[147] Batch[1000] Loss[3.728]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[1000]	rmse=0.020293	lr=0.175919
[1,0]<stderr>:INFO:root:Epoch[147] Batch[1100] Loss[3.108]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[1100]	rmse=0.020298	lr=0.175731
[1,0]<stderr>:INFO:root:Epoch[147] Batch[1200] Loss[3.138]
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[1200]	rmse=0.020309	lr=0.175543
[1,0]<stderr>:INFO:root:Epoch[147] Rank[0] Batch[1251]	Time cost=404.35	Train-metric=0.020319
[1,0]<stderr>:INFO:root:Epoch[147]	Speed: 3168.12 samples/sec
[1,0]<stderr>:INFO:root:Epoch[148] Batch[100] Loss[4.380]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[100]	rmse=0.020142	lr=0.175259
[1,0]<stderr>:INFO:root:Epoch[148] Batch[200] Loss[3.245]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[200]	rmse=0.020235	lr=0.175071
[1,0]<stderr>:INFO:root:Epoch[148] Batch[300] Loss[2.649]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[300]	rmse=0.020277	lr=0.174883
[1,0]<stderr>:INFO:root:Epoch[148] Batch[400] Loss[2.739]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[400]	rmse=0.020258	lr=0.174695
[1,0]<stderr>:INFO:root:Epoch[148] Batch[500] Loss[2.924]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[500]	rmse=0.020248	lr=0.174507
[1,0]<stderr>:INFO:root:Epoch[148] Batch[600] Loss[2.567]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[600]	rmse=0.020286	lr=0.174319
[1,0]<stderr>:INFO:root:Epoch[148] Batch[700] Loss[2.915]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[700]	rmse=0.020305	lr=0.174132
[1,0]<stderr>:INFO:root:Epoch[148] Batch[800] Loss[2.830]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[800]	rmse=0.020298	lr=0.173944
[1,0]<stderr>:INFO:root:Epoch[148] Batch[900] Loss[2.903]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[900]	rmse=0.020299	lr=0.173756
[1,0]<stderr>:INFO:root:Epoch[148] Batch[1000] Loss[5.075]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[1000]	rmse=0.020307	lr=0.173568
[1,0]<stderr>:INFO:root:Epoch[148] Batch[1100] Loss[3.012]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[1100]	rmse=0.020325	lr=0.173381
[1,0]<stderr>:INFO:root:Epoch[148] Batch[1200] Loss[4.857]
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[1200]	rmse=0.020332	lr=0.173193
[1,0]<stderr>:INFO:root:Epoch[148] Rank[0] Batch[1251]	Time cost=401.04	Train-metric=0.020343
[1,0]<stderr>:INFO:root:Epoch[148]	Speed: 3194.26 samples/sec
[1,0]<stderr>:INFO:root:Epoch[149] Batch[100] Loss[2.498]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[100]	rmse=0.020209	lr=0.172910
[1,0]<stderr>:INFO:root:Epoch[149] Batch[200] Loss[4.700]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[200]	rmse=0.020240	lr=0.172722
[1,0]<stderr>:INFO:root:Epoch[149] Batch[300] Loss[3.654]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[300]	rmse=0.020230	lr=0.172534
[1,0]<stderr>:INFO:root:Epoch[149] Batch[400] Loss[2.784]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[400]	rmse=0.020211	lr=0.172347
[1,0]<stderr>:INFO:root:Epoch[149] Batch[500] Loss[5.228]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[500]	rmse=0.020222	lr=0.172159
[1,0]<stderr>:INFO:root:Epoch[149] Batch[600] Loss[2.619]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[600]	rmse=0.020225	lr=0.171972
[1,0]<stderr>:INFO:root:Epoch[149] Batch[700] Loss[2.656]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[700]	rmse=0.020249	lr=0.171784
[1,0]<stderr>:INFO:root:Epoch[149] Batch[800] Loss[4.053]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[800]	rmse=0.020284	lr=0.171597
[1,0]<stderr>:INFO:root:Epoch[149] Batch[900] Loss[4.806]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[900]	rmse=0.020303	lr=0.171409
[1,0]<stderr>:INFO:root:Epoch[149] Batch[1000] Loss[3.109]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[1000]	rmse=0.020323	lr=0.171222
[1,0]<stderr>:INFO:root:Epoch[149] Batch[1100] Loss[4.427]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[1100]	rmse=0.020337	lr=0.171035
[1,0]<stderr>:INFO:root:Epoch[149] Batch[1200] Loss[3.086]
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[1200]	rmse=0.020332	lr=0.170847
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0] Batch[1251]	Time cost=402.03	Train-metric=0.020327
[1,0]<stderr>:INFO:root:Epoch[149]	Speed: 3186.40 samples/sec
[1,0]<stderr>:INFO:root:Epoch[149] Rank[0]	Validation-accuracy=0.648860	Validation-top_k_accuracy_5=0.866020
[1,0]<stderr>:INFO:root:Epoch[150] Batch[100] Loss[3.380]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[100]	rmse=0.019989	lr=0.170564
[1,0]<stderr>:INFO:root:Epoch[150] Batch[200] Loss[2.980]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[200]	rmse=0.020178	lr=0.170377
[1,0]<stderr>:INFO:root:Epoch[150] Batch[300] Loss[4.684]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[300]	rmse=0.020166	lr=0.170190
[1,0]<stderr>:INFO:root:Epoch[150] Batch[400] Loss[2.772]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[400]	rmse=0.020202	lr=0.170003
[1,0]<stderr>:INFO:root:Epoch[150] Batch[500] Loss[3.681]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[500]	rmse=0.020268	lr=0.169815
[1,0]<stderr>:INFO:root:Epoch[150] Batch[600] Loss[4.110]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[600]	rmse=0.020271	lr=0.169628
[1,0]<stderr>:INFO:root:Epoch[150] Batch[700] Loss[2.985]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[700]	rmse=0.020273	lr=0.169441
[1,0]<stderr>:INFO:root:Epoch[150] Batch[800] Loss[3.607]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[800]	rmse=0.020302	lr=0.169254
[1,0]<stderr>:INFO:root:Epoch[150] Batch[900] Loss[5.175]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[900]	rmse=0.020315	lr=0.169067
[1,0]<stderr>:INFO:root:Epoch[150] Batch[1000] Loss[2.833]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[1000]	rmse=0.020312	lr=0.168880
[1,0]<stderr>:INFO:root:Epoch[150] Batch[1100] Loss[3.437]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[1100]	rmse=0.020318	lr=0.168693
[1,0]<stderr>:INFO:root:Epoch[150] Batch[1200] Loss[2.461]
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[1200]	rmse=0.020323	lr=0.168506
[1,0]<stderr>:INFO:root:Epoch[150] Rank[0] Batch[1251]	Time cost=401.40	Train-metric=0.020331
[1,0]<stderr>:INFO:root:Epoch[150]	Speed: 3191.40 samples/sec
[1,0]<stderr>:INFO:root:Epoch[151] Batch[100] Loss[4.970]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[100]	rmse=0.020214	lr=0.168223
[1,0]<stderr>:INFO:root:Epoch[151] Batch[200] Loss[4.643]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[200]	rmse=0.020201	lr=0.168036
[1,0]<stderr>:INFO:root:Epoch[151] Batch[300] Loss[5.151]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[300]	rmse=0.020212	lr=0.167849
[1,0]<stderr>:INFO:root:Epoch[151] Batch[400] Loss[3.576]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[400]	rmse=0.020269	lr=0.167662
[1,0]<stderr>:INFO:root:Epoch[151] Batch[500] Loss[2.898]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[500]	rmse=0.020284	lr=0.167475
[1,0]<stderr>:INFO:root:Epoch[151] Batch[600] Loss[2.742]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[600]	rmse=0.020285	lr=0.167289
[1,0]<stderr>:INFO:root:Epoch[151] Batch[700] Loss[5.054]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[700]	rmse=0.020308	lr=0.167102
[1,0]<stderr>:INFO:root:Epoch[151] Batch[800] Loss[2.883]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[800]	rmse=0.020311	lr=0.166915
[1,0]<stderr>:INFO:root:Epoch[151] Batch[900] Loss[4.283]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[900]	rmse=0.020324	lr=0.166728
[1,0]<stderr>:INFO:root:Epoch[151] Batch[1000] Loss[4.464]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[1000]	rmse=0.020329	lr=0.166542
[1,0]<stderr>:INFO:root:Epoch[151] Batch[1100] Loss[2.727]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[1100]	rmse=0.020341	lr=0.166355
[1,0]<stderr>:INFO:root:Epoch[151] Batch[1200] Loss[2.607]
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[1200]	rmse=0.020330	lr=0.166168
[1,0]<stderr>:INFO:root:Epoch[151] Rank[0] Batch[1251]	Time cost=399.24	Train-metric=0.020335
[1,0]<stderr>:INFO:root:Epoch[151]	Speed: 3208.66 samples/sec
[1,0]<stderr>:INFO:root:Epoch[152] Batch[100] Loss[4.715]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[100]	rmse=0.020245	lr=0.165886
[1,0]<stderr>:INFO:root:Epoch[152] Batch[200] Loss[3.175]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[200]	rmse=0.020204	lr=0.165700
[1,0]<stderr>:INFO:root:Epoch[152] Batch[300] Loss[2.904]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[300]	rmse=0.020204	lr=0.165513
[1,0]<stderr>:INFO:root:Epoch[152] Batch[400] Loss[2.510]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[400]	rmse=0.020231	lr=0.165327
[1,0]<stderr>:INFO:root:Epoch[152] Batch[500] Loss[3.408]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[500]	rmse=0.020241	lr=0.165140
[1,0]<stderr>:INFO:root:Epoch[152] Batch[600] Loss[4.702]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[600]	rmse=0.020215	lr=0.164954
[1,0]<stderr>:INFO:root:Epoch[152] Batch[700] Loss[2.784]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[700]	rmse=0.020230	lr=0.164767
[1,0]<stderr>:INFO:root:Epoch[152] Batch[800] Loss[2.555]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[800]	rmse=0.020250	lr=0.164581
[1,0]<stderr>:INFO:root:Epoch[152] Batch[900] Loss[3.142]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[900]	rmse=0.020240	lr=0.164395
[1,0]<stderr>:INFO:root:Epoch[152] Batch[1000] Loss[2.596]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[1000]	rmse=0.020240	lr=0.164208
[1,0]<stderr>:INFO:root:Epoch[152] Batch[1100] Loss[3.213]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[1100]	rmse=0.020240	lr=0.164022
[1,0]<stderr>:INFO:root:Epoch[152] Batch[1200] Loss[2.945]
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[1200]	rmse=0.020240	lr=0.163836
[1,0]<stderr>:INFO:root:Epoch[152] Rank[0] Batch[1251]	Time cost=398.98	Train-metric=0.020236
[1,0]<stderr>:INFO:root:Epoch[152]	Speed: 3210.73 samples/sec
[1,0]<stderr>:INFO:root:Epoch[153] Batch[100] Loss[4.846]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[100]	rmse=0.020242	lr=0.163554
[1,0]<stderr>:INFO:root:Epoch[153] Batch[200] Loss[2.611]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[200]	rmse=0.020179	lr=0.163368
[1,0]<stderr>:INFO:root:Epoch[153] Batch[300] Loss[3.374]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[300]	rmse=0.020206	lr=0.163182
[1,0]<stderr>:INFO:root:Epoch[153] Batch[400] Loss[3.784]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[400]	rmse=0.020180	lr=0.162996
[1,0]<stderr>:INFO:root:Epoch[153] Batch[500] Loss[5.000]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[500]	rmse=0.020181	lr=0.162810
[1,0]<stderr>:INFO:root:Epoch[153] Batch[600] Loss[3.536]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[600]	rmse=0.020179	lr=0.162624
[1,0]<stderr>:INFO:root:Epoch[153] Batch[700] Loss[3.961]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[700]	rmse=0.020173	lr=0.162438
[1,0]<stderr>:INFO:root:Epoch[153] Batch[800] Loss[2.787]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[800]	rmse=0.020191	lr=0.162252
[1,0]<stderr>:INFO:root:Epoch[153] Batch[900] Loss[2.754]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[900]	rmse=0.020214	lr=0.162066
[1,0]<stderr>:INFO:root:Epoch[153] Batch[1000] Loss[5.287]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[1000]	rmse=0.020234	lr=0.161880
[1,0]<stderr>:INFO:root:Epoch[153] Batch[1100] Loss[4.343]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[1100]	rmse=0.020245	lr=0.161694
[1,0]<stderr>:INFO:root:Epoch[153] Batch[1200] Loss[4.877]
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[1200]	rmse=0.020252	lr=0.161508
[1,0]<stderr>:INFO:root:Epoch[153] Rank[0] Batch[1251]	Time cost=399.32	Train-metric=0.020254
[1,0]<stderr>:INFO:root:Epoch[153]	Speed: 3208.00 samples/sec
[1,0]<stderr>:INFO:root:Epoch[154] Batch[100] Loss[3.421]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[100]	rmse=0.020001	lr=0.161228
[1,0]<stderr>:INFO:root:Epoch[154] Batch[200] Loss[2.706]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[200]	rmse=0.020203	lr=0.161042
[1,0]<stderr>:INFO:root:Epoch[154] Batch[300] Loss[2.606]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[300]	rmse=0.020189	lr=0.160856
[1,0]<stderr>:INFO:root:Epoch[154] Batch[400] Loss[5.021]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[400]	rmse=0.020212	lr=0.160670
[1,0]<stderr>:INFO:root:Epoch[154] Batch[500] Loss[4.257]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[500]	rmse=0.020177	lr=0.160485
[1,0]<stderr>:INFO:root:Epoch[154] Batch[600] Loss[4.493]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[600]	rmse=0.020207	lr=0.160299
[1,0]<stderr>:INFO:root:Epoch[154] Batch[700] Loss[2.588]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[700]	rmse=0.020204	lr=0.160114
[1,0]<stderr>:INFO:root:Epoch[154] Batch[800] Loss[2.770]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[800]	rmse=0.020205	lr=0.159928
[1,0]<stderr>:INFO:root:Epoch[154] Batch[900] Loss[4.775]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[900]	rmse=0.020226	lr=0.159742
[1,0]<stderr>:INFO:root:Epoch[154] Batch[1000] Loss[2.707]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[1000]	rmse=0.020230	lr=0.159557
[1,0]<stderr>:INFO:root:Epoch[154] Batch[1100] Loss[4.969]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[1100]	rmse=0.020265	lr=0.159372
[1,0]<stderr>:INFO:root:Epoch[154] Batch[1200] Loss[2.751]
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[1200]	rmse=0.020283	lr=0.159186
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0] Batch[1251]	Time cost=399.33	Train-metric=0.020274
[1,0]<stderr>:INFO:root:Epoch[154]	Speed: 3207.91 samples/sec
[1,0]<stderr>:INFO:root:Epoch[154] Rank[0]	Validation-accuracy=0.647400	Validation-top_k_accuracy_5=0.865260
[1,0]<stderr>:INFO:root:Epoch[155] Batch[100] Loss[2.531]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[100]	rmse=0.020182	lr=0.158906
[1,0]<stderr>:INFO:root:Epoch[155] Batch[200] Loss[2.765]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[200]	rmse=0.020178	lr=0.158721
[1,0]<stderr>:INFO:root:Epoch[155] Batch[300] Loss[4.566]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[300]	rmse=0.020177	lr=0.158536
[1,0]<stderr>:INFO:root:Epoch[155] Batch[400] Loss[2.799]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[400]	rmse=0.020167	lr=0.158350
[1,0]<stderr>:INFO:root:Epoch[155] Batch[500] Loss[4.911]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[500]	rmse=0.020195	lr=0.158165
[1,0]<stderr>:INFO:root:Epoch[155] Batch[600] Loss[3.744]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[600]	rmse=0.020146	lr=0.157980
[1,0]<stderr>:INFO:root:Epoch[155] Batch[700] Loss[4.707]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[700]	rmse=0.020162	lr=0.157795
[1,0]<stderr>:INFO:root:Epoch[155] Batch[800] Loss[4.942]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[800]	rmse=0.020188	lr=0.157610
[1,0]<stderr>:INFO:root:Epoch[155] Batch[900] Loss[2.615]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[900]	rmse=0.020182	lr=0.157425
[1,0]<stderr>:INFO:root:Epoch[155] Batch[1000] Loss[2.935]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[1000]	rmse=0.020201	lr=0.157240
[1,0]<stderr>:INFO:root:Epoch[155] Batch[1100] Loss[2.857]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[1100]	rmse=0.020217	lr=0.157055
[1,0]<stderr>:INFO:root:Epoch[155] Batch[1200] Loss[4.613]
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[1200]	rmse=0.020218	lr=0.156870
[1,0]<stderr>:INFO:root:Epoch[155] Rank[0] Batch[1251]	Time cost=398.96	Train-metric=0.020229
[1,0]<stderr>:INFO:root:Epoch[155]	Speed: 3210.94 samples/sec
[1,0]<stderr>:INFO:root:Epoch[156] Batch[100] Loss[2.830]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[100]	rmse=0.020041	lr=0.156591
[1,0]<stderr>:INFO:root:Epoch[156] Batch[200] Loss[2.871]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[200]	rmse=0.020121	lr=0.156406
[1,0]<stderr>:INFO:root:Epoch[156] Batch[300] Loss[2.748]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[300]	rmse=0.020118	lr=0.156221
[1,0]<stderr>:INFO:root:Epoch[156] Batch[400] Loss[2.543]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[400]	rmse=0.020134	lr=0.156036
[1,0]<stderr>:INFO:root:Epoch[156] Batch[500] Loss[3.118]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[500]	rmse=0.020131	lr=0.155851
[1,0]<stderr>:INFO:root:Epoch[156] Batch[600] Loss[4.967]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[600]	rmse=0.020155	lr=0.155667
[1,0]<stderr>:INFO:root:Epoch[156] Batch[700] Loss[2.584]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[700]	rmse=0.020166	lr=0.155482
[1,0]<stderr>:INFO:root:Epoch[156] Batch[800] Loss[2.789]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[800]	rmse=0.020146	lr=0.155297
[1,0]<stderr>:INFO:root:Epoch[156] Batch[900] Loss[3.797]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[900]	rmse=0.020151	lr=0.155113
[1,0]<stderr>:INFO:root:Epoch[156] Batch[1000] Loss[2.791]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[1000]	rmse=0.020164	lr=0.154928
[1,0]<stderr>:INFO:root:Epoch[156] Batch[1100] Loss[3.248]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[1100]	rmse=0.020176	lr=0.154744
[1,0]<stderr>:INFO:root:Epoch[156] Batch[1200] Loss[3.028]
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[1200]	rmse=0.020178	lr=0.154559
[1,0]<stderr>:INFO:root:Epoch[156] Rank[0] Batch[1251]	Time cost=399.17	Train-metric=0.020187
[1,0]<stderr>:INFO:root:Epoch[156]	Speed: 3209.22 samples/sec
[1,0]<stderr>:INFO:root:Epoch[157] Batch[100] Loss[2.764]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[100]	rmse=0.020223	lr=0.154281
[1,0]<stderr>:INFO:root:Epoch[157] Batch[200] Loss[3.041]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[200]	rmse=0.020185	lr=0.154097
[1,0]<stderr>:INFO:root:Epoch[157] Batch[300] Loss[2.458]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[300]	rmse=0.020159	lr=0.153912
[1,0]<stderr>:INFO:root:Epoch[157] Batch[400] Loss[3.169]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[400]	rmse=0.020168	lr=0.153728
[1,0]<stderr>:INFO:root:Epoch[157] Batch[500] Loss[4.152]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[500]	rmse=0.020184	lr=0.153544
[1,0]<stderr>:INFO:root:Epoch[157] Batch[600] Loss[2.871]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[600]	rmse=0.020166	lr=0.153360
[1,0]<stderr>:INFO:root:Epoch[157] Batch[700] Loss[2.798]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[700]	rmse=0.020160	lr=0.153176
[1,0]<stderr>:INFO:root:Epoch[157] Batch[800] Loss[2.650]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[800]	rmse=0.020161	lr=0.152991
[1,0]<stderr>:INFO:root:Epoch[157] Batch[900] Loss[2.466]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[900]	rmse=0.020169	lr=0.152807
[1,0]<stderr>:INFO:root:Epoch[157] Batch[1000] Loss[2.941]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[1000]	rmse=0.020183	lr=0.152623
[1,0]<stderr>:INFO:root:Epoch[157] Batch[1100] Loss[2.752]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[1100]	rmse=0.020177	lr=0.152439
[1,0]<stderr>:INFO:root:Epoch[157] Batch[1200] Loss[2.735]
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[1200]	rmse=0.020174	lr=0.152256
[1,0]<stderr>:INFO:root:Epoch[157] Rank[0] Batch[1251]	Time cost=400.26	Train-metric=0.020179
[1,0]<stderr>:INFO:root:Epoch[157]	Speed: 3200.52 samples/sec
[1,0]<stderr>:INFO:root:Epoch[158] Batch[100] Loss[4.900]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[100]	rmse=0.020045	lr=0.151978
[1,0]<stderr>:INFO:root:Epoch[158] Batch[200] Loss[4.489]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[200]	rmse=0.020108	lr=0.151794
[1,0]<stderr>:INFO:root:Epoch[158] Batch[300] Loss[2.704]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[300]	rmse=0.020148	lr=0.151610
[1,0]<stderr>:INFO:root:Epoch[158] Batch[400] Loss[3.159]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[400]	rmse=0.020101	lr=0.151427
[1,0]<stderr>:INFO:root:Epoch[158] Batch[500] Loss[4.598]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[500]	rmse=0.020079	lr=0.151243
[1,0]<stderr>:INFO:root:Epoch[158] Batch[600] Loss[2.579]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[600]	rmse=0.020089	lr=0.151059
[1,0]<stderr>:INFO:root:Epoch[158] Batch[700] Loss[2.494]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[700]	rmse=0.020096	lr=0.150876
[1,0]<stderr>:INFO:root:Epoch[158] Batch[800] Loss[3.339]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[800]	rmse=0.020093	lr=0.150692
[1,0]<stderr>:INFO:root:Epoch[158] Batch[900] Loss[2.812]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[900]	rmse=0.020103	lr=0.150509
[1,0]<stderr>:INFO:root:Epoch[158] Batch[1000] Loss[2.648]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[1000]	rmse=0.020102	lr=0.150325
[1,0]<stderr>:INFO:root:Epoch[158] Batch[1100] Loss[2.586]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[1100]	rmse=0.020120	lr=0.150142
[1,0]<stderr>:INFO:root:Epoch[158] Batch[1200] Loss[2.886]
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[1200]	rmse=0.020133	lr=0.149958
[1,0]<stderr>:INFO:root:Epoch[158] Rank[0] Batch[1251]	Time cost=400.20	Train-metric=0.020140
[1,0]<stderr>:INFO:root:Epoch[158]	Speed: 3200.92 samples/sec
[1,0]<stderr>:INFO:root:Epoch[159] Batch[100] Loss[3.097]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[100]	rmse=0.020010	lr=0.149682
[1,0]<stderr>:INFO:root:Epoch[159] Batch[200] Loss[2.519]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[200]	rmse=0.020051	lr=0.149498
[1,0]<stderr>:INFO:root:Epoch[159] Batch[300] Loss[2.713]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[300]	rmse=0.020108	lr=0.149315
[1,0]<stderr>:INFO:root:Epoch[159] Batch[400] Loss[3.084]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[400]	rmse=0.020080	lr=0.149132
[1,0]<stderr>:INFO:root:Epoch[159] Batch[500] Loss[2.517]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[500]	rmse=0.020106	lr=0.148949
[1,0]<stderr>:INFO:root:Epoch[159] Batch[600] Loss[2.895]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[600]	rmse=0.020095	lr=0.148766
[1,0]<stderr>:INFO:root:Epoch[159] Batch[700] Loss[4.423]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[700]	rmse=0.020100	lr=0.148583
[1,0]<stderr>:INFO:root:Epoch[159] Batch[800] Loss[3.050]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[800]	rmse=0.020108	lr=0.148400
[1,0]<stderr>:INFO:root:Epoch[159] Batch[900] Loss[2.825]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[900]	rmse=0.020124	lr=0.148217
[1,0]<stderr>:INFO:root:Epoch[159] Batch[1000] Loss[4.368]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[1000]	rmse=0.020116	lr=0.148034
[1,0]<stderr>:INFO:root:Epoch[159] Batch[1100] Loss[2.845]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[1100]	rmse=0.020123	lr=0.147851
[1,0]<stderr>:INFO:root:Epoch[159] Batch[1200] Loss[2.756]
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[1200]	rmse=0.020126	lr=0.147668
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0] Batch[1251]	Time cost=398.98	Train-metric=0.020131
[1,0]<stderr>:INFO:root:Epoch[159]	Speed: 3210.72 samples/sec
[1,0]<stderr>:INFO:root:Epoch[159] Rank[0]	Validation-accuracy=0.653660	Validation-top_k_accuracy_5=0.870760
[1,0]<stderr>:INFO:root:Epoch[160] Batch[100] Loss[5.206]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[100]	rmse=0.020133	lr=0.147392
[1,0]<stderr>:INFO:root:Epoch[160] Batch[200] Loss[2.722]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[200]	rmse=0.020021	lr=0.147209
[1,0]<stderr>:INFO:root:Epoch[160] Batch[300] Loss[2.637]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[300]	rmse=0.020028	lr=0.147027
[1,0]<stderr>:INFO:root:Epoch[160] Batch[400] Loss[2.926]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[400]	rmse=0.020039	lr=0.146844
[1,0]<stderr>:INFO:root:Epoch[160] Batch[500] Loss[2.895]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[500]	rmse=0.020050	lr=0.146662
[1,0]<stderr>:INFO:root:Epoch[160] Batch[600] Loss[4.385]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[600]	rmse=0.020053	lr=0.146479
[1,0]<stderr>:INFO:root:Epoch[160] Batch[700] Loss[5.230]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[700]	rmse=0.020078	lr=0.146297
[1,0]<stderr>:INFO:root:Epoch[160] Batch[800] Loss[2.920]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[800]	rmse=0.020109	lr=0.146114
[1,0]<stderr>:INFO:root:Epoch[160] Batch[900] Loss[2.813]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[900]	rmse=0.020128	lr=0.145932
[1,0]<stderr>:INFO:root:Epoch[160] Batch[1000] Loss[2.538]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[1000]	rmse=0.020160	lr=0.145750
[1,0]<stderr>:INFO:root:Epoch[160] Batch[1100] Loss[3.816]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[1100]	rmse=0.020159	lr=0.145567
[1,0]<stderr>:INFO:root:Epoch[160] Batch[1200] Loss[3.140]
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[1200]	rmse=0.020171	lr=0.145385
[1,0]<stderr>:INFO:root:Epoch[160] Rank[0] Batch[1251]	Time cost=397.75	Train-metric=0.020178
[1,0]<stderr>:INFO:root:Epoch[160]	Speed: 3220.65 samples/sec
[1,0]<stderr>:INFO:root:Epoch[161] Batch[100] Loss[2.658]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[100]	rmse=0.020097	lr=0.145110
[1,0]<stderr>:INFO:root:Epoch[161] Batch[200] Loss[2.916]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[200]	rmse=0.020108	lr=0.144928
[1,0]<stderr>:INFO:root:Epoch[161] Batch[300] Loss[4.910]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[300]	rmse=0.020080	lr=0.144746
[1,0]<stderr>:INFO:root:Epoch[161] Batch[400] Loss[3.146]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[400]	rmse=0.020079	lr=0.144564
[1,0]<stderr>:INFO:root:Epoch[161] Batch[500] Loss[4.639]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[500]	rmse=0.020101	lr=0.144382
[1,0]<stderr>:INFO:root:Epoch[161] Batch[600] Loss[3.110]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[600]	rmse=0.020111	lr=0.144200
[1,0]<stderr>:INFO:root:Epoch[161] Batch[700] Loss[4.997]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[700]	rmse=0.020096	lr=0.144018
[1,0]<stderr>:INFO:root:Epoch[161] Batch[800] Loss[2.776]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[800]	rmse=0.020076	lr=0.143837
[1,0]<stderr>:INFO:root:Epoch[161] Batch[900] Loss[2.884]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[900]	rmse=0.020096	lr=0.143655
[1,0]<stderr>:INFO:root:Epoch[161] Batch[1000] Loss[2.809]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[1000]	rmse=0.020105	lr=0.143473
[1,0]<stderr>:INFO:root:Epoch[161] Batch[1100] Loss[2.484]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[1100]	rmse=0.020108	lr=0.143292
[1,0]<stderr>:INFO:root:Epoch[161] Batch[1200] Loss[5.201]
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[1200]	rmse=0.020116	lr=0.143110
[1,0]<stderr>:INFO:root:Epoch[161] Rank[0] Batch[1251]	Time cost=399.43	Train-metric=0.020115
[1,0]<stderr>:INFO:root:Epoch[161]	Speed: 3207.13 samples/sec
[1,0]<stderr>:INFO:root:Epoch[162] Batch[100] Loss[3.604]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[100]	rmse=0.020027	lr=0.142836
[1,0]<stderr>:INFO:root:Epoch[162] Batch[200] Loss[4.960]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[200]	rmse=0.019963	lr=0.142655
[1,0]<stderr>:INFO:root:Epoch[162] Batch[300] Loss[2.786]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[300]	rmse=0.019980	lr=0.142473
[1,0]<stderr>:INFO:root:Epoch[162] Batch[400] Loss[3.019]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[400]	rmse=0.019966	lr=0.142292
[1,0]<stderr>:INFO:root:Epoch[162] Batch[500] Loss[2.480]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[500]	rmse=0.020010	lr=0.142110
[1,0]<stderr>:INFO:root:Epoch[162] Batch[600] Loss[2.714]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[600]	rmse=0.019992	lr=0.141929
[1,0]<stderr>:INFO:root:Epoch[162] Batch[700] Loss[4.838]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[700]	rmse=0.019961	lr=0.141748
[1,0]<stderr>:INFO:root:Epoch[162] Batch[800] Loss[2.544]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[800]	rmse=0.019962	lr=0.141567
[1,0]<stderr>:INFO:root:Epoch[162] Batch[900] Loss[2.996]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[900]	rmse=0.020004	lr=0.141386
[1,0]<stderr>:INFO:root:Epoch[162] Batch[1000] Loss[3.081]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[1000]	rmse=0.020010	lr=0.141205
[1,0]<stderr>:INFO:root:Epoch[162] Batch[1100] Loss[2.542]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[1100]	rmse=0.020021	lr=0.141024
[1,0]<stderr>:INFO:root:Epoch[162] Batch[1200] Loss[2.802]
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[1200]	rmse=0.020044	lr=0.140843
[1,0]<stderr>:INFO:root:Epoch[162] Rank[0] Batch[1251]	Time cost=398.89	Train-metric=0.020049
[1,0]<stderr>:INFO:root:Epoch[162]	Speed: 3211.51 samples/sec
[1,0]<stderr>:INFO:root:Epoch[163] Batch[100] Loss[2.805]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[100]	rmse=0.019920	lr=0.140570
[1,0]<stderr>:INFO:root:Epoch[163] Batch[200] Loss[2.693]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[200]	rmse=0.019875	lr=0.140389
[1,0]<stderr>:INFO:root:Epoch[163] Batch[300] Loss[2.679]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[300]	rmse=0.019882	lr=0.140208
[1,0]<stderr>:INFO:root:Epoch[163] Batch[400] Loss[2.899]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[400]	rmse=0.019939	lr=0.140028
[1,0]<stderr>:INFO:root:Epoch[163] Batch[500] Loss[3.141]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[500]	rmse=0.019960	lr=0.139847
[1,0]<stderr>:INFO:root:Epoch[163] Batch[600] Loss[4.366]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[600]	rmse=0.019968	lr=0.139666
[1,0]<stderr>:INFO:root:Epoch[163] Batch[700] Loss[2.704]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[700]	rmse=0.019983	lr=0.139486
[1,0]<stderr>:INFO:root:Epoch[163] Batch[800] Loss[2.648]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[800]	rmse=0.020015	lr=0.139305
[1,0]<stderr>:INFO:root:Epoch[163] Batch[900] Loss[4.365]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[900]	rmse=0.020022	lr=0.139125
[1,0]<stderr>:INFO:root:Epoch[163] Batch[1000] Loss[2.640]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[1000]	rmse=0.020033	lr=0.138945
[1,0]<stderr>:INFO:root:Epoch[163] Batch[1100] Loss[4.444]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[1100]	rmse=0.020058	lr=0.138764
[1,0]<stderr>:INFO:root:Epoch[163] Batch[1200] Loss[3.987]
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[1200]	rmse=0.020069	lr=0.138584
[1,0]<stderr>:INFO:root:Epoch[163] Rank[0] Batch[1251]	Time cost=399.32	Train-metric=0.020068
[1,0]<stderr>:INFO:root:Epoch[163]	Speed: 3208.00 samples/sec
[1,0]<stderr>:INFO:root:Epoch[164] Batch[100] Loss[3.053]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[100]	rmse=0.019857	lr=0.138312
[1,0]<stderr>:INFO:root:Epoch[164] Batch[200] Loss[2.551]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[200]	rmse=0.019955	lr=0.138132
[1,0]<stderr>:INFO:root:Epoch[164] Batch[300] Loss[2.778]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[300]	rmse=0.019992	lr=0.137952
[1,0]<stderr>:INFO:root:Epoch[164] Batch[400] Loss[3.667]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[400]	rmse=0.020021	lr=0.137772
[1,0]<stderr>:INFO:root:Epoch[164] Batch[500] Loss[5.164]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[500]	rmse=0.020007	lr=0.137592
[1,0]<stderr>:INFO:root:Epoch[164] Batch[600] Loss[4.637]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[600]	rmse=0.020040	lr=0.137412
[1,0]<stderr>:INFO:root:Epoch[164] Batch[700] Loss[2.486]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[700]	rmse=0.020052	lr=0.137232
[1,0]<stderr>:INFO:root:Epoch[164] Batch[800] Loss[4.439]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[800]	rmse=0.020049	lr=0.137052
[1,0]<stderr>:INFO:root:Epoch[164] Batch[900] Loss[2.562]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[900]	rmse=0.020044	lr=0.136873
[1,0]<stderr>:INFO:root:Epoch[164] Batch[1000] Loss[3.832]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[1000]	rmse=0.020043	lr=0.136693
[1,0]<stderr>:INFO:root:Epoch[164] Batch[1100] Loss[2.520]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[1100]	rmse=0.020043	lr=0.136513
[1,0]<stderr>:INFO:root:Epoch[164] Batch[1200] Loss[2.528]
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[1200]	rmse=0.020064	lr=0.136334
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0] Batch[1251]	Time cost=399.21	Train-metric=0.020063
[1,0]<stderr>:INFO:root:Epoch[164]	Speed: 3208.91 samples/sec
[1,0]<stderr>:INFO:root:Epoch[164] Rank[0]	Validation-accuracy=0.666320	Validation-top_k_accuracy_5=0.876800
[1,0]<stderr>:INFO:root:Epoch[165] Batch[100] Loss[2.770]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[100]	rmse=0.020084	lr=0.136063
[1,0]<stderr>:INFO:root:Epoch[165] Batch[200] Loss[2.843]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[200]	rmse=0.020120	lr=0.135883
[1,0]<stderr>:INFO:root:Epoch[165] Batch[300] Loss[4.839]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[300]	rmse=0.020024	lr=0.135704
[1,0]<stderr>:INFO:root:Epoch[165] Batch[400] Loss[2.667]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[400]	rmse=0.020020	lr=0.135525
[1,0]<stderr>:INFO:root:Epoch[165] Batch[500] Loss[2.839]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[500]	rmse=0.020038	lr=0.135345
[1,0]<stderr>:INFO:root:Epoch[165] Batch[600] Loss[2.691]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[600]	rmse=0.020049	lr=0.135166
[1,0]<stderr>:INFO:root:Epoch[165] Batch[700] Loss[4.706]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[700]	rmse=0.020083	lr=0.134987
[1,0]<stderr>:INFO:root:Epoch[165] Batch[800] Loss[2.962]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[800]	rmse=0.020082	lr=0.134808
[1,0]<stderr>:INFO:root:Epoch[165] Batch[900] Loss[2.756]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[900]	rmse=0.020063	lr=0.134629
[1,0]<stderr>:INFO:root:Epoch[165] Batch[1000] Loss[2.771]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[1000]	rmse=0.020064	lr=0.134450
[1,0]<stderr>:INFO:root:Epoch[165] Batch[1100] Loss[3.368]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[1100]	rmse=0.020059	lr=0.134271
[1,0]<stderr>:INFO:root:Epoch[165] Batch[1200] Loss[3.129]
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[1200]	rmse=0.020062	lr=0.134092
[1,0]<stderr>:INFO:root:Epoch[165] Rank[0] Batch[1251]	Time cost=398.86	Train-metric=0.020067
[1,0]<stderr>:INFO:root:Epoch[165]	Speed: 3211.72 samples/sec
[1,0]<stderr>:INFO:root:Epoch[166] Batch[100] Loss[2.910]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[100]	rmse=0.019862	lr=0.133822
[1,0]<stderr>:INFO:root:Epoch[166] Batch[200] Loss[2.422]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[200]	rmse=0.019895	lr=0.133644
[1,0]<stderr>:INFO:root:Epoch[166] Batch[300] Loss[3.838]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[300]	rmse=0.019864	lr=0.133465
[1,0]<stderr>:INFO:root:Epoch[166] Batch[400] Loss[2.832]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[400]	rmse=0.019942	lr=0.133287
[1,0]<stderr>:INFO:root:Epoch[166] Batch[500] Loss[2.518]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[500]	rmse=0.019952	lr=0.133108
[1,0]<stderr>:INFO:root:Epoch[166] Batch[600] Loss[2.388]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[600]	rmse=0.019962	lr=0.132930
[1,0]<stderr>:INFO:root:Epoch[166] Batch[700] Loss[2.677]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[700]	rmse=0.019963	lr=0.132751
[1,0]<stderr>:INFO:root:Epoch[166] Batch[800] Loss[2.972]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[800]	rmse=0.019965	lr=0.132573
[1,0]<stderr>:INFO:root:Epoch[166] Batch[900] Loss[4.061]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[900]	rmse=0.019965	lr=0.132395
[1,0]<stderr>:INFO:root:Epoch[166] Batch[1000] Loss[2.644]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[1000]	rmse=0.019983	lr=0.132216
[1,0]<stderr>:INFO:root:Epoch[166] Batch[1100] Loss[3.418]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[1100]	rmse=0.019991	lr=0.132038
[1,0]<stderr>:INFO:root:Epoch[166] Batch[1200] Loss[5.146]
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[1200]	rmse=0.019994	lr=0.131860
[1,0]<stderr>:INFO:root:Epoch[166] Rank[0] Batch[1251]	Time cost=399.41	Train-metric=0.019988
[1,0]<stderr>:INFO:root:Epoch[166]	Speed: 3207.33 samples/sec
[1,0]<stderr>:INFO:root:Epoch[167] Batch[100] Loss[3.686]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[100]	rmse=0.020065	lr=0.131591
[1,0]<stderr>:INFO:root:Epoch[167] Batch[200] Loss[2.567]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[200]	rmse=0.019956	lr=0.131413
[1,0]<stderr>:INFO:root:Epoch[167] Batch[300] Loss[2.833]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[300]	rmse=0.019973	lr=0.131236
[1,0]<stderr>:INFO:root:Epoch[167] Batch[400] Loss[2.809]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[400]	rmse=0.020004	lr=0.131058
[1,0]<stderr>:INFO:root:Epoch[167] Batch[500] Loss[5.042]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[500]	rmse=0.019969	lr=0.130880
[1,0]<stderr>:INFO:root:Epoch[167] Batch[600] Loss[3.454]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[600]	rmse=0.019959	lr=0.130702
[1,0]<stderr>:INFO:root:Epoch[167] Batch[700] Loss[2.825]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[700]	rmse=0.019956	lr=0.130525
[1,0]<stderr>:INFO:root:Epoch[167] Batch[800] Loss[2.681]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[800]	rmse=0.019982	lr=0.130347
[1,0]<stderr>:INFO:root:Epoch[167] Batch[900] Loss[2.298]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[900]	rmse=0.020002	lr=0.130170
[1,0]<stderr>:INFO:root:Epoch[167] Batch[1000] Loss[5.095]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[1000]	rmse=0.019993	lr=0.129992
[1,0]<stderr>:INFO:root:Epoch[167] Batch[1100] Loss[2.619]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[1100]	rmse=0.020003	lr=0.129815
[1,0]<stderr>:INFO:root:Epoch[167] Batch[1200] Loss[4.088]
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[1200]	rmse=0.020004	lr=0.129638
[1,0]<stderr>:INFO:root:Epoch[167] Rank[0] Batch[1251]	Time cost=398.70	Train-metric=0.020007
[1,0]<stderr>:INFO:root:Epoch[167]	Speed: 3212.97 samples/sec
[1,0]<stderr>:INFO:root:Epoch[168] Batch[100] Loss[3.974]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[100]	rmse=0.019924	lr=0.129370
[1,0]<stderr>:INFO:root:Epoch[168] Batch[200] Loss[2.297]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[200]	rmse=0.019870	lr=0.129193
[1,0]<stderr>:INFO:root:Epoch[168] Batch[300] Loss[2.515]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[300]	rmse=0.019905	lr=0.129016
[1,0]<stderr>:INFO:root:Epoch[168] Batch[400] Loss[5.248]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[400]	rmse=0.019974	lr=0.128839
[1,0]<stderr>:INFO:root:Epoch[168] Batch[500] Loss[2.791]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[500]	rmse=0.019977	lr=0.128662
[1,0]<stderr>:INFO:root:Epoch[168] Batch[600] Loss[2.570]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[600]	rmse=0.019959	lr=0.128485
[1,0]<stderr>:INFO:root:Epoch[168] Batch[700] Loss[3.134]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[700]	rmse=0.019954	lr=0.128308
[1,0]<stderr>:INFO:root:Epoch[168] Batch[800] Loss[5.162]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[800]	rmse=0.019960	lr=0.128131
[1,0]<stderr>:INFO:root:Epoch[168] Batch[900] Loss[2.654]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[900]	rmse=0.019965	lr=0.127955
[1,0]<stderr>:INFO:root:Epoch[168] Batch[1000] Loss[4.985]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[1000]	rmse=0.019972	lr=0.127778
[1,0]<stderr>:INFO:root:Epoch[168] Batch[1100] Loss[2.563]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[1100]	rmse=0.019972	lr=0.127601
[1,0]<stderr>:INFO:root:Epoch[168] Batch[1200] Loss[4.843]
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[1200]	rmse=0.019969	lr=0.127425
[1,0]<stderr>:INFO:root:Epoch[168] Rank[0] Batch[1251]	Time cost=399.63	Train-metric=0.019973
[1,0]<stderr>:INFO:root:Epoch[168]	Speed: 3205.56 samples/sec
[1,0]<stderr>:INFO:root:Epoch[169] Batch[100] Loss[2.376]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[100]	rmse=0.020012	lr=0.127159
[1,0]<stderr>:INFO:root:Epoch[169] Batch[200] Loss[2.583]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[200]	rmse=0.019950	lr=0.126982
[1,0]<stderr>:INFO:root:Epoch[169] Batch[300] Loss[2.825]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[300]	rmse=0.019918	lr=0.126806
[1,0]<stderr>:INFO:root:Epoch[169] Batch[400] Loss[3.515]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[400]	rmse=0.019911	lr=0.126630
[1,0]<stderr>:INFO:root:Epoch[169] Batch[500] Loss[2.689]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[500]	rmse=0.019917	lr=0.126454
[1,0]<stderr>:INFO:root:Epoch[169] Batch[600] Loss[2.712]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[600]	rmse=0.019946	lr=0.126278
[1,0]<stderr>:INFO:root:Epoch[169] Batch[700] Loss[3.310]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[700]	rmse=0.019940	lr=0.126102
[1,0]<stderr>:INFO:root:Epoch[169] Batch[800] Loss[2.787]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[800]	rmse=0.019958	lr=0.125926
[1,0]<stderr>:INFO:root:Epoch[169] Batch[900] Loss[2.677]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[900]	rmse=0.019963	lr=0.125750
[1,0]<stderr>:INFO:root:Epoch[169] Batch[1000] Loss[3.244]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[1000]	rmse=0.019966	lr=0.125574
[1,0]<stderr>:INFO:root:Epoch[169] Batch[1100] Loss[4.889]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[1100]	rmse=0.019950	lr=0.125398
[1,0]<stderr>:INFO:root:Epoch[169] Batch[1200] Loss[4.823]
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[1200]	rmse=0.019947	lr=0.125222
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0] Batch[1251]	Time cost=398.96	Train-metric=0.019950
[1,0]<stderr>:INFO:root:Epoch[169]	Speed: 3210.90 samples/sec
[1,0]<stderr>:INFO:root:Epoch[169] Rank[0]	Validation-accuracy=0.663560	Validation-top_k_accuracy_5=0.874680
[1,0]<stderr>:INFO:root:Epoch[170] Batch[100] Loss[2.016]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[100]	rmse=0.019700	lr=0.124957
[1,0]<stderr>:INFO:root:Epoch[170] Batch[200] Loss[4.366]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[200]	rmse=0.019801	lr=0.124782
[1,0]<stderr>:INFO:root:Epoch[170] Batch[300] Loss[2.712]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[300]	rmse=0.019803	lr=0.124606
[1,0]<stderr>:INFO:root:Epoch[170] Batch[400] Loss[4.411]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[400]	rmse=0.019838	lr=0.124431
[1,0]<stderr>:INFO:root:Epoch[170] Batch[500] Loss[2.532]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[500]	rmse=0.019825	lr=0.124256
[1,0]<stderr>:INFO:root:Epoch[170] Batch[600] Loss[2.801]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[600]	rmse=0.019848	lr=0.124080
[1,0]<stderr>:INFO:root:Epoch[170] Batch[700] Loss[2.522]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[700]	rmse=0.019885	lr=0.123905
[1,0]<stderr>:INFO:root:Epoch[170] Batch[800] Loss[4.866]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[800]	rmse=0.019915	lr=0.123730
[1,0]<stderr>:INFO:root:Epoch[170] Batch[900] Loss[3.748]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[900]	rmse=0.019924	lr=0.123555
[1,0]<stderr>:INFO:root:Epoch[170] Batch[1000] Loss[2.645]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[1000]	rmse=0.019927	lr=0.123380
[1,0]<stderr>:INFO:root:Epoch[170] Batch[1100] Loss[2.645]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[1100]	rmse=0.019929	lr=0.123205
[1,0]<stderr>:INFO:root:Epoch[170] Batch[1200] Loss[2.955]
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[1200]	rmse=0.019932	lr=0.123030
[1,0]<stderr>:INFO:root:Epoch[170] Rank[0] Batch[1251]	Time cost=399.12	Train-metric=0.019928
[1,0]<stderr>:INFO:root:Epoch[170]	Speed: 3209.62 samples/sec
[1,0]<stderr>:INFO:root:Epoch[171] Batch[100] Loss[2.837]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[100]	rmse=0.019879	lr=0.122767
[1,0]<stderr>:INFO:root:Epoch[171] Batch[200] Loss[2.600]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[200]	rmse=0.019838	lr=0.122592
[1,0]<stderr>:INFO:root:Epoch[171] Batch[300] Loss[2.604]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[300]	rmse=0.019843	lr=0.122417
[1,0]<stderr>:INFO:root:Epoch[171] Batch[400] Loss[4.837]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[400]	rmse=0.019857	lr=0.122243
[1,0]<stderr>:INFO:root:Epoch[171] Batch[500] Loss[3.275]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[500]	rmse=0.019883	lr=0.122068
[1,0]<stderr>:INFO:root:Epoch[171] Batch[600] Loss[2.741]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[600]	rmse=0.019894	lr=0.121894
[1,0]<stderr>:INFO:root:Epoch[171] Batch[700] Loss[2.454]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[700]	rmse=0.019922	lr=0.121720
[1,0]<stderr>:INFO:root:Epoch[171] Batch[800] Loss[3.448]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[800]	rmse=0.019933	lr=0.121546
[1,0]<stderr>:INFO:root:Epoch[171] Batch[900] Loss[2.919]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[900]	rmse=0.019914	lr=0.121371
[1,0]<stderr>:INFO:root:Epoch[171] Batch[1000] Loss[3.825]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[1000]	rmse=0.019929	lr=0.121197
[1,0]<stderr>:INFO:root:Epoch[171] Batch[1100] Loss[4.823]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[1100]	rmse=0.019941	lr=0.121023
[1,0]<stderr>:INFO:root:Epoch[171] Batch[1200] Loss[3.387]
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[1200]	rmse=0.019953	lr=0.120849
[1,0]<stderr>:INFO:root:Epoch[171] Rank[0] Batch[1251]	Time cost=398.91	Train-metric=0.019944
[1,0]<stderr>:INFO:root:Epoch[171]	Speed: 3211.32 samples/sec
[1,0]<stderr>:INFO:root:Epoch[172] Batch[100] Loss[2.630]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[100]	rmse=0.019798	lr=0.120587
[1,0]<stderr>:INFO:root:Epoch[172] Batch[200] Loss[2.516]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[200]	rmse=0.019769	lr=0.120413
[1,0]<stderr>:INFO:root:Epoch[172] Batch[300] Loss[5.311]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[300]	rmse=0.019844	lr=0.120239
[1,0]<stderr>:INFO:root:Epoch[172] Batch[400] Loss[3.260]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[400]	rmse=0.019895	lr=0.120066
[1,0]<stderr>:INFO:root:Epoch[172] Batch[500] Loss[2.578]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[500]	rmse=0.019874	lr=0.119892
[1,0]<stderr>:INFO:root:Epoch[172] Batch[600] Loss[4.999]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[600]	rmse=0.019876	lr=0.119719
[1,0]<stderr>:INFO:root:Epoch[172] Batch[700] Loss[2.613]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[700]	rmse=0.019893	lr=0.119545
[1,0]<stderr>:INFO:root:Epoch[172] Batch[800] Loss[2.819]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[800]	rmse=0.019891	lr=0.119372
[1,0]<stderr>:INFO:root:Epoch[172] Batch[900] Loss[2.603]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[900]	rmse=0.019894	lr=0.119199
[1,0]<stderr>:INFO:root:Epoch[172] Batch[1000] Loss[3.184]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[1000]	rmse=0.019885	lr=0.119025
[1,0]<stderr>:INFO:root:Epoch[172] Batch[1100] Loss[2.692]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[1100]	rmse=0.019878	lr=0.118852
[1,0]<stderr>:INFO:root:Epoch[172] Batch[1200] Loss[2.648]
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[1200]	rmse=0.019872	lr=0.118679
[1,0]<stderr>:INFO:root:Epoch[172] Rank[0] Batch[1251]	Time cost=401.36	Train-metric=0.019881
[1,0]<stderr>:INFO:root:Epoch[172]	Speed: 3191.72 samples/sec
[1,0]<stderr>:INFO:root:Epoch[173] Batch[100] Loss[2.531]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[100]	rmse=0.019803	lr=0.118418
[1,0]<stderr>:INFO:root:Epoch[173] Batch[200] Loss[2.874]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[200]	rmse=0.019904	lr=0.118245
[1,0]<stderr>:INFO:root:Epoch[173] Batch[300] Loss[3.131]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[300]	rmse=0.019854	lr=0.118072
[1,0]<stderr>:INFO:root:Epoch[173] Batch[400] Loss[3.782]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[400]	rmse=0.019914	lr=0.117900
[1,0]<stderr>:INFO:root:Epoch[173] Batch[500] Loss[2.816]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[500]	rmse=0.019910	lr=0.117727
[1,0]<stderr>:INFO:root:Epoch[173] Batch[600] Loss[4.630]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[600]	rmse=0.019907	lr=0.117554
[1,0]<stderr>:INFO:root:Epoch[173] Batch[700] Loss[5.250]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[700]	rmse=0.019884	lr=0.117382
[1,0]<stderr>:INFO:root:Epoch[173] Batch[800] Loss[2.743]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[800]	rmse=0.019883	lr=0.117209
[1,0]<stderr>:INFO:root:Epoch[173] Batch[900] Loss[2.736]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[900]	rmse=0.019903	lr=0.117037
[1,0]<stderr>:INFO:root:Epoch[173] Batch[1000] Loss[2.899]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[1000]	rmse=0.019908	lr=0.116865
[1,0]<stderr>:INFO:root:Epoch[173] Batch[1100] Loss[3.230]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[1100]	rmse=0.019918	lr=0.116693
[1,0]<stderr>:INFO:root:Epoch[173] Batch[1200] Loss[4.559]
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[1200]	rmse=0.019933	lr=0.116521
[1,0]<stderr>:INFO:root:Epoch[173] Rank[0] Batch[1251]	Time cost=402.18	Train-metric=0.019924
[1,0]<stderr>:INFO:root:Epoch[173]	Speed: 3185.22 samples/sec
[1,0]<stderr>:INFO:root:Epoch[174] Batch[100] Loss[2.680]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[100]	rmse=0.019791	lr=0.116261
[1,0]<stderr>:INFO:root:Epoch[174] Batch[200] Loss[2.783]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[200]	rmse=0.019816	lr=0.116089
[1,0]<stderr>:INFO:root:Epoch[174] Batch[300] Loss[2.376]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[300]	rmse=0.019760	lr=0.115917
[1,0]<stderr>:INFO:root:Epoch[174] Batch[400] Loss[2.681]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[400]	rmse=0.019744	lr=0.115745
[1,0]<stderr>:INFO:root:Epoch[174] Batch[500] Loss[2.570]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[500]	rmse=0.019780	lr=0.115573
[1,0]<stderr>:INFO:root:Epoch[174] Batch[600] Loss[4.721]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[600]	rmse=0.019805	lr=0.115402
[1,0]<stderr>:INFO:root:Epoch[174] Batch[700] Loss[2.842]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[700]	rmse=0.019825	lr=0.115230
[1,0]<stderr>:INFO:root:Epoch[174] Batch[800] Loss[4.467]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[800]	rmse=0.019842	lr=0.115059
[1,0]<stderr>:INFO:root:Epoch[174] Batch[900] Loss[3.430]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[900]	rmse=0.019833	lr=0.114887
[1,0]<stderr>:INFO:root:Epoch[174] Batch[1000] Loss[5.032]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[1000]	rmse=0.019854	lr=0.114716
[1,0]<stderr>:INFO:root:Epoch[174] Batch[1100] Loss[2.660]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[1100]	rmse=0.019874	lr=0.114545
[1,0]<stderr>:INFO:root:Epoch[174] Batch[1200] Loss[2.790]
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[1200]	rmse=0.019891	lr=0.114374
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0] Batch[1251]	Time cost=402.32	Train-metric=0.019886
[1,0]<stderr>:INFO:root:Epoch[174]	Speed: 3184.07 samples/sec
[1,0]<stderr>:INFO:root:Epoch[174] Rank[0]	Validation-accuracy=0.669840	Validation-top_k_accuracy_5=0.880340
[1,0]<stderr>:INFO:root:Epoch[175] Batch[100] Loss[2.652]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[100]	rmse=0.019703	lr=0.114115
[1,0]<stderr>:INFO:root:Epoch[175] Batch[200] Loss[2.311]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[200]	rmse=0.019643	lr=0.113944
[1,0]<stderr>:INFO:root:Epoch[175] Batch[300] Loss[2.852]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[300]	rmse=0.019689	lr=0.113773
[1,0]<stderr>:INFO:root:Epoch[175] Batch[400] Loss[2.993]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[400]	rmse=0.019736	lr=0.113602
[1,0]<stderr>:INFO:root:Epoch[175] Batch[500] Loss[4.791]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[500]	rmse=0.019737	lr=0.113432
[1,0]<stderr>:INFO:root:Epoch[175] Batch[600] Loss[2.588]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[600]	rmse=0.019730	lr=0.113261
[1,0]<stderr>:INFO:root:Epoch[175] Batch[700] Loss[2.692]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[700]	rmse=0.019738	lr=0.113090
[1,0]<stderr>:INFO:root:Epoch[175] Batch[800] Loss[3.464]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[800]	rmse=0.019773	lr=0.112920
[1,0]<stderr>:INFO:root:Epoch[175] Batch[900] Loss[2.773]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[900]	rmse=0.019789	lr=0.112749
[1,0]<stderr>:INFO:root:Epoch[175] Batch[1000] Loss[2.971]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[1000]	rmse=0.019792	lr=0.112579
[1,0]<stderr>:INFO:root:Epoch[175] Batch[1100] Loss[2.588]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[1100]	rmse=0.019795	lr=0.112409
[1,0]<stderr>:INFO:root:Epoch[175] Batch[1200] Loss[2.612]
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[1200]	rmse=0.019810	lr=0.112239
[1,0]<stderr>:INFO:root:Epoch[175] Rank[0] Batch[1251]	Time cost=402.05	Train-metric=0.019806
[1,0]<stderr>:INFO:root:Epoch[175]	Speed: 3186.22 samples/sec
[1,0]<stderr>:INFO:root:Epoch[176] Batch[100] Loss[2.435]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[100]	rmse=0.019924	lr=0.111982
[1,0]<stderr>:INFO:root:Epoch[176] Batch[200] Loss[2.526]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[200]	rmse=0.019841	lr=0.111812
[1,0]<stderr>:INFO:root:Epoch[176] Batch[300] Loss[4.743]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[300]	rmse=0.019850	lr=0.111642
[1,0]<stderr>:INFO:root:Epoch[176] Batch[400] Loss[3.237]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[400]	rmse=0.019828	lr=0.111472
[1,0]<stderr>:INFO:root:Epoch[176] Batch[500] Loss[5.188]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[500]	rmse=0.019807	lr=0.111302
[1,0]<stderr>:INFO:root:Epoch[176] Batch[600] Loss[3.429]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[600]	rmse=0.019846	lr=0.111132
[1,0]<stderr>:INFO:root:Epoch[176] Batch[700] Loss[3.748]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[700]	rmse=0.019846	lr=0.110963
[1,0]<stderr>:INFO:root:Epoch[176] Batch[800] Loss[5.019]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[800]	rmse=0.019863	lr=0.110793
[1,0]<stderr>:INFO:root:Epoch[176] Batch[900] Loss[5.048]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[900]	rmse=0.019859	lr=0.110624
[1,0]<stderr>:INFO:root:Epoch[176] Batch[1000] Loss[3.069]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[1000]	rmse=0.019852	lr=0.110454
[1,0]<stderr>:INFO:root:Epoch[176] Batch[1100] Loss[3.071]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[1100]	rmse=0.019844	lr=0.110285
[1,0]<stderr>:INFO:root:Epoch[176] Batch[1200] Loss[3.147]
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[1200]	rmse=0.019862	lr=0.110116
[1,0]<stderr>:INFO:root:Epoch[176] Rank[0] Batch[1251]	Time cost=400.02	Train-metric=0.019871
[1,0]<stderr>:INFO:root:Epoch[176]	Speed: 3202.36 samples/sec
[1,0]<stderr>:INFO:root:Epoch[177] Batch[100] Loss[2.764]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[100]	rmse=0.019630	lr=0.109861
[1,0]<stderr>:INFO:root:Epoch[177] Batch[200] Loss[2.612]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[200]	rmse=0.019770	lr=0.109692
[1,0]<stderr>:INFO:root:Epoch[177] Batch[300] Loss[4.360]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[300]	rmse=0.019820	lr=0.109523
[1,0]<stderr>:INFO:root:Epoch[177] Batch[400] Loss[2.681]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[400]	rmse=0.019786	lr=0.109354
[1,0]<stderr>:INFO:root:Epoch[177] Batch[500] Loss[2.591]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[500]	rmse=0.019813	lr=0.109185
[1,0]<stderr>:INFO:root:Epoch[177] Batch[600] Loss[5.038]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[600]	rmse=0.019808	lr=0.109016
[1,0]<stderr>:INFO:root:Epoch[177] Batch[700] Loss[2.389]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[700]	rmse=0.019802	lr=0.108848
[1,0]<stderr>:INFO:root:Epoch[177] Batch[800] Loss[2.522]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[800]	rmse=0.019802	lr=0.108679
[1,0]<stderr>:INFO:root:Epoch[177] Batch[900] Loss[2.968]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[900]	rmse=0.019826	lr=0.108511
[1,0]<stderr>:INFO:root:Epoch[177] Batch[1000] Loss[2.609]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[1000]	rmse=0.019825	lr=0.108342
[1,0]<stderr>:INFO:root:Epoch[177] Batch[1100] Loss[3.836]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[1100]	rmse=0.019842	lr=0.108174
[1,0]<stderr>:INFO:root:Epoch[177] Batch[1200] Loss[4.361]
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[1200]	rmse=0.019838	lr=0.108006
[1,0]<stderr>:INFO:root:Epoch[177] Rank[0] Batch[1251]	Time cost=400.91	Train-metric=0.019835
[1,0]<stderr>:INFO:root:Epoch[177]	Speed: 3195.25 samples/sec
[1,0]<stderr>:INFO:root:Epoch[178] Batch[100] Loss[4.785]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[100]	rmse=0.019715	lr=0.107752
[1,0]<stderr>:INFO:root:Epoch[178] Batch[200] Loss[2.626]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[200]	rmse=0.019699	lr=0.107584
[1,0]<stderr>:INFO:root:Epoch[178] Batch[300] Loss[4.103]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[300]	rmse=0.019670	lr=0.107416
[1,0]<stderr>:INFO:root:Epoch[178] Batch[400] Loss[4.938]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[400]	rmse=0.019651	lr=0.107248
[1,0]<stderr>:INFO:root:Epoch[178] Batch[500] Loss[2.782]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[500]	rmse=0.019696	lr=0.107081
[1,0]<stderr>:INFO:root:Epoch[178] Batch[600] Loss[2.632]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[600]	rmse=0.019710	lr=0.106913
[1,0]<stderr>:INFO:root:Epoch[178] Batch[700] Loss[2.536]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[700]	rmse=0.019729	lr=0.106745
[1,0]<stderr>:INFO:root:Epoch[178] Batch[800] Loss[4.524]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[800]	rmse=0.019753	lr=0.106578
[1,0]<stderr>:INFO:root:Epoch[178] Batch[900] Loss[3.815]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[900]	rmse=0.019768	lr=0.106411
[1,0]<stderr>:INFO:root:Epoch[178] Batch[1000] Loss[2.473]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[1000]	rmse=0.019769	lr=0.106243
[1,0]<stderr>:INFO:root:Epoch[178] Batch[1100] Loss[2.571]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[1100]	rmse=0.019790	lr=0.106076
[1,0]<stderr>:INFO:root:Epoch[178] Batch[1200] Loss[4.631]
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[1200]	rmse=0.019789	lr=0.105909
[1,0]<stderr>:INFO:root:Epoch[178] Rank[0] Batch[1251]	Time cost=399.04	Train-metric=0.019790
[1,0]<stderr>:INFO:root:Epoch[178]	Speed: 3210.25 samples/sec
[1,0]<stderr>:INFO:root:Epoch[179] Batch[100] Loss[2.453]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[100]	rmse=0.019583	lr=0.105657
[1,0]<stderr>:INFO:root:Epoch[179] Batch[200] Loss[2.813]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[200]	rmse=0.019690	lr=0.105490
[1,0]<stderr>:INFO:root:Epoch[179] Batch[300] Loss[2.675]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[300]	rmse=0.019743	lr=0.105323
[1,0]<stderr>:INFO:root:Epoch[179] Batch[400] Loss[2.925]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[400]	rmse=0.019725	lr=0.105156
[1,0]<stderr>:INFO:root:Epoch[179] Batch[500] Loss[2.477]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[500]	rmse=0.019741	lr=0.104989
[1,0]<stderr>:INFO:root:Epoch[179] Batch[600] Loss[2.522]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[600]	rmse=0.019746	lr=0.104823
[1,0]<stderr>:INFO:root:Epoch[179] Batch[700] Loss[3.501]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[700]	rmse=0.019753	lr=0.104656
[1,0]<stderr>:INFO:root:Epoch[179] Batch[800] Loss[2.500]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[800]	rmse=0.019759	lr=0.104490
[1,0]<stderr>:INFO:root:Epoch[179] Batch[900] Loss[4.839]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[900]	rmse=0.019762	lr=0.104323
[1,0]<stderr>:INFO:root:Epoch[179] Batch[1000] Loss[2.583]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[1000]	rmse=0.019764	lr=0.104157
[1,0]<stderr>:INFO:root:Epoch[179] Batch[1100] Loss[3.936]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[1100]	rmse=0.019775	lr=0.103991
[1,0]<stderr>:INFO:root:Epoch[179] Batch[1200] Loss[3.072]
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[1200]	rmse=0.019783	lr=0.103825
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0] Batch[1251]	Time cost=400.20	Train-metric=0.019793
[1,0]<stderr>:INFO:root:Epoch[179]	Speed: 3200.96 samples/sec
[1,0]<stderr>:INFO:root:Epoch[179] Rank[0]	Validation-accuracy=0.677220	Validation-top_k_accuracy_5=0.880600
[1,0]<stderr>:INFO:root:Epoch[180] Batch[100] Loss[2.526]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[100]	rmse=0.019732	lr=0.103574
[1,0]<stderr>:INFO:root:Epoch[180] Batch[200] Loss[2.628]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[200]	rmse=0.019670	lr=0.103408
[1,0]<stderr>:INFO:root:Epoch[180] Batch[300] Loss[2.595]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[300]	rmse=0.019704	lr=0.103243
[1,0]<stderr>:INFO:root:Epoch[180] Batch[400] Loss[3.280]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[400]	rmse=0.019724	lr=0.103077
[1,0]<stderr>:INFO:root:Epoch[180] Batch[500] Loss[4.102]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[500]	rmse=0.019754	lr=0.102911
[1,0]<stderr>:INFO:root:Epoch[180] Batch[600] Loss[3.379]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[600]	rmse=0.019755	lr=0.102746
[1,0]<stderr>:INFO:root:Epoch[180] Batch[700] Loss[2.551]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[700]	rmse=0.019743	lr=0.102580
[1,0]<stderr>:INFO:root:Epoch[180] Batch[800] Loss[4.468]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[800]	rmse=0.019737	lr=0.102415
[1,0]<stderr>:INFO:root:Epoch[180] Batch[900] Loss[4.049]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[900]	rmse=0.019744	lr=0.102250
[1,0]<stderr>:INFO:root:Epoch[180] Batch[1000] Loss[4.153]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[1000]	rmse=0.019727	lr=0.102085
[1,0]<stderr>:INFO:root:Epoch[180] Batch[1100] Loss[2.542]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[1100]	rmse=0.019751	lr=0.101919
[1,0]<stderr>:INFO:root:Epoch[180] Batch[1200] Loss[4.052]
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[1200]	rmse=0.019747	lr=0.101754
[1,0]<stderr>:INFO:root:Epoch[180] Rank[0] Batch[1251]	Time cost=398.25	Train-metric=0.019749
[1,0]<stderr>:INFO:root:Epoch[180]	Speed: 3216.60 samples/sec
[1,0]<stderr>:INFO:root:Epoch[181] Batch[100] Loss[2.622]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[100]	rmse=0.019692	lr=0.101505
[1,0]<stderr>:INFO:root:Epoch[181] Batch[200] Loss[3.363]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[200]	rmse=0.019692	lr=0.101341
[1,0]<stderr>:INFO:root:Epoch[181] Batch[300] Loss[2.699]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[300]	rmse=0.019670	lr=0.101176
[1,0]<stderr>:INFO:root:Epoch[181] Batch[400] Loss[2.645]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[400]	rmse=0.019649	lr=0.101011
[1,0]<stderr>:INFO:root:Epoch[181] Batch[500] Loss[2.634]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[500]	rmse=0.019670	lr=0.100847
[1,0]<stderr>:INFO:root:Epoch[181] Batch[600] Loss[5.049]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[600]	rmse=0.019676	lr=0.100682
[1,0]<stderr>:INFO:root:Epoch[181] Batch[700] Loss[2.350]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[700]	rmse=0.019671	lr=0.100518
[1,0]<stderr>:INFO:root:Epoch[181] Batch[800] Loss[2.727]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[800]	rmse=0.019668	lr=0.100354
[1,0]<stderr>:INFO:root:Epoch[181] Batch[900] Loss[2.657]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[900]	rmse=0.019692	lr=0.100190
[1,0]<stderr>:INFO:root:Epoch[181] Batch[1000] Loss[4.876]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[1000]	rmse=0.019715	lr=0.100026
[1,0]<stderr>:INFO:root:Epoch[181] Batch[1100] Loss[4.580]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[1100]	rmse=0.019724	lr=0.099862
[1,0]<stderr>:INFO:root:Epoch[181] Batch[1200] Loss[4.721]
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[1200]	rmse=0.019725	lr=0.099698
[1,0]<stderr>:INFO:root:Epoch[181] Rank[0] Batch[1251]	Time cost=399.16	Train-metric=0.019716
[1,0]<stderr>:INFO:root:Epoch[181]	Speed: 3209.27 samples/sec
[1,0]<stderr>:INFO:root:Epoch[182] Batch[100] Loss[2.742]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[100]	rmse=0.019599	lr=0.099451
[1,0]<stderr>:INFO:root:Epoch[182] Batch[200] Loss[5.080]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[200]	rmse=0.019565	lr=0.099287
[1,0]<stderr>:INFO:root:Epoch[182] Batch[300] Loss[2.847]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[300]	rmse=0.019610	lr=0.099123
[1,0]<stderr>:INFO:root:Epoch[182] Batch[400] Loss[4.859]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[400]	rmse=0.019611	lr=0.098960
[1,0]<stderr>:INFO:root:Epoch[182] Batch[500] Loss[2.471]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[500]	rmse=0.019635	lr=0.098796
[1,0]<stderr>:INFO:root:Epoch[182] Batch[600] Loss[2.528]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[600]	rmse=0.019680	lr=0.098633
[1,0]<stderr>:INFO:root:Epoch[182] Batch[700] Loss[3.794]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[700]	rmse=0.019710	lr=0.098470
[1,0]<stderr>:INFO:root:Epoch[182] Batch[800] Loss[2.696]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[800]	rmse=0.019715	lr=0.098307
[1,0]<stderr>:INFO:root:Epoch[182] Batch[900] Loss[2.828]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[900]	rmse=0.019696	lr=0.098144
[1,0]<stderr>:INFO:root:Epoch[182] Batch[1000] Loss[3.016]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[1000]	rmse=0.019691	lr=0.097981
[1,0]<stderr>:INFO:root:Epoch[182] Batch[1100] Loss[2.320]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[1100]	rmse=0.019705	lr=0.097818
[1,0]<stderr>:INFO:root:Epoch[182] Batch[1200] Loss[2.812]
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[1200]	rmse=0.019700	lr=0.097655
[1,0]<stderr>:INFO:root:Epoch[182] Rank[0] Batch[1251]	Time cost=398.94	Train-metric=0.019697
[1,0]<stderr>:INFO:root:Epoch[182]	Speed: 3211.08 samples/sec
[1,0]<stderr>:INFO:root:Epoch[183] Batch[100] Loss[5.002]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[100]	rmse=0.019642	lr=0.097410
[1,0]<stderr>:INFO:root:Epoch[183] Batch[200] Loss[2.830]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[200]	rmse=0.019562	lr=0.097247
[1,0]<stderr>:INFO:root:Epoch[183] Batch[300] Loss[2.405]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[300]	rmse=0.019580	lr=0.097085
[1,0]<stderr>:INFO:root:Epoch[183] Batch[400] Loss[2.928]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[400]	rmse=0.019585	lr=0.096922
[1,0]<stderr>:INFO:root:Epoch[183] Batch[500] Loss[2.476]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[500]	rmse=0.019614	lr=0.096760
[1,0]<stderr>:INFO:root:Epoch[183] Batch[600] Loss[2.470]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[600]	rmse=0.019612	lr=0.096598
[1,0]<stderr>:INFO:root:Epoch[183] Batch[700] Loss[4.808]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[700]	rmse=0.019626	lr=0.096436
[1,0]<stderr>:INFO:root:Epoch[183] Batch[800] Loss[2.576]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[800]	rmse=0.019616	lr=0.096274
[1,0]<stderr>:INFO:root:Epoch[183] Batch[900] Loss[2.666]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[900]	rmse=0.019637	lr=0.096112
[1,0]<stderr>:INFO:root:Epoch[183] Batch[1000] Loss[4.919]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[1000]	rmse=0.019655	lr=0.095950
[1,0]<stderr>:INFO:root:Epoch[183] Batch[1100] Loss[3.399]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[1100]	rmse=0.019671	lr=0.095789
[1,0]<stderr>:INFO:root:Epoch[183] Batch[1200] Loss[3.953]
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[1200]	rmse=0.019673	lr=0.095627
[1,0]<stderr>:INFO:root:Epoch[183] Rank[0] Batch[1251]	Time cost=399.27	Train-metric=0.019674
[1,0]<stderr>:INFO:root:Epoch[183]	Speed: 3208.44 samples/sec
[1,0]<stderr>:INFO:root:Epoch[184] Batch[100] Loss[2.384]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[100]	rmse=0.019477	lr=0.095383
[1,0]<stderr>:INFO:root:Epoch[184] Batch[200] Loss[2.314]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[200]	rmse=0.019439	lr=0.095222
[1,0]<stderr>:INFO:root:Epoch[184] Batch[300] Loss[2.611]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[300]	rmse=0.019485	lr=0.095061
[1,0]<stderr>:INFO:root:Epoch[184] Batch[400] Loss[4.652]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[400]	rmse=0.019535	lr=0.094899
[1,0]<stderr>:INFO:root:Epoch[184] Batch[500] Loss[2.979]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[500]	rmse=0.019579	lr=0.094738
[1,0]<stderr>:INFO:root:Epoch[184] Batch[600] Loss[3.448]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[600]	rmse=0.019571	lr=0.094577
[1,0]<stderr>:INFO:root:Epoch[184] Batch[700] Loss[2.401]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[700]	rmse=0.019582	lr=0.094417
[1,0]<stderr>:INFO:root:Epoch[184] Batch[800] Loss[3.187]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[800]	rmse=0.019601	lr=0.094256
[1,0]<stderr>:INFO:root:Epoch[184] Batch[900] Loss[2.420]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[900]	rmse=0.019611	lr=0.094095
[1,0]<stderr>:INFO:root:Epoch[184] Batch[1000] Loss[3.447]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[1000]	rmse=0.019629	lr=0.093934
[1,0]<stderr>:INFO:root:Epoch[184] Batch[1100] Loss[3.678]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[1100]	rmse=0.019613	lr=0.093774
[1,0]<stderr>:INFO:root:Epoch[184] Batch[1200] Loss[2.431]
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[1200]	rmse=0.019614	lr=0.093614
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0] Batch[1251]	Time cost=399.48	Train-metric=0.019616
[1,0]<stderr>:INFO:root:Epoch[184]	Speed: 3206.71 samples/sec
[1,0]<stderr>:INFO:root:Epoch[184] Rank[0]	Validation-accuracy=0.683280	Validation-top_k_accuracy_5=0.886280
[1,0]<stderr>:INFO:root:Epoch[185] Batch[100] Loss[2.622]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[100]	rmse=0.019698	lr=0.093371
[1,0]<stderr>:INFO:root:Epoch[185] Batch[200] Loss[4.981]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[200]	rmse=0.019691	lr=0.093211
[1,0]<stderr>:INFO:root:Epoch[185] Batch[300] Loss[2.476]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[300]	rmse=0.019690	lr=0.093051
[1,0]<stderr>:INFO:root:Epoch[185] Batch[400] Loss[2.759]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[400]	rmse=0.019676	lr=0.092891
[1,0]<stderr>:INFO:root:Epoch[185] Batch[500] Loss[2.411]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[500]	rmse=0.019653	lr=0.092731
[1,0]<stderr>:INFO:root:Epoch[185] Batch[600] Loss[3.478]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[600]	rmse=0.019675	lr=0.092572
[1,0]<stderr>:INFO:root:Epoch[185] Batch[700] Loss[2.735]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[700]	rmse=0.019680	lr=0.092412
[1,0]<stderr>:INFO:root:Epoch[185] Batch[800] Loss[3.261]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[800]	rmse=0.019679	lr=0.092252
[1,0]<stderr>:INFO:root:Epoch[185] Batch[900] Loss[2.575]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[900]	rmse=0.019680	lr=0.092093
[1,0]<stderr>:INFO:root:Epoch[185] Batch[1000] Loss[3.667]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[1000]	rmse=0.019680	lr=0.091933
[1,0]<stderr>:INFO:root:Epoch[185] Batch[1100] Loss[4.797]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[1100]	rmse=0.019677	lr=0.091774
[1,0]<stderr>:INFO:root:Epoch[185] Batch[1200] Loss[2.723]
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[1200]	rmse=0.019672	lr=0.091615
[1,0]<stderr>:INFO:root:Epoch[185] Rank[0] Batch[1251]	Time cost=398.62	Train-metric=0.019669
[1,0]<stderr>:INFO:root:Epoch[185]	Speed: 3213.67 samples/sec
[1,0]<stderr>:INFO:root:Epoch[186] Batch[100] Loss[2.652]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[100]	rmse=0.019692	lr=0.091375
[1,0]<stderr>:INFO:root:Epoch[186] Batch[200] Loss[2.856]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[200]	rmse=0.019569	lr=0.091216
[1,0]<stderr>:INFO:root:Epoch[186] Batch[300] Loss[4.685]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[300]	rmse=0.019559	lr=0.091057
[1,0]<stderr>:INFO:root:Epoch[186] Batch[400] Loss[2.675]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[400]	rmse=0.019567	lr=0.090898
[1,0]<stderr>:INFO:root:Epoch[186] Batch[500] Loss[4.075]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[500]	rmse=0.019529	lr=0.090739
[1,0]<stderr>:INFO:root:Epoch[186] Batch[600] Loss[5.139]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[600]	rmse=0.019528	lr=0.090581
[1,0]<stderr>:INFO:root:Epoch[186] Batch[700] Loss[2.626]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[700]	rmse=0.019548	lr=0.090422
[1,0]<stderr>:INFO:root:Epoch[186] Batch[800] Loss[3.088]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[800]	rmse=0.019525	lr=0.090264
[1,0]<stderr>:INFO:root:Epoch[186] Batch[900] Loss[2.967]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[900]	rmse=0.019537	lr=0.090106
[1,0]<stderr>:INFO:root:Epoch[186] Batch[1000] Loss[2.564]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[1000]	rmse=0.019554	lr=0.089948
[1,0]<stderr>:INFO:root:Epoch[186] Batch[1100] Loss[4.287]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[1100]	rmse=0.019560	lr=0.089789
[1,0]<stderr>:INFO:root:Epoch[186] Batch[1200] Loss[2.711]
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[1200]	rmse=0.019574	lr=0.089631
[1,0]<stderr>:INFO:root:Epoch[186] Rank[0] Batch[1251]	Time cost=399.26	Train-metric=0.019583
[1,0]<stderr>:INFO:root:Epoch[186]	Speed: 3208.50 samples/sec
[1,0]<stderr>:INFO:root:Epoch[187] Batch[100] Loss[4.557]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[100]	rmse=0.019580	lr=0.089393
[1,0]<stderr>:INFO:root:Epoch[187] Batch[200] Loss[3.309]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[200]	rmse=0.019593	lr=0.089235
[1,0]<stderr>:INFO:root:Epoch[187] Batch[300] Loss[2.629]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[300]	rmse=0.019564	lr=0.089078
[1,0]<stderr>:INFO:root:Epoch[187] Batch[400] Loss[2.401]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[400]	rmse=0.019562	lr=0.088920
[1,0]<stderr>:INFO:root:Epoch[187] Batch[500] Loss[3.028]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[500]	rmse=0.019581	lr=0.088763
[1,0]<stderr>:INFO:root:Epoch[187] Batch[600] Loss[4.766]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[600]	rmse=0.019592	lr=0.088605
[1,0]<stderr>:INFO:root:Epoch[187] Batch[700] Loss[2.442]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[700]	rmse=0.019600	lr=0.088448
[1,0]<stderr>:INFO:root:Epoch[187] Batch[800] Loss[2.544]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[800]	rmse=0.019613	lr=0.088291
[1,0]<stderr>:INFO:root:Epoch[187] Batch[900] Loss[4.682]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[900]	rmse=0.019601	lr=0.088134
[1,0]<stderr>:INFO:root:Epoch[187] Batch[1000] Loss[3.086]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[1000]	rmse=0.019611	lr=0.087977
[1,0]<stderr>:INFO:root:Epoch[187] Batch[1100] Loss[2.760]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[1100]	rmse=0.019616	lr=0.087820
[1,0]<stderr>:INFO:root:Epoch[187] Batch[1200] Loss[4.116]
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[1200]	rmse=0.019605	lr=0.087664
[1,0]<stderr>:INFO:root:Epoch[187] Rank[0] Batch[1251]	Time cost=399.42	Train-metric=0.019603
[1,0]<stderr>:INFO:root:Epoch[187]	Speed: 3207.18 samples/sec
[1,0]<stderr>:INFO:root:Epoch[188] Batch[100] Loss[4.732]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[100]	rmse=0.019445	lr=0.087427
[1,0]<stderr>:INFO:root:Epoch[188] Batch[200] Loss[2.767]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[200]	rmse=0.019503	lr=0.087271
[1,0]<stderr>:INFO:root:Epoch[188] Batch[300] Loss[2.952]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[300]	rmse=0.019482	lr=0.087114
[1,0]<stderr>:INFO:root:Epoch[188] Batch[400] Loss[2.681]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[400]	rmse=0.019486	lr=0.086958
[1,0]<stderr>:INFO:root:Epoch[188] Batch[500] Loss[3.114]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[500]	rmse=0.019514	lr=0.086802
[1,0]<stderr>:INFO:root:Epoch[188] Batch[600] Loss[4.881]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[600]	rmse=0.019523	lr=0.086646
[1,0]<stderr>:INFO:root:Epoch[188] Batch[700] Loss[2.651]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[700]	rmse=0.019533	lr=0.086490
[1,0]<stderr>:INFO:root:Epoch[188] Batch[800] Loss[2.737]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[800]	rmse=0.019551	lr=0.086334
[1,0]<stderr>:INFO:root:Epoch[188] Batch[900] Loss[3.804]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[900]	rmse=0.019573	lr=0.086178
[1,0]<stderr>:INFO:root:Epoch[188] Batch[1000] Loss[2.885]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[1000]	rmse=0.019559	lr=0.086022
[1,0]<stderr>:INFO:root:Epoch[188] Batch[1100] Loss[3.014]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[1100]	rmse=0.019561	lr=0.085867
[1,0]<stderr>:INFO:root:Epoch[188] Batch[1200] Loss[2.832]
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[1200]	rmse=0.019563	lr=0.085711
[1,0]<stderr>:INFO:root:Epoch[188] Rank[0] Batch[1251]	Time cost=400.32	Train-metric=0.019572
[1,0]<stderr>:INFO:root:Epoch[188]	Speed: 3200.02 samples/sec
[1,0]<stderr>:INFO:root:Epoch[189] Batch[100] Loss[2.659]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[100]	rmse=0.019396	lr=0.085477
[1,0]<stderr>:INFO:root:Epoch[189] Batch[200] Loss[4.683]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[200]	rmse=0.019354	lr=0.085322
[1,0]<stderr>:INFO:root:Epoch[189] Batch[300] Loss[2.444]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[300]	rmse=0.019408	lr=0.085166
[1,0]<stderr>:INFO:root:Epoch[189] Batch[400] Loss[4.681]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[400]	rmse=0.019425	lr=0.085012
[1,0]<stderr>:INFO:root:Epoch[189] Batch[500] Loss[4.784]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[500]	rmse=0.019447	lr=0.084857
[1,0]<stderr>:INFO:root:Epoch[189] Batch[600] Loss[4.084]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[600]	rmse=0.019448	lr=0.084702
[1,0]<stderr>:INFO:root:Epoch[189] Batch[700] Loss[3.829]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[700]	rmse=0.019461	lr=0.084547
[1,0]<stderr>:INFO:root:Epoch[189] Batch[800] Loss[2.469]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[800]	rmse=0.019475	lr=0.084393
[1,0]<stderr>:INFO:root:Epoch[189] Batch[900] Loss[2.441]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[900]	rmse=0.019488	lr=0.084238
[1,0]<stderr>:INFO:root:Epoch[189] Batch[1000] Loss[2.474]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[1000]	rmse=0.019472	lr=0.084084
[1,0]<stderr>:INFO:root:Epoch[189] Batch[1100] Loss[5.087]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[1100]	rmse=0.019493	lr=0.083929
[1,0]<stderr>:INFO:root:Epoch[189] Batch[1200] Loss[2.360]
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[1200]	rmse=0.019498	lr=0.083775
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0] Batch[1251]	Time cost=399.63	Train-metric=0.019501
[1,0]<stderr>:INFO:root:Epoch[189]	Speed: 3205.50 samples/sec
[1,0]<stderr>:INFO:root:Epoch[189] Rank[0]	Validation-accuracy=0.685360	Validation-top_k_accuracy_5=0.887720
[1,0]<stderr>:INFO:root:Epoch[190] Batch[100] Loss[2.561]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[100]	rmse=0.019361	lr=0.083543
[1,0]<stderr>:INFO:root:Epoch[190] Batch[200] Loss[2.671]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[200]	rmse=0.019344	lr=0.083389
[1,0]<stderr>:INFO:root:Epoch[190] Batch[300] Loss[3.975]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[300]	rmse=0.019376	lr=0.083235
[1,0]<stderr>:INFO:root:Epoch[190] Batch[400] Loss[4.282]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[400]	rmse=0.019408	lr=0.083081
[1,0]<stderr>:INFO:root:Epoch[190] Batch[500] Loss[3.060]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[500]	rmse=0.019403	lr=0.082928
[1,0]<stderr>:INFO:root:Epoch[190] Batch[600] Loss[3.314]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[600]	rmse=0.019463	lr=0.082774
[1,0]<stderr>:INFO:root:Epoch[190] Batch[700] Loss[2.451]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[700]	rmse=0.019469	lr=0.082621
[1,0]<stderr>:INFO:root:Epoch[190] Batch[800] Loss[4.566]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[800]	rmse=0.019472	lr=0.082467
[1,0]<stderr>:INFO:root:Epoch[190] Batch[900] Loss[4.436]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[900]	rmse=0.019481	lr=0.082314
[1,0]<stderr>:INFO:root:Epoch[190] Batch[1000] Loss[2.775]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[1000]	rmse=0.019486	lr=0.082161
[1,0]<stderr>:INFO:root:Epoch[190] Batch[1100] Loss[2.787]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[1100]	rmse=0.019500	lr=0.082008
[1,0]<stderr>:INFO:root:Epoch[190] Batch[1200] Loss[4.274]
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[1200]	rmse=0.019507	lr=0.081855
[1,0]<stderr>:INFO:root:Epoch[190] Rank[0] Batch[1251]	Time cost=398.79	Train-metric=0.019514
[1,0]<stderr>:INFO:root:Epoch[190]	Speed: 3212.24 samples/sec
[1,0]<stderr>:INFO:root:Epoch[191] Batch[100] Loss[2.539]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[100]	rmse=0.019310	lr=0.081625
[1,0]<stderr>:INFO:root:Epoch[191] Batch[200] Loss[2.564]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[200]	rmse=0.019424	lr=0.081472
[1,0]<stderr>:INFO:root:Epoch[191] Batch[300] Loss[2.723]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[300]	rmse=0.019444	lr=0.081320
[1,0]<stderr>:INFO:root:Epoch[191] Batch[400] Loss[2.307]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[400]	rmse=0.019432	lr=0.081167
[1,0]<stderr>:INFO:root:Epoch[191] Batch[500] Loss[2.934]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[500]	rmse=0.019417	lr=0.081015
[1,0]<stderr>:INFO:root:Epoch[191] Batch[600] Loss[2.872]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[600]	rmse=0.019423	lr=0.080863
[1,0]<stderr>:INFO:root:Epoch[191] Batch[700] Loss[2.525]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[700]	rmse=0.019450	lr=0.080711
[1,0]<stderr>:INFO:root:Epoch[191] Batch[800] Loss[2.609]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[800]	rmse=0.019453	lr=0.080559
[1,0]<stderr>:INFO:root:Epoch[191] Batch[900] Loss[2.304]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[900]	rmse=0.019468	lr=0.080407
[1,0]<stderr>:INFO:root:Epoch[191] Batch[1000] Loss[4.266]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[1000]	rmse=0.019472	lr=0.080255
[1,0]<stderr>:INFO:root:Epoch[191] Batch[1100] Loss[2.684]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[1100]	rmse=0.019465	lr=0.080104
[1,0]<stderr>:INFO:root:Epoch[191] Batch[1200] Loss[2.541]
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[1200]	rmse=0.019479	lr=0.079952
[1,0]<stderr>:INFO:root:Epoch[191] Rank[0] Batch[1251]	Time cost=399.33	Train-metric=0.019487
[1,0]<stderr>:INFO:root:Epoch[191]	Speed: 3207.92 samples/sec
[1,0]<stderr>:INFO:root:Epoch[192] Batch[100] Loss[2.465]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[100]	rmse=0.019266	lr=0.079724
[1,0]<stderr>:INFO:root:Epoch[192] Batch[200] Loss[2.682]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[200]	rmse=0.019386	lr=0.079572
[1,0]<stderr>:INFO:root:Epoch[192] Batch[300] Loss[2.647]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[300]	rmse=0.019394	lr=0.079421
[1,0]<stderr>:INFO:root:Epoch[192] Batch[400] Loss[2.368]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[400]	rmse=0.019453	lr=0.079270
[1,0]<stderr>:INFO:root:Epoch[192] Batch[500] Loss[2.542]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[500]	rmse=0.019440	lr=0.079119
[1,2]<stderr>:[ip-172-31-29-212][[55333,1],2][btl_tcp.c:559:mca_btl_tcp_recv_blocking] recv(116) failed: Connection reset by peer (104)
[1,0]<stderr>:INFO:root:Epoch[192] Batch[600] Loss[4.542]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[600]	rmse=0.019413	lr=0.078968
[1,0]<stderr>:INFO:root:Epoch[192] Batch[700] Loss[3.157]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[700]	rmse=0.019420	lr=0.078818
[1,0]<stderr>:INFO:root:Epoch[192] Batch[800] Loss[4.956]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[800]	rmse=0.019442	lr=0.078667
[1,0]<stderr>:INFO:root:Epoch[192] Batch[900] Loss[4.319]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[900]	rmse=0.019437	lr=0.078517
[1,0]<stderr>:INFO:root:Epoch[192] Batch[1000] Loss[4.140]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[1000]	rmse=0.019429	lr=0.078366
[1,0]<stderr>:INFO:root:Epoch[192] Batch[1100] Loss[4.664]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[1100]	rmse=0.019433	lr=0.078216
[1,0]<stderr>:INFO:root:Epoch[192] Batch[1200] Loss[2.564]
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[1200]	rmse=0.019437	lr=0.078066
[1,0]<stderr>:INFO:root:Epoch[192] Rank[0] Batch[1251]	Time cost=398.52	Train-metric=0.019438
[1,0]<stderr>:INFO:root:Epoch[192]	Speed: 3214.47 samples/sec
[1,0]<stderr>:INFO:root:Epoch[193] Batch[100] Loss[2.590]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[100]	rmse=0.019240	lr=0.077839
[1,0]<stderr>:INFO:root:Epoch[193] Batch[200] Loss[2.512]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[200]	rmse=0.019280	lr=0.077689
[1,0]<stderr>:INFO:root:Epoch[193] Batch[300] Loss[2.602]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[300]	rmse=0.019364	lr=0.077540
[1,0]<stderr>:INFO:root:Epoch[193] Batch[400] Loss[4.966]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[400]	rmse=0.019381	lr=0.077390
[1,0]<stderr>:INFO:root:Epoch[193] Batch[500] Loss[4.897]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[500]	rmse=0.019376	lr=0.077240
[1,0]<stderr>:INFO:root:Epoch[193] Batch[600] Loss[3.420]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[600]	rmse=0.019372	lr=0.077091
[1,0]<stderr>:INFO:root:Epoch[193] Batch[700] Loss[2.587]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[700]	rmse=0.019384	lr=0.076942
[1,0]<stderr>:INFO:root:Epoch[193] Batch[800] Loss[5.273]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[800]	rmse=0.019403	lr=0.076792
[1,0]<stderr>:INFO:root:Epoch[193] Batch[900] Loss[2.638]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[900]	rmse=0.019399	lr=0.076643
[1,0]<stderr>:INFO:root:Epoch[193] Batch[1000] Loss[3.403]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[1000]	rmse=0.019418	lr=0.076494
[1,0]<stderr>:INFO:root:Epoch[193] Batch[1100] Loss[2.616]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[1100]	rmse=0.019415	lr=0.076345
[1,0]<stderr>:INFO:root:Epoch[193] Batch[1200] Loss[3.782]
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[1200]	rmse=0.019404	lr=0.076196
[1,0]<stderr>:INFO:root:Epoch[193] Rank[0] Batch[1251]	Time cost=398.90	Train-metric=0.019417
[1,0]<stderr>:INFO:root:Epoch[193]	Speed: 3211.39 samples/sec
[1,0]<stderr>:INFO:root:Epoch[194] Batch[100] Loss[4.145]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[100]	rmse=0.019385	lr=0.075972
[1,0]<stderr>:INFO:root:Epoch[194] Batch[200] Loss[2.343]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[200]	rmse=0.019352	lr=0.075823
[1,0]<stderr>:INFO:root:Epoch[194] Batch[300] Loss[2.595]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[300]	rmse=0.019370	lr=0.075675
[1,0]<stderr>:INFO:root:Epoch[194] Batch[400] Loss[2.529]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[400]	rmse=0.019392	lr=0.075527
[1,0]<stderr>:INFO:root:Epoch[194] Batch[500] Loss[2.564]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[500]	rmse=0.019392	lr=0.075379
[1,0]<stderr>:INFO:root:Epoch[194] Batch[600] Loss[4.069]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[600]	rmse=0.019422	lr=0.075231
[1,0]<stderr>:INFO:root:Epoch[194] Batch[700] Loss[3.533]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[700]	rmse=0.019423	lr=0.075083
[1,0]<stderr>:INFO:root:Epoch[194] Batch[800] Loss[2.485]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[800]	rmse=0.019427	lr=0.074935
[1,0]<stderr>:INFO:root:Epoch[194] Batch[900] Loss[2.507]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[900]	rmse=0.019414	lr=0.074787
[1,0]<stderr>:INFO:root:Epoch[194] Batch[1000] Loss[2.988]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[1000]	rmse=0.019424	lr=0.074639
[1,0]<stderr>:INFO:root:Epoch[194] Batch[1100] Loss[2.484]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[1100]	rmse=0.019438	lr=0.074492
[1,0]<stderr>:INFO:root:Epoch[194] Batch[1200] Loss[3.202]
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[1200]	rmse=0.019434	lr=0.074345
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0] Batch[1251]	Time cost=398.74	Train-metric=0.019440
[1,0]<stderr>:INFO:root:Epoch[194]	Speed: 3212.65 samples/sec
[1,0]<stderr>:INFO:root:Epoch[194] Rank[0]	Validation-accuracy=0.694700	Validation-top_k_accuracy_5=0.894060
[1,0]<stderr>:INFO:root:Epoch[195] Batch[100] Loss[4.954]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[100]	rmse=0.019319	lr=0.074122
[1,0]<stderr>:INFO:root:Epoch[195] Batch[200] Loss[2.335]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[200]	rmse=0.019321	lr=0.073975
[1,0]<stderr>:INFO:root:Epoch[195] Batch[300] Loss[4.183]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[300]	rmse=0.019301	lr=0.073828
[1,0]<stderr>:INFO:root:Epoch[195] Batch[400] Loss[2.412]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[400]	rmse=0.019350	lr=0.073681
[1,0]<stderr>:INFO:root:Epoch[195] Batch[500] Loss[2.840]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[500]	rmse=0.019338	lr=0.073534
[1,0]<stderr>:INFO:root:Epoch[195] Batch[600] Loss[2.473]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[600]	rmse=0.019373	lr=0.073388
[1,0]<stderr>:INFO:root:Epoch[195] Batch[700] Loss[4.279]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[700]	rmse=0.019376	lr=0.073241
[1,0]<stderr>:INFO:root:Epoch[195] Batch[800] Loss[3.362]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[800]	rmse=0.019383	lr=0.073095
[1,0]<stderr>:INFO:root:Epoch[195] Batch[900] Loss[3.538]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[900]	rmse=0.019384	lr=0.072948
[1,0]<stderr>:INFO:root:Epoch[195] Batch[1000] Loss[2.687]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[1000]	rmse=0.019403	lr=0.072802
[1,0]<stderr>:INFO:root:Epoch[195] Batch[1100] Loss[2.312]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[1100]	rmse=0.019407	lr=0.072656
[1,0]<stderr>:INFO:root:Epoch[195] Batch[1200] Loss[3.666]
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[1200]	rmse=0.019412	lr=0.072510
[1,0]<stderr>:INFO:root:Epoch[195] Rank[0] Batch[1251]	Time cost=398.65	Train-metric=0.019418
[1,0]<stderr>:INFO:root:Epoch[195]	Speed: 3213.37 samples/sec
[1,0]<stderr>:INFO:root:Epoch[196] Batch[100] Loss[2.546]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[100]	rmse=0.019294	lr=0.072290
[1,0]<stderr>:INFO:root:Epoch[196] Batch[200] Loss[2.595]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[200]	rmse=0.019302	lr=0.072144
[1,0]<stderr>:INFO:root:Epoch[196] Batch[300] Loss[2.867]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[300]	rmse=0.019372	lr=0.071999
[1,0]<stderr>:INFO:root:Epoch[196] Batch[400] Loss[4.615]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[400]	rmse=0.019343	lr=0.071853
[1,0]<stderr>:INFO:root:Epoch[196] Batch[500] Loss[2.675]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[500]	rmse=0.019321	lr=0.071708
[1,0]<stderr>:INFO:root:Epoch[196] Batch[600] Loss[2.897]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[600]	rmse=0.019333	lr=0.071563
[1,0]<stderr>:INFO:root:Epoch[196] Batch[700] Loss[3.235]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[700]	rmse=0.019333	lr=0.071418
[1,0]<stderr>:INFO:root:Epoch[196] Batch[800] Loss[2.357]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[800]	rmse=0.019326	lr=0.071273
[1,0]<stderr>:INFO:root:Epoch[196] Batch[900] Loss[2.816]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[900]	rmse=0.019329	lr=0.071128
[1,0]<stderr>:INFO:root:Epoch[196] Batch[1000] Loss[2.453]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[1000]	rmse=0.019339	lr=0.070983
[1,0]<stderr>:INFO:root:Epoch[196] Batch[1100] Loss[5.062]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[1100]	rmse=0.019357	lr=0.070838
[1,0]<stderr>:INFO:root:Epoch[196] Batch[1200] Loss[2.772]
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[1200]	rmse=0.019377	lr=0.070694
[1,0]<stderr>:INFO:root:Epoch[196] Rank[0] Batch[1251]	Time cost=398.91	Train-metric=0.019373
[1,0]<stderr>:INFO:root:Epoch[196]	Speed: 3211.31 samples/sec
[1,0]<stderr>:INFO:root:Epoch[197] Batch[100] Loss[2.466]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[100]	rmse=0.019284	lr=0.070476
[1,0]<stderr>:INFO:root:Epoch[197] Batch[200] Loss[3.091]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[200]	rmse=0.019273	lr=0.070332
[1,0]<stderr>:INFO:root:Epoch[197] Batch[300] Loss[2.371]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[300]	rmse=0.019286	lr=0.070187
[1,0]<stderr>:INFO:root:Epoch[197] Batch[400] Loss[4.283]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[400]	rmse=0.019345	lr=0.070043
[1,0]<stderr>:INFO:root:Epoch[197] Batch[500] Loss[2.529]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[500]	rmse=0.019329	lr=0.069900
[1,0]<stderr>:INFO:root:Epoch[197] Batch[600] Loss[2.731]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[600]	rmse=0.019333	lr=0.069756
[1,0]<stderr>:INFO:root:Epoch[197] Batch[700] Loss[2.612]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[700]	rmse=0.019332	lr=0.069612
[1,0]<stderr>:INFO:root:Epoch[197] Batch[800] Loss[3.058]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[800]	rmse=0.019342	lr=0.069469
[1,0]<stderr>:INFO:root:Epoch[197] Batch[900] Loss[3.293]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[900]	rmse=0.019331	lr=0.069325
[1,0]<stderr>:INFO:root:Epoch[197] Batch[1000] Loss[2.630]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[1000]	rmse=0.019339	lr=0.069182
[1,0]<stderr>:INFO:root:Epoch[197] Batch[1100] Loss[4.400]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[1100]	rmse=0.019346	lr=0.069039
[1,0]<stderr>:INFO:root:Epoch[197] Batch[1200] Loss[2.890]
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[1200]	rmse=0.019334	lr=0.068896
[1,0]<stderr>:INFO:root:Epoch[197] Rank[0] Batch[1251]	Time cost=399.07	Train-metric=0.019336
[1,0]<stderr>:INFO:root:Epoch[197]	Speed: 3210.03 samples/sec
[1,0]<stderr>:INFO:root:Epoch[198] Batch[100] Loss[4.309]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[100]	rmse=0.019157	lr=0.068680
[1,0]<stderr>:INFO:root:Epoch[198] Batch[200] Loss[2.738]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[200]	rmse=0.019191	lr=0.068537
[1,0]<stderr>:INFO:root:Epoch[198] Batch[300] Loss[2.336]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[300]	rmse=0.019214	lr=0.068394
[1,0]<stderr>:INFO:root:Epoch[198] Batch[400] Loss[4.716]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[400]	rmse=0.019225	lr=0.068252
[1,0]<stderr>:INFO:root:Epoch[198] Batch[500] Loss[2.717]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[500]	rmse=0.019239	lr=0.068109
[1,0]<stderr>:INFO:root:Epoch[198] Batch[600] Loss[2.987]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[600]	rmse=0.019267	lr=0.067967
[1,0]<stderr>:INFO:root:Epoch[198] Batch[700] Loss[2.111]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[700]	rmse=0.019261	lr=0.067825
[1,0]<stderr>:INFO:root:Epoch[198] Batch[800] Loss[2.530]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[800]	rmse=0.019284	lr=0.067683
[1,0]<stderr>:INFO:root:Epoch[198] Batch[900] Loss[3.174]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[900]	rmse=0.019300	lr=0.067541
[1,0]<stderr>:INFO:root:Epoch[198] Batch[1000] Loss[3.435]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[1000]	rmse=0.019317	lr=0.067399
[1,0]<stderr>:INFO:root:Epoch[198] Batch[1100] Loss[3.687]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[1100]	rmse=0.019339	lr=0.067257
[1,0]<stderr>:INFO:root:Epoch[198] Batch[1200] Loss[2.418]
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[1200]	rmse=0.019346	lr=0.067116
[1,0]<stderr>:INFO:root:Epoch[198] Rank[0] Batch[1251]	Time cost=401.20	Train-metric=0.019339
[1,0]<stderr>:INFO:root:Epoch[198]	Speed: 3193.02 samples/sec
[1,0]<stderr>:INFO:root:Epoch[199] Batch[100] Loss[4.314]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[100]	rmse=0.019321	lr=0.066902
[1,0]<stderr>:INFO:root:Epoch[199] Batch[200] Loss[2.261]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[200]	rmse=0.019269	lr=0.066761
[1,0]<stderr>:INFO:root:Epoch[199] Batch[300] Loss[2.401]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[300]	rmse=0.019262	lr=0.066620
[1,0]<stderr>:INFO:root:Epoch[199] Batch[400] Loss[3.011]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[400]	rmse=0.019222	lr=0.066479
[1,0]<stderr>:INFO:root:Epoch[199] Batch[500] Loss[2.949]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[500]	rmse=0.019225	lr=0.066338
[1,0]<stderr>:INFO:root:Epoch[199] Batch[600] Loss[2.534]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[600]	rmse=0.019224	lr=0.066197
[1,0]<stderr>:INFO:root:Epoch[199] Batch[700] Loss[4.578]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[700]	rmse=0.019243	lr=0.066056
[1,0]<stderr>:INFO:root:Epoch[199] Batch[800] Loss[2.554]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[800]	rmse=0.019240	lr=0.065916
[1,0]<stderr>:INFO:root:Epoch[199] Batch[900] Loss[2.421]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[900]	rmse=0.019253	lr=0.065775
[1,0]<stderr>:INFO:root:Epoch[199] Batch[1000] Loss[3.723]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[1000]	rmse=0.019256	lr=0.065635
[1,0]<stderr>:INFO:root:Epoch[199] Batch[1100] Loss[3.081]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[1100]	rmse=0.019267	lr=0.065495
[1,0]<stderr>:INFO:root:Epoch[199] Batch[1200] Loss[2.621]
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[1200]	rmse=0.019271	lr=0.065355
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0] Batch[1251]	Time cost=401.49	Train-metric=0.019277
[1,0]<stderr>:INFO:root:Epoch[199]	Speed: 3190.66 samples/sec
[1,0]<stderr>:INFO:root:Epoch[199] Rank[0]	Validation-accuracy=0.698140	Validation-top_k_accuracy_5=0.894400
[1,0]<stderr>:INFO:root:Epoch[200] Batch[100] Loss[3.660]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[100]	rmse=0.019037	lr=0.065143
[1,0]<stderr>:INFO:root:Epoch[200] Batch[200] Loss[2.923]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[200]	rmse=0.019170	lr=0.065003
[1,0]<stderr>:INFO:root:Epoch[200] Batch[300] Loss[4.595]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[300]	rmse=0.019231	lr=0.064864
[1,0]<stderr>:INFO:root:Epoch[200] Batch[400] Loss[2.820]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[400]	rmse=0.019229	lr=0.064724
[1,0]<stderr>:INFO:root:Epoch[200] Batch[500] Loss[2.583]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[500]	rmse=0.019229	lr=0.064585
[1,0]<stderr>:INFO:root:Epoch[200] Batch[600] Loss[4.580]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[600]	rmse=0.019247	lr=0.064445
[1,0]<stderr>:INFO:root:Epoch[200] Batch[700] Loss[2.668]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[700]	rmse=0.019259	lr=0.064306
[1,0]<stderr>:INFO:root:Epoch[200] Batch[800] Loss[2.507]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[800]	rmse=0.019246	lr=0.064167
[1,0]<stderr>:INFO:root:Epoch[200] Batch[900] Loss[2.769]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[900]	rmse=0.019242	lr=0.064028
[1,0]<stderr>:INFO:root:Epoch[200] Batch[1000] Loss[2.520]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[1000]	rmse=0.019236	lr=0.063889
[1,0]<stderr>:INFO:root:Epoch[200] Batch[1100] Loss[3.217]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[1100]	rmse=0.019259	lr=0.063751
[1,0]<stderr>:INFO:root:Epoch[200] Batch[1200] Loss[2.475]
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[1200]	rmse=0.019259	lr=0.063612
[1,0]<stderr>:INFO:root:Epoch[200] Rank[0] Batch[1251]	Time cost=401.57	Train-metric=0.019256
[1,0]<stderr>:INFO:root:Epoch[200]	Speed: 3190.02 samples/sec
[1,0]<stderr>:INFO:root:Epoch[201] Batch[100] Loss[5.077]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[100]	rmse=0.019120	lr=0.063403
[1,0]<stderr>:INFO:root:Epoch[201] Batch[200] Loss[2.344]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[200]	rmse=0.019160	lr=0.063265
[1,0]<stderr>:INFO:root:Epoch[201] Batch[300] Loss[2.550]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[300]	rmse=0.019141	lr=0.063127
[1,0]<stderr>:INFO:root:Epoch[201] Batch[400] Loss[3.036]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[400]	rmse=0.019147	lr=0.062989
[1,0]<stderr>:INFO:root:Epoch[201] Batch[500] Loss[4.454]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[500]	rmse=0.019158	lr=0.062851
[1,0]<stderr>:INFO:root:Epoch[201] Batch[600] Loss[5.054]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[600]	rmse=0.019172	lr=0.062713
[1,0]<stderr>:INFO:root:Epoch[201] Batch[700] Loss[4.118]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[700]	rmse=0.019181	lr=0.062575
[1,0]<stderr>:INFO:root:Epoch[201] Batch[800] Loss[2.780]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[800]	rmse=0.019183	lr=0.062438
[1,0]<stderr>:INFO:root:Epoch[201] Batch[900] Loss[2.433]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[900]	rmse=0.019196	lr=0.062300
[1,0]<stderr>:INFO:root:Epoch[201] Batch[1000] Loss[2.607]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[1000]	rmse=0.019195	lr=0.062163
[1,0]<stderr>:INFO:root:Epoch[201] Batch[1100] Loss[2.214]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[1100]	rmse=0.019196	lr=0.062026
[1,0]<stderr>:INFO:root:Epoch[201] Batch[1200] Loss[4.973]
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[1200]	rmse=0.019191	lr=0.061889
[1,0]<stderr>:INFO:root:Epoch[201] Rank[0] Batch[1251]	Time cost=403.37	Train-metric=0.019193
[1,0]<stderr>:INFO:root:Epoch[201]	Speed: 3175.83 samples/sec
[1,0]<stderr>:INFO:root:Epoch[202] Batch[100] Loss[4.071]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[100]	rmse=0.019184	lr=0.061682
[1,0]<stderr>:INFO:root:Epoch[202] Batch[200] Loss[2.585]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[200]	rmse=0.019200	lr=0.061546
[1,0]<stderr>:INFO:root:Epoch[202] Batch[300] Loss[3.013]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[300]	rmse=0.019154	lr=0.061409
[1,0]<stderr>:INFO:root:Epoch[202] Batch[400] Loss[2.357]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[400]	rmse=0.019171	lr=0.061273
[1,0]<stderr>:INFO:root:Epoch[202] Batch[500] Loss[4.812]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[500]	rmse=0.019186	lr=0.061136
[1,0]<stderr>:INFO:root:Epoch[202] Batch[600] Loss[4.839]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[600]	rmse=0.019181	lr=0.061000
[1,0]<stderr>:INFO:root:Epoch[202] Batch[700] Loss[4.511]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[700]	rmse=0.019200	lr=0.060864
[1,0]<stderr>:INFO:root:Epoch[202] Batch[800] Loss[2.190]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[800]	rmse=0.019200	lr=0.060728
[1,0]<stderr>:INFO:root:Epoch[202] Batch[900] Loss[2.742]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[900]	rmse=0.019194	lr=0.060592
[1,0]<stderr>:INFO:root:Epoch[202] Batch[1000] Loss[2.703]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[1000]	rmse=0.019199	lr=0.060456
[1,0]<stderr>:INFO:root:Epoch[202] Batch[1100] Loss[4.907]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[1100]	rmse=0.019191	lr=0.060321
[1,0]<stderr>:INFO:root:Epoch[202] Batch[1200] Loss[4.531]
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[1200]	rmse=0.019206	lr=0.060185
[1,0]<stderr>:INFO:root:Epoch[202] Rank[0] Batch[1251]	Time cost=400.35	Train-metric=0.019205
[1,0]<stderr>:INFO:root:Epoch[202]	Speed: 3199.75 samples/sec
[1,0]<stderr>:INFO:root:Epoch[203] Batch[100] Loss[2.534]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[100]	rmse=0.019040	lr=0.059981
[1,0]<stderr>:INFO:root:Epoch[203] Batch[200] Loss[2.482]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[200]	rmse=0.019020	lr=0.059846
[1,0]<stderr>:INFO:root:Epoch[203] Batch[300] Loss[3.949]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[300]	rmse=0.019026	lr=0.059711
[1,0]<stderr>:INFO:root:Epoch[203] Batch[400] Loss[2.579]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[400]	rmse=0.019031	lr=0.059576
[1,0]<stderr>:INFO:root:Epoch[203] Batch[500] Loss[2.678]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[500]	rmse=0.019046	lr=0.059441
[1,0]<stderr>:INFO:root:Epoch[203] Batch[600] Loss[2.784]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[600]	rmse=0.019060	lr=0.059306
[1,0]<stderr>:INFO:root:Epoch[203] Batch[700] Loss[2.569]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[700]	rmse=0.019076	lr=0.059172
[1,0]<stderr>:INFO:root:Epoch[203] Batch[800] Loss[3.587]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[800]	rmse=0.019127	lr=0.059037
[1,0]<stderr>:INFO:root:Epoch[203] Batch[900] Loss[4.768]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[900]	rmse=0.019140	lr=0.058903
[1,0]<stderr>:INFO:root:Epoch[203] Batch[1000] Loss[4.202]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[1000]	rmse=0.019150	lr=0.058769
[1,0]<stderr>:INFO:root:Epoch[203] Batch[1100] Loss[5.072]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[1100]	rmse=0.019153	lr=0.058635
[1,0]<stderr>:INFO:root:Epoch[203] Batch[1200] Loss[3.913]
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[1200]	rmse=0.019166	lr=0.058501
[1,0]<stderr>:INFO:root:Epoch[203] Rank[0] Batch[1251]	Time cost=399.99	Train-metric=0.019160
[1,0]<stderr>:INFO:root:Epoch[203]	Speed: 3202.66 samples/sec
[1,0]<stderr>:INFO:root:Epoch[204] Batch[100] Loss[2.009]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[100]	rmse=0.018992	lr=0.058299
[1,0]<stderr>:INFO:root:Epoch[204] Batch[200] Loss[2.497]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[200]	rmse=0.019015	lr=0.058165
[1,0]<stderr>:INFO:root:Epoch[204] Batch[300] Loss[2.446]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[300]	rmse=0.019066	lr=0.058032
[1,0]<stderr>:INFO:root:Epoch[204] Batch[400] Loss[4.072]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[400]	rmse=0.019108	lr=0.057899
[1,0]<stderr>:INFO:root:Epoch[204] Batch[500] Loss[2.659]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[500]	rmse=0.019104	lr=0.057765
[1,0]<stderr>:INFO:root:Epoch[204] Batch[600] Loss[3.047]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[600]	rmse=0.019124	lr=0.057632
[1,0]<stderr>:INFO:root:Epoch[204] Batch[700] Loss[2.808]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[700]	rmse=0.019137	lr=0.057499
[1,0]<stderr>:INFO:root:Epoch[204] Batch[800] Loss[2.880]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[800]	rmse=0.019138	lr=0.057367
[1,0]<stderr>:INFO:root:Epoch[204] Batch[900] Loss[2.547]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[900]	rmse=0.019152	lr=0.057234
[1,0]<stderr>:INFO:root:Epoch[204] Batch[1000] Loss[3.037]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[1000]	rmse=0.019151	lr=0.057101
[1,0]<stderr>:INFO:root:Epoch[204] Batch[1100] Loss[2.323]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[1100]	rmse=0.019161	lr=0.056969
[1,0]<stderr>:INFO:root:Epoch[204] Batch[1200] Loss[2.647]
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[1200]	rmse=0.019169	lr=0.056837
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0] Batch[1251]	Time cost=398.85	Train-metric=0.019168
[1,0]<stderr>:INFO:root:Epoch[204]	Speed: 3211.82 samples/sec
[1,0]<stderr>:INFO:root:Epoch[204] Rank[0]	Validation-accuracy=0.702940	Validation-top_k_accuracy_5=0.898860
[1,0]<stderr>:INFO:root:Epoch[205] Batch[100] Loss[4.613]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[100]	rmse=0.019023	lr=0.056637
[1,0]<stderr>:INFO:root:Epoch[205] Batch[200] Loss[3.752]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[200]	rmse=0.019057	lr=0.056505
[1,0]<stderr>:INFO:root:Epoch[205] Batch[300] Loss[3.026]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[300]	rmse=0.019050	lr=0.056373
[1,0]<stderr>:INFO:root:Epoch[205] Batch[400] Loss[4.325]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[400]	rmse=0.019037	lr=0.056242
[1,0]<stderr>:INFO:root:Epoch[205] Batch[500] Loss[3.576]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[500]	rmse=0.019024	lr=0.056110
[1,0]<stderr>:INFO:root:Epoch[205] Batch[600] Loss[2.414]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[600]	rmse=0.019029	lr=0.055979
[1,0]<stderr>:INFO:root:Epoch[205] Batch[700] Loss[2.709]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[700]	rmse=0.019038	lr=0.055847
[1,0]<stderr>:INFO:root:Epoch[205] Batch[800] Loss[4.025]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[800]	rmse=0.019066	lr=0.055716
[1,0]<stderr>:INFO:root:Epoch[205] Batch[900] Loss[4.236]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[900]	rmse=0.019064	lr=0.055585
[1,0]<stderr>:INFO:root:Epoch[205] Batch[1000] Loss[2.516]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[1000]	rmse=0.019088	lr=0.055454
[1,0]<stderr>:INFO:root:Epoch[205] Batch[1100] Loss[4.850]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[1100]	rmse=0.019100	lr=0.055323
[1,0]<stderr>:INFO:root:Epoch[205] Batch[1200] Loss[2.387]
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[1200]	rmse=0.019108	lr=0.055192
[1,0]<stderr>:INFO:root:Epoch[205] Rank[0] Batch[1251]	Time cost=398.60	Train-metric=0.019105
[1,0]<stderr>:INFO:root:Epoch[205]	Speed: 3213.79 samples/sec
[1,0]<stderr>:INFO:root:Epoch[206] Batch[100] Loss[2.559]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[100]	rmse=0.019131	lr=0.054995
[1,0]<stderr>:INFO:root:Epoch[206] Batch[200] Loss[3.618]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[200]	rmse=0.019129	lr=0.054865
[1,0]<stderr>:INFO:root:Epoch[206] Batch[300] Loss[2.651]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[300]	rmse=0.019152	lr=0.054735
[1,0]<stderr>:INFO:root:Epoch[206] Batch[400] Loss[4.035]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[400]	rmse=0.019132	lr=0.054605
[1,0]<stderr>:INFO:root:Epoch[206] Batch[500] Loss[3.025]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[500]	rmse=0.019101	lr=0.054475
[1,0]<stderr>:INFO:root:Epoch[206] Batch[600] Loss[2.491]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[600]	rmse=0.019095	lr=0.054345
[1,0]<stderr>:INFO:root:Epoch[206] Batch[700] Loss[2.475]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[700]	rmse=0.019090	lr=0.054215
[1,0]<stderr>:INFO:root:Epoch[206] Batch[800] Loss[2.530]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[800]	rmse=0.019093	lr=0.054086
[1,0]<stderr>:INFO:root:Epoch[206] Batch[900] Loss[2.591]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[900]	rmse=0.019088	lr=0.053956
[1,0]<stderr>:INFO:root:Epoch[206] Batch[1000] Loss[2.541]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[1000]	rmse=0.019091	lr=0.053827
[1,0]<stderr>:INFO:root:Epoch[206] Batch[1100] Loss[2.920]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[1100]	rmse=0.019096	lr=0.053698
[1,0]<stderr>:INFO:root:Epoch[206] Batch[1200] Loss[2.320]
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[1200]	rmse=0.019108	lr=0.053568
[1,0]<stderr>:INFO:root:Epoch[206] Rank[0] Batch[1251]	Time cost=398.41	Train-metric=0.019105
[1,0]<stderr>:INFO:root:Epoch[206]	Speed: 3215.32 samples/sec
[1,0]<stderr>:INFO:root:Epoch[207] Batch[100] Loss[2.685]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[100]	rmse=0.019015	lr=0.053374
[1,0]<stderr>:INFO:root:Epoch[207] Batch[200] Loss[2.963]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[200]	rmse=0.019043	lr=0.053245
[1,0]<stderr>:INFO:root:Epoch[207] Batch[300] Loss[2.579]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[300]	rmse=0.019063	lr=0.053117
[1,0]<stderr>:INFO:root:Epoch[207] Batch[400] Loss[2.565]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[400]	rmse=0.019053	lr=0.052988
[1,0]<stderr>:INFO:root:Epoch[207] Batch[500] Loss[2.256]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[500]	rmse=0.019055	lr=0.052860
[1,0]<stderr>:INFO:root:Epoch[207] Batch[600] Loss[2.415]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[600]	rmse=0.019035	lr=0.052732
[1,0]<stderr>:INFO:root:Epoch[207] Batch[700] Loss[2.555]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[700]	rmse=0.019043	lr=0.052603
[1,0]<stderr>:INFO:root:Epoch[207] Batch[800] Loss[4.473]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[800]	rmse=0.019050	lr=0.052476
[1,0]<stderr>:INFO:root:Epoch[207] Batch[900] Loss[4.691]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[900]	rmse=0.019053	lr=0.052348
[1,0]<stderr>:INFO:root:Epoch[207] Batch[1000] Loss[2.247]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[1000]	rmse=0.019051	lr=0.052220
[1,0]<stderr>:INFO:root:Epoch[207] Batch[1100] Loss[2.920]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[1100]	rmse=0.019061	lr=0.052093
[1,0]<stderr>:INFO:root:Epoch[207] Batch[1200] Loss[2.460]
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[1200]	rmse=0.019070	lr=0.051965
[1,0]<stderr>:INFO:root:Epoch[207] Rank[0] Batch[1251]	Time cost=398.95	Train-metric=0.019075
[1,0]<stderr>:INFO:root:Epoch[207]	Speed: 3210.97 samples/sec
[1,0]<stderr>:INFO:root:Epoch[208] Batch[100] Loss[2.392]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[100]	rmse=0.018859	lr=0.051773
[1,0]<stderr>:INFO:root:Epoch[208] Batch[200] Loss[2.609]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[200]	rmse=0.018918	lr=0.051646
[1,0]<stderr>:INFO:root:Epoch[208] Batch[300] Loss[2.668]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[300]	rmse=0.018993	lr=0.051519
[1,0]<stderr>:INFO:root:Epoch[208] Batch[400] Loss[2.062]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[400]	rmse=0.019008	lr=0.051392
[1,0]<stderr>:INFO:root:Epoch[208] Batch[500] Loss[2.659]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[500]	rmse=0.019064	lr=0.051265
[1,0]<stderr>:INFO:root:Epoch[208] Batch[600] Loss[3.645]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[600]	rmse=0.019068	lr=0.051139
[1,0]<stderr>:INFO:root:Epoch[208] Batch[700] Loss[2.408]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[700]	rmse=0.019063	lr=0.051013
[1,0]<stderr>:INFO:root:Epoch[208] Batch[800] Loss[2.772]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[800]	rmse=0.019076	lr=0.050886
[1,0]<stderr>:INFO:root:Epoch[208] Batch[900] Loss[4.364]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[900]	rmse=0.019075	lr=0.050760
[1,0]<stderr>:INFO:root:Epoch[208] Batch[1000] Loss[2.616]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[1000]	rmse=0.019079	lr=0.050634
[1,0]<stderr>:INFO:root:Epoch[208] Batch[1100] Loss[2.204]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[1100]	rmse=0.019077	lr=0.050508
[1,0]<stderr>:INFO:root:Epoch[208] Batch[1200] Loss[4.946]
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[1200]	rmse=0.019073	lr=0.050383
[1,0]<stderr>:INFO:root:Epoch[208] Rank[0] Batch[1251]	Time cost=399.34	Train-metric=0.019069
[1,0]<stderr>:INFO:root:Epoch[208]	Speed: 3207.85 samples/sec
[1,0]<stderr>:INFO:root:Epoch[209] Batch[100] Loss[3.771]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[100]	rmse=0.018878	lr=0.050193
[1,0]<stderr>:INFO:root:Epoch[209] Batch[200] Loss[2.349]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[200]	rmse=0.018892	lr=0.050067
[1,0]<stderr>:INFO:root:Epoch[209] Batch[300] Loss[2.568]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[300]	rmse=0.018921	lr=0.049942
[1,0]<stderr>:INFO:root:Epoch[209] Batch[400] Loss[2.784]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[400]	rmse=0.018904	lr=0.049817
[1,0]<stderr>:INFO:root:Epoch[209] Batch[500] Loss[2.236]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[500]	rmse=0.018917	lr=0.049692
[1,0]<stderr>:INFO:root:Epoch[209] Batch[600] Loss[4.509]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[600]	rmse=0.018939	lr=0.049567
[1,0]<stderr>:INFO:root:Epoch[209] Batch[700] Loss[2.739]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[700]	rmse=0.018947	lr=0.049443
[1,0]<stderr>:INFO:root:Epoch[209] Batch[800] Loss[3.449]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[800]	rmse=0.018971	lr=0.049318
[1,0]<stderr>:INFO:root:Epoch[209] Batch[900] Loss[2.229]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[900]	rmse=0.018979	lr=0.049193
[1,0]<stderr>:INFO:root:Epoch[209] Batch[1000] Loss[2.527]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[1000]	rmse=0.018992	lr=0.049069
[1,0]<stderr>:INFO:root:Epoch[209] Batch[1100] Loss[4.521]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[1100]	rmse=0.019002	lr=0.048945
[1,0]<stderr>:INFO:root:Epoch[209] Batch[1200] Loss[2.436]
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[1200]	rmse=0.019001	lr=0.048821
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0] Batch[1251]	Time cost=399.51	Train-metric=0.019016
[1,0]<stderr>:INFO:root:Epoch[209]	Speed: 3206.47 samples/sec
[1,0]<stderr>:INFO:root:Epoch[209] Rank[0]	Validation-accuracy=0.711000	Validation-top_k_accuracy_5=0.902380
[1,0]<stderr>:INFO:root:Epoch[210] Batch[100] Loss[4.221]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[100]	rmse=0.018878	lr=0.048634
[1,0]<stderr>:INFO:root:Epoch[210] Batch[200] Loss[2.171]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[200]	rmse=0.018900	lr=0.048510
[1,0]<stderr>:INFO:root:Epoch[210] Batch[300] Loss[2.783]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[300]	rmse=0.018893	lr=0.048387
[1,0]<stderr>:INFO:root:Epoch[210] Batch[400] Loss[2.627]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[400]	rmse=0.018923	lr=0.048263
[1,0]<stderr>:INFO:root:Epoch[210] Batch[500] Loss[2.743]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[500]	rmse=0.018950	lr=0.048140
[1,0]<stderr>:INFO:root:Epoch[210] Batch[600] Loss[2.319]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[600]	rmse=0.018975	lr=0.048017
[1,0]<stderr>:INFO:root:Epoch[210] Batch[700] Loss[2.292]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[700]	rmse=0.019000	lr=0.047894
[1,0]<stderr>:INFO:root:Epoch[210] Batch[800] Loss[4.869]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[800]	rmse=0.018993	lr=0.047771
[1,0]<stderr>:INFO:root:Epoch[210] Batch[900] Loss[2.514]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[900]	rmse=0.018995	lr=0.047648
[1,0]<stderr>:INFO:root:Epoch[210] Batch[1000] Loss[4.834]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[1000]	rmse=0.019011	lr=0.047525
[1,0]<stderr>:INFO:root:Epoch[210] Batch[1100] Loss[2.560]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[1100]	rmse=0.019011	lr=0.047403
[1,0]<stderr>:INFO:root:Epoch[210] Batch[1200] Loss[2.788]
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[1200]	rmse=0.019027	lr=0.047280
[1,0]<stderr>:INFO:root:Epoch[210] Rank[0] Batch[1251]	Time cost=413.22	Train-metric=0.019029
[1,0]<stderr>:INFO:root:Epoch[210]	Speed: 3100.08 samples/sec
[1,0]<stderr>:INFO:root:Epoch[211] Batch[100] Loss[2.480]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[100]	rmse=0.018746	lr=0.047096
[1,0]<stderr>:INFO:root:Epoch[211] Batch[200] Loss[2.574]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[200]	rmse=0.018833	lr=0.046974
[1,0]<stderr>:INFO:root:Epoch[211] Batch[300] Loss[2.250]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[300]	rmse=0.018877	lr=0.046852
[1,0]<stderr>:INFO:root:Epoch[211] Batch[400] Loss[5.086]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[400]	rmse=0.018873	lr=0.046730
[1,0]<stderr>:INFO:root:Epoch[211] Batch[500] Loss[3.953]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[500]	rmse=0.018869	lr=0.046609
[1,0]<stderr>:INFO:root:Epoch[211] Batch[600] Loss[2.175]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[600]	rmse=0.018879	lr=0.046487
[1,0]<stderr>:INFO:root:Epoch[211] Batch[700] Loss[5.018]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[700]	rmse=0.018885	lr=0.046366
[1,0]<stderr>:INFO:root:Epoch[211] Batch[800] Loss[2.366]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[800]	rmse=0.018909	lr=0.046245
[1,0]<stderr>:INFO:root:Epoch[211] Batch[900] Loss[4.290]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[900]	rmse=0.018917	lr=0.046124
[1,0]<stderr>:INFO:root:Epoch[211] Batch[1000] Loss[2.360]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[1000]	rmse=0.018923	lr=0.046003
[1,0]<stderr>:INFO:root:Epoch[211] Batch[1100] Loss[4.552]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[1100]	rmse=0.018928	lr=0.045882
[1,0]<stderr>:INFO:root:Epoch[211] Batch[1200] Loss[2.790]
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[1200]	rmse=0.018929	lr=0.045762
[1,0]<stderr>:INFO:root:Epoch[211] Rank[0] Batch[1251]	Time cost=400.01	Train-metric=0.018940
[1,0]<stderr>:INFO:root:Epoch[211]	Speed: 3202.46 samples/sec
[1,0]<stderr>:INFO:root:Epoch[212] Batch[100] Loss[3.054]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[100]	rmse=0.018807	lr=0.045580
[1,0]<stderr>:INFO:root:Epoch[212] Batch[200] Loss[2.342]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[200]	rmse=0.018833	lr=0.045459
[1,0]<stderr>:INFO:root:Epoch[212] Batch[300] Loss[3.573]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[300]	rmse=0.018843	lr=0.045339
[1,0]<stderr>:INFO:root:Epoch[212] Batch[400] Loss[4.786]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[400]	rmse=0.018900	lr=0.045219
[1,0]<stderr>:INFO:root:Epoch[212] Batch[500] Loss[4.406]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[500]	rmse=0.018918	lr=0.045099
[1,0]<stderr>:INFO:root:Epoch[212] Batch[600] Loss[2.388]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[600]	rmse=0.018940	lr=0.044980
[1,0]<stderr>:INFO:root:Epoch[212] Batch[700] Loss[3.410]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[700]	rmse=0.018931	lr=0.044860
[1,0]<stderr>:INFO:root:Epoch[212] Batch[800] Loss[4.162]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[800]	rmse=0.018938	lr=0.044741
[1,0]<stderr>:INFO:root:Epoch[212] Batch[900] Loss[2.806]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[900]	rmse=0.018940	lr=0.044621
[1,0]<stderr>:INFO:root:Epoch[212] Batch[1000] Loss[2.431]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[1000]	rmse=0.018931	lr=0.044502
[1,0]<stderr>:INFO:root:Epoch[212] Batch[1100] Loss[2.635]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[1100]	rmse=0.018931	lr=0.044383
[1,0]<stderr>:INFO:root:Epoch[212] Batch[1200] Loss[3.911]
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[1200]	rmse=0.018933	lr=0.044264
[1,0]<stderr>:INFO:root:Epoch[212] Rank[0] Batch[1251]	Time cost=399.09	Train-metric=0.018931
[1,0]<stderr>:INFO:root:Epoch[212]	Speed: 3209.84 samples/sec
[1,0]<stderr>:INFO:root:Epoch[213] Batch[100] Loss[2.937]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[100]	rmse=0.018873	lr=0.044085
[1,0]<stderr>:INFO:root:Epoch[213] Batch[200] Loss[5.110]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[200]	rmse=0.018819	lr=0.043966
[1,0]<stderr>:INFO:root:Epoch[213] Batch[300] Loss[2.479]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[300]	rmse=0.018812	lr=0.043848
[1,0]<stderr>:INFO:root:Epoch[213] Batch[400] Loss[3.732]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[400]	rmse=0.018812	lr=0.043730
[1,0]<stderr>:INFO:root:Epoch[213] Batch[500] Loss[2.553]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[500]	rmse=0.018833	lr=0.043612
[1,0]<stderr>:INFO:root:Epoch[213] Batch[600] Loss[3.374]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[600]	rmse=0.018847	lr=0.043494
[1,0]<stderr>:INFO:root:Epoch[213] Batch[700] Loss[4.163]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[700]	rmse=0.018842	lr=0.043376
[1,0]<stderr>:INFO:root:Epoch[213] Batch[800] Loss[2.677]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[800]	rmse=0.018851	lr=0.043258
[1,0]<stderr>:INFO:root:Epoch[213] Batch[900] Loss[3.478]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[900]	rmse=0.018854	lr=0.043141
[1,0]<stderr>:INFO:root:Epoch[213] Batch[1000] Loss[2.510]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[1000]	rmse=0.018885	lr=0.043023
[1,0]<stderr>:INFO:root:Epoch[213] Batch[1100] Loss[2.637]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[1100]	rmse=0.018881	lr=0.042906
[1,0]<stderr>:INFO:root:Epoch[213] Batch[1200] Loss[4.795]
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[1200]	rmse=0.018886	lr=0.042789
[1,0]<stderr>:INFO:root:Epoch[213] Rank[0] Batch[1251]	Time cost=399.52	Train-metric=0.018890
[1,0]<stderr>:INFO:root:Epoch[213]	Speed: 3206.40 samples/sec
[1,0]<stderr>:INFO:root:Epoch[214] Batch[100] Loss[2.960]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[100]	rmse=0.018789	lr=0.042612
[1,0]<stderr>:INFO:root:Epoch[214] Batch[200] Loss[2.231]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[200]	rmse=0.018736	lr=0.042495
[1,0]<stderr>:INFO:root:Epoch[214] Batch[300] Loss[4.058]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[300]	rmse=0.018857	lr=0.042379
[1,0]<stderr>:INFO:root:Epoch[214] Batch[400] Loss[2.409]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[400]	rmse=0.018847	lr=0.042262
[1,0]<stderr>:INFO:root:Epoch[214] Batch[500] Loss[2.238]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[500]	rmse=0.018806	lr=0.042146
[1,0]<stderr>:INFO:root:Epoch[214] Batch[600] Loss[2.547]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[600]	rmse=0.018824	lr=0.042030
[1,0]<stderr>:INFO:root:Epoch[214] Batch[700] Loss[2.415]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[700]	rmse=0.018812	lr=0.041914
[1,0]<stderr>:INFO:root:Epoch[214] Batch[800] Loss[2.924]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[800]	rmse=0.018815	lr=0.041798
[1,0]<stderr>:INFO:root:Epoch[214] Batch[900] Loss[2.308]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[900]	rmse=0.018806	lr=0.041682
[1,0]<stderr>:INFO:root:Epoch[214] Batch[1000] Loss[2.437]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[1000]	rmse=0.018813	lr=0.041566
[1,0]<stderr>:INFO:root:Epoch[214] Batch[1100] Loss[3.914]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[1100]	rmse=0.018820	lr=0.041451
[1,0]<stderr>:INFO:root:Epoch[214] Batch[1200] Loss[2.589]
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[1200]	rmse=0.018816	lr=0.041335
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0] Batch[1251]	Time cost=399.01	Train-metric=0.018829
[1,0]<stderr>:INFO:root:Epoch[214]	Speed: 3210.52 samples/sec
[1,0]<stderr>:INFO:root:Epoch[214] Rank[0]	Validation-accuracy=0.714180	Validation-top_k_accuracy_5=0.902960
[1,0]<stderr>:INFO:root:Epoch[215] Batch[100] Loss[2.410]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[100]	rmse=0.018812	lr=0.041161
[1,0]<stderr>:INFO:root:Epoch[215] Batch[200] Loss[3.049]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[200]	rmse=0.018798	lr=0.041046
[1,0]<stderr>:INFO:root:Epoch[215] Batch[300] Loss[2.542]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[300]	rmse=0.018792	lr=0.040931
[1,0]<stderr>:INFO:root:Epoch[215] Batch[400] Loss[4.849]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[400]	rmse=0.018767	lr=0.040817
[1,0]<stderr>:INFO:root:Epoch[215] Batch[500] Loss[2.347]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[500]	rmse=0.018790	lr=0.040702
[1,0]<stderr>:INFO:root:Epoch[215] Batch[600] Loss[2.391]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[600]	rmse=0.018795	lr=0.040588
[1,0]<stderr>:INFO:root:Epoch[215] Batch[700] Loss[2.351]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[700]	rmse=0.018811	lr=0.040473
[1,0]<stderr>:INFO:root:Epoch[215] Batch[800] Loss[2.467]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[800]	rmse=0.018808	lr=0.040359
[1,0]<stderr>:INFO:root:Epoch[215] Batch[900] Loss[2.492]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[900]	rmse=0.018805	lr=0.040245
[1,0]<stderr>:INFO:root:Epoch[215] Batch[1000] Loss[2.965]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[1000]	rmse=0.018814	lr=0.040131
[1,0]<stderr>:INFO:root:Epoch[215] Batch[1100] Loss[3.447]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[1100]	rmse=0.018812	lr=0.040018
[1,0]<stderr>:INFO:root:Epoch[215] Batch[1200] Loss[4.862]
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[1200]	rmse=0.018819	lr=0.039904
[1,0]<stderr>:INFO:root:Epoch[215] Rank[0] Batch[1251]	Time cost=398.32	Train-metric=0.018827
[1,0]<stderr>:INFO:root:Epoch[215]	Speed: 3216.06 samples/sec
[1,0]<stderr>:INFO:root:Epoch[216] Batch[100] Loss[4.603]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[100]	rmse=0.018644	lr=0.039733
[1,0]<stderr>:INFO:root:Epoch[216] Batch[200] Loss[2.370]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[200]	rmse=0.018657	lr=0.039620
[1,0]<stderr>:INFO:root:Epoch[216] Batch[300] Loss[2.264]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[300]	rmse=0.018677	lr=0.039507
[1,0]<stderr>:INFO:root:Epoch[216] Batch[400] Loss[2.876]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[400]	rmse=0.018743	lr=0.039394
[1,0]<stderr>:INFO:root:Epoch[216] Batch[500] Loss[4.250]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[500]	rmse=0.018758	lr=0.039281
[1,0]<stderr>:INFO:root:Epoch[216] Batch[600] Loss[2.320]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[600]	rmse=0.018773	lr=0.039168
[1,0]<stderr>:INFO:root:Epoch[216] Batch[700] Loss[3.586]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[700]	rmse=0.018762	lr=0.039056
[1,0]<stderr>:INFO:root:Epoch[216] Batch[800] Loss[2.403]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[800]	rmse=0.018765	lr=0.038943
[1,0]<stderr>:INFO:root:Epoch[216] Batch[900] Loss[3.370]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[900]	rmse=0.018786	lr=0.038831
[1,0]<stderr>:INFO:root:Epoch[216] Batch[1000] Loss[2.448]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[1000]	rmse=0.018781	lr=0.038719
[1,0]<stderr>:INFO:root:Epoch[216] Batch[1100] Loss[2.539]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[1100]	rmse=0.018778	lr=0.038607
[1,0]<stderr>:INFO:root:Epoch[216] Batch[1200] Loss[2.407]
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[1200]	rmse=0.018784	lr=0.038495
[1,0]<stderr>:INFO:root:Epoch[216] Rank[0] Batch[1251]	Time cost=398.81	Train-metric=0.018779
[1,0]<stderr>:INFO:root:Epoch[216]	Speed: 3212.14 samples/sec
[1,0]<stderr>:INFO:root:Epoch[217] Batch[100] Loss[2.352]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[100]	rmse=0.018843	lr=0.038327
[1,0]<stderr>:INFO:root:Epoch[217] Batch[200] Loss[4.636]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[200]	rmse=0.018767	lr=0.038216
[1,0]<stderr>:INFO:root:Epoch[217] Batch[300] Loss[2.307]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[300]	rmse=0.018808	lr=0.038104
[1,0]<stderr>:INFO:root:Epoch[217] Batch[400] Loss[2.378]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[400]	rmse=0.018801	lr=0.037993
[1,0]<stderr>:INFO:root:Epoch[217] Batch[500] Loss[2.729]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[500]	rmse=0.018773	lr=0.037882
[1,0]<stderr>:INFO:root:Epoch[217] Batch[600] Loss[2.686]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[600]	rmse=0.018755	lr=0.037771
[1,0]<stderr>:INFO:root:Epoch[217] Batch[700] Loss[2.550]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[700]	rmse=0.018756	lr=0.037661
[1,0]<stderr>:INFO:root:Epoch[217] Batch[800] Loss[2.190]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[800]	rmse=0.018763	lr=0.037550
[1,0]<stderr>:INFO:root:Epoch[217] Batch[900] Loss[2.743]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[900]	rmse=0.018770	lr=0.037440
[1,0]<stderr>:INFO:root:Epoch[217] Batch[1000] Loss[2.368]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[1000]	rmse=0.018761	lr=0.037329
[1,0]<stderr>:INFO:root:Epoch[217] Batch[1100] Loss[5.036]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[1100]	rmse=0.018763	lr=0.037219
[1,0]<stderr>:INFO:root:Epoch[217] Batch[1200] Loss[2.303]
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[1200]	rmse=0.018778	lr=0.037109
[1,0]<stderr>:INFO:root:Epoch[217] Rank[0] Batch[1251]	Time cost=399.67	Train-metric=0.018780
[1,0]<stderr>:INFO:root:Epoch[217]	Speed: 3205.24 samples/sec
[1,0]<stderr>:INFO:root:Epoch[218] Batch[100] Loss[2.316]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[100]	rmse=0.018654	lr=0.036944
[1,0]<stderr>:INFO:root:Epoch[218] Batch[200] Loss[3.415]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[200]	rmse=0.018691	lr=0.036834
[1,0]<stderr>:INFO:root:Epoch[218] Batch[300] Loss[3.525]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[300]	rmse=0.018691	lr=0.036725
[1,0]<stderr>:INFO:root:Epoch[218] Batch[400] Loss[2.207]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[400]	rmse=0.018705	lr=0.036615
[1,0]<stderr>:INFO:root:Epoch[218] Batch[500] Loss[2.664]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[500]	rmse=0.018717	lr=0.036506
[1,0]<stderr>:INFO:root:Epoch[218] Batch[600] Loss[2.950]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[600]	rmse=0.018721	lr=0.036397
[1,0]<stderr>:INFO:root:Epoch[218] Batch[700] Loss[2.517]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[700]	rmse=0.018725	lr=0.036288
[1,0]<stderr>:INFO:root:Epoch[218] Batch[800] Loss[2.427]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[800]	rmse=0.018736	lr=0.036180
[1,0]<stderr>:INFO:root:Epoch[218] Batch[900] Loss[2.469]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[900]	rmse=0.018745	lr=0.036071
[1,0]<stderr>:INFO:root:Epoch[218] Batch[1000] Loss[2.339]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[1000]	rmse=0.018754	lr=0.035963
[1,0]<stderr>:INFO:root:Epoch[218] Batch[1100] Loss[2.251]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[1100]	rmse=0.018763	lr=0.035854
[1,0]<stderr>:INFO:root:Epoch[218] Batch[1200] Loss[2.343]
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[1200]	rmse=0.018769	lr=0.035746
[1,0]<stderr>:INFO:root:Epoch[218] Rank[0] Batch[1251]	Time cost=398.72	Train-metric=0.018765
[1,0]<stderr>:INFO:root:Epoch[218]	Speed: 3212.81 samples/sec
[1,0]<stderr>:INFO:root:Epoch[219] Batch[100] Loss[4.922]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[100]	rmse=0.018665	lr=0.035583
[1,0]<stderr>:INFO:root:Epoch[219] Batch[200] Loss[2.223]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[200]	rmse=0.018693	lr=0.035476
[1,0]<stderr>:INFO:root:Epoch[219] Batch[300] Loss[3.424]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[300]	rmse=0.018679	lr=0.035368
[1,0]<stderr>:INFO:root:Epoch[219] Batch[400] Loss[2.113]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[400]	rmse=0.018698	lr=0.035260
[1,0]<stderr>:INFO:root:Epoch[219] Batch[500] Loss[2.137]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[500]	rmse=0.018669	lr=0.035153
[1,0]<stderr>:INFO:root:Epoch[219] Batch[600] Loss[3.436]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[600]	rmse=0.018692	lr=0.035046
[1,0]<stderr>:INFO:root:Epoch[219] Batch[700] Loss[2.354]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[700]	rmse=0.018685	lr=0.034939
[1,0]<stderr>:INFO:root:Epoch[219] Batch[800] Loss[2.111]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[800]	rmse=0.018675	lr=0.034832
[1,0]<stderr>:INFO:root:Epoch[219] Batch[900] Loss[4.927]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[900]	rmse=0.018685	lr=0.034725
[1,0]<stderr>:INFO:root:Epoch[219] Batch[1000] Loss[2.469]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[1000]	rmse=0.018684	lr=0.034619
[1,0]<stderr>:INFO:root:Epoch[219] Batch[1100] Loss[4.399]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[1100]	rmse=0.018701	lr=0.034512
[1,0]<stderr>:INFO:root:Epoch[219] Batch[1200] Loss[2.396]
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[1200]	rmse=0.018701	lr=0.034406
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0] Batch[1251]	Time cost=399.47	Train-metric=0.018701
[1,0]<stderr>:INFO:root:Epoch[219]	Speed: 3206.82 samples/sec
[1,0]<stderr>:INFO:root:Epoch[219] Rank[0]	Validation-accuracy=0.719280	Validation-top_k_accuracy_5=0.905940
[1,0]<stderr>:INFO:root:Epoch[220] Batch[100] Loss[2.534]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[100]	rmse=0.018389	lr=0.034246
[1,0]<stderr>:INFO:root:Epoch[220] Batch[200] Loss[4.317]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[200]	rmse=0.018540	lr=0.034140
[1,0]<stderr>:INFO:root:Epoch[220] Batch[300] Loss[2.282]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[300]	rmse=0.018579	lr=0.034034
[1,0]<stderr>:INFO:root:Epoch[220] Batch[400] Loss[2.062]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[400]	rmse=0.018575	lr=0.033929
[1,0]<stderr>:INFO:root:Epoch[220] Batch[500] Loss[4.439]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[500]	rmse=0.018587	lr=0.033823
[1,0]<stderr>:INFO:root:Epoch[220] Batch[600] Loss[3.135]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[600]	rmse=0.018601	lr=0.033718
[1,0]<stderr>:INFO:root:Epoch[220] Batch[700] Loss[4.665]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[700]	rmse=0.018613	lr=0.033613
[1,0]<stderr>:INFO:root:Epoch[220] Batch[800] Loss[3.517]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[800]	rmse=0.018634	lr=0.033508
[1,0]<stderr>:INFO:root:Epoch[220] Batch[900] Loss[2.455]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[900]	rmse=0.018646	lr=0.033403
[1,0]<stderr>:INFO:root:Epoch[220] Batch[1000] Loss[4.268]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[1000]	rmse=0.018650	lr=0.033298
[1,0]<stderr>:INFO:root:Epoch[220] Batch[1100] Loss[2.637]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[1100]	rmse=0.018654	lr=0.033194
[1,0]<stderr>:INFO:root:Epoch[220] Batch[1200] Loss[2.337]
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[1200]	rmse=0.018662	lr=0.033089
[1,0]<stderr>:INFO:root:Epoch[220] Rank[0] Batch[1251]	Time cost=397.31	Train-metric=0.018666
[1,0]<stderr>:INFO:root:Epoch[220]	Speed: 3224.26 samples/sec
[1,0]<stderr>:INFO:root:Epoch[221] Batch[100] Loss[3.883]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[100]	rmse=0.018498	lr=0.032932
[1,0]<stderr>:INFO:root:Epoch[221] Batch[200] Loss[2.502]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[200]	rmse=0.018516	lr=0.032828
[1,0]<stderr>:INFO:root:Epoch[221] Batch[300] Loss[3.286]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[300]	rmse=0.018503	lr=0.032724
[1,0]<stderr>:INFO:root:Epoch[221] Batch[400] Loss[2.308]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[400]	rmse=0.018569	lr=0.032620
[1,0]<stderr>:INFO:root:Epoch[221] Batch[500] Loss[2.304]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[500]	rmse=0.018589	lr=0.032517
[1,0]<stderr>:INFO:root:Epoch[221] Batch[600] Loss[3.123]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[600]	rmse=0.018584	lr=0.032413
[1,0]<stderr>:INFO:root:Epoch[221] Batch[700] Loss[3.001]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[700]	rmse=0.018577	lr=0.032310
[1,0]<stderr>:INFO:root:Epoch[221] Batch[800] Loss[4.645]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[800]	rmse=0.018578	lr=0.032207
[1,0]<stderr>:INFO:root:Epoch[221] Batch[900] Loss[2.455]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[900]	rmse=0.018584	lr=0.032104
[1,0]<stderr>:INFO:root:Epoch[221] Batch[1000] Loss[2.459]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[1000]	rmse=0.018595	lr=0.032001
[1,0]<stderr>:INFO:root:Epoch[221] Batch[1100] Loss[2.720]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[1100]	rmse=0.018602	lr=0.031898
[1,0]<stderr>:INFO:root:Epoch[221] Batch[1200] Loss[2.302]
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[1200]	rmse=0.018618	lr=0.031796
[1,0]<stderr>:INFO:root:Epoch[221] Rank[0] Batch[1251]	Time cost=399.24	Train-metric=0.018615
[1,0]<stderr>:INFO:root:Epoch[221]	Speed: 3208.62 samples/sec
[1,0]<stderr>:INFO:root:Epoch[222] Batch[100] Loss[2.204]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[100]	rmse=0.018551	lr=0.031641
[1,0]<stderr>:INFO:root:Epoch[222] Batch[200] Loss[3.333]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[200]	rmse=0.018520	lr=0.031539
[1,0]<stderr>:INFO:root:Epoch[222] Batch[300] Loss[3.352]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[300]	rmse=0.018539	lr=0.031437
[1,0]<stderr>:INFO:root:Epoch[222] Batch[400] Loss[2.037]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[400]	rmse=0.018539	lr=0.031335
[1,0]<stderr>:INFO:root:Epoch[222] Batch[500] Loss[4.405]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[500]	rmse=0.018536	lr=0.031234
[1,0]<stderr>:INFO:root:Epoch[222] Batch[600] Loss[4.667]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[600]	rmse=0.018541	lr=0.031132
[1,0]<stderr>:INFO:root:Epoch[222] Batch[700] Loss[2.595]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[700]	rmse=0.018519	lr=0.031031
[1,0]<stderr>:INFO:root:Epoch[222] Batch[800] Loss[4.149]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[800]	rmse=0.018552	lr=0.030929
[1,0]<stderr>:INFO:root:Epoch[222] Batch[900] Loss[2.503]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[900]	rmse=0.018558	lr=0.030828
[1,0]<stderr>:INFO:root:Epoch[222] Batch[1000] Loss[2.612]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[1000]	rmse=0.018580	lr=0.030727
[1,0]<stderr>:INFO:root:Epoch[222] Batch[1100] Loss[2.208]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[1100]	rmse=0.018584	lr=0.030627
[1,0]<stderr>:INFO:root:Epoch[222] Batch[1200] Loss[4.722]
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[1200]	rmse=0.018592	lr=0.030526
[1,0]<stderr>:INFO:root:Epoch[222] Rank[0] Batch[1251]	Time cost=399.50	Train-metric=0.018590
[1,0]<stderr>:INFO:root:Epoch[222]	Speed: 3206.55 samples/sec
[1,0]<stderr>:INFO:root:Epoch[223] Batch[100] Loss[2.617]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[100]	rmse=0.018343	lr=0.030374
[1,0]<stderr>:INFO:root:Epoch[223] Batch[200] Loss[2.803]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[200]	rmse=0.018509	lr=0.030274
[1,0]<stderr>:INFO:root:Epoch[223] Batch[300] Loss[2.575]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[300]	rmse=0.018451	lr=0.030174
[1,0]<stderr>:INFO:root:Epoch[223] Batch[400] Loss[2.276]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[400]	rmse=0.018456	lr=0.030074
[1,0]<stderr>:INFO:root:Epoch[223] Batch[500] Loss[4.923]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[500]	rmse=0.018464	lr=0.029974
[1,0]<stderr>:INFO:root:Epoch[223] Batch[600] Loss[2.417]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[600]	rmse=0.018483	lr=0.029874
[1,0]<stderr>:INFO:root:Epoch[223] Batch[700] Loss[2.334]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[700]	rmse=0.018506	lr=0.029775
[1,0]<stderr>:INFO:root:Epoch[223] Batch[800] Loss[4.293]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[800]	rmse=0.018500	lr=0.029676
[1,0]<stderr>:INFO:root:Epoch[223] Batch[900] Loss[2.605]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[900]	rmse=0.018506	lr=0.029576
[1,0]<stderr>:INFO:root:Epoch[223] Batch[1000] Loss[3.259]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[1000]	rmse=0.018527	lr=0.029477
[1,0]<stderr>:INFO:root:Epoch[223] Batch[1100] Loss[2.440]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[1100]	rmse=0.018537	lr=0.029379
[1,0]<stderr>:INFO:root:Epoch[223] Batch[1200] Loss[2.357]
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[1200]	rmse=0.018551	lr=0.029280
[1,0]<stderr>:INFO:root:Epoch[223] Rank[0] Batch[1251]	Time cost=398.98	Train-metric=0.018554
[1,0]<stderr>:INFO:root:Epoch[223]	Speed: 3210.73 samples/sec
[1,0]<stderr>:INFO:root:Epoch[224] Batch[100] Loss[3.846]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[100]	rmse=0.018342	lr=0.029131
[1,0]<stderr>:INFO:root:Epoch[224] Batch[200] Loss[2.554]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[200]	rmse=0.018415	lr=0.029033
[1,0]<stderr>:INFO:root:Epoch[224] Batch[300] Loss[2.065]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[300]	rmse=0.018441	lr=0.028934
[1,0]<stderr>:INFO:root:Epoch[224] Batch[400] Loss[2.477]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[400]	rmse=0.018469	lr=0.028836
[1,0]<stderr>:INFO:root:Epoch[224] Batch[500] Loss[2.317]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[500]	rmse=0.018459	lr=0.028739
[1,0]<stderr>:INFO:root:Epoch[224] Batch[600] Loss[4.635]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[600]	rmse=0.018453	lr=0.028641
[1,0]<stderr>:INFO:root:Epoch[224] Batch[700] Loss[2.487]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[700]	rmse=0.018439	lr=0.028543
[1,0]<stderr>:INFO:root:Epoch[224] Batch[800] Loss[3.706]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[800]	rmse=0.018455	lr=0.028446
[1,0]<stderr>:INFO:root:Epoch[224] Batch[900] Loss[3.827]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[900]	rmse=0.018473	lr=0.028349
[1,0]<stderr>:INFO:root:Epoch[224] Batch[1000] Loss[2.238]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[1000]	rmse=0.018482	lr=0.028251
[1,0]<stderr>:INFO:root:Epoch[224] Batch[1100] Loss[2.342]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[1100]	rmse=0.018493	lr=0.028154
[1,0]<stderr>:INFO:root:Epoch[224] Batch[1200] Loss[2.278]
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[1200]	rmse=0.018505	lr=0.028058
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0] Batch[1251]	Time cost=399.92	Train-metric=0.018514
[1,0]<stderr>:INFO:root:Epoch[224]	Speed: 3203.19 samples/sec
[1,0]<stderr>:INFO:root:Epoch[224] Rank[0]	Validation-accuracy=0.727260	Validation-top_k_accuracy_5=0.910560
[1,0]<stderr>:INFO:root:Epoch[225] Batch[100] Loss[2.467]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[100]	rmse=0.018521	lr=0.027912
[1,0]<stderr>:INFO:root:Epoch[225] Batch[200] Loss[4.256]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[200]	rmse=0.018534	lr=0.027815
[1,0]<stderr>:INFO:root:Epoch[225] Batch[300] Loss[2.552]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[300]	rmse=0.018523	lr=0.027719
[1,0]<stderr>:INFO:root:Epoch[225] Batch[400] Loss[3.071]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[400]	rmse=0.018502	lr=0.027623
[1,0]<stderr>:INFO:root:Epoch[225] Batch[500] Loss[2.331]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[500]	rmse=0.018504	lr=0.027527
[1,0]<stderr>:INFO:root:Epoch[225] Batch[600] Loss[2.661]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[600]	rmse=0.018538	lr=0.027431
[1,0]<stderr>:INFO:root:Epoch[225] Batch[700] Loss[2.923]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[700]	rmse=0.018547	lr=0.027336
[1,0]<stderr>:INFO:root:Epoch[225] Batch[800] Loss[3.497]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[800]	rmse=0.018548	lr=0.027240
[1,0]<stderr>:INFO:root:Epoch[225] Batch[900] Loss[2.330]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[900]	rmse=0.018536	lr=0.027145
[1,0]<stderr>:INFO:root:Epoch[225] Batch[1000] Loss[2.331]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[1000]	rmse=0.018537	lr=0.027050
[1,0]<stderr>:INFO:root:Epoch[225] Batch[1100] Loss[4.681]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[1100]	rmse=0.018532	lr=0.026955
[1,0]<stderr>:INFO:root:Epoch[225] Batch[1200] Loss[2.183]
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[1200]	rmse=0.018540	lr=0.026860
[1,0]<stderr>:INFO:root:Epoch[225] Rank[0] Batch[1251]	Time cost=403.08	Train-metric=0.018539
[1,0]<stderr>:INFO:root:Epoch[225]	Speed: 3178.06 samples/sec
[1,0]<stderr>:INFO:root:Epoch[226] Batch[100] Loss[4.573]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[100]	rmse=0.018389	lr=0.026717
[1,0]<stderr>:INFO:root:Epoch[226] Batch[200] Loss[2.419]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[200]	rmse=0.018420	lr=0.026622
[1,0]<stderr>:INFO:root:Epoch[226] Batch[300] Loss[2.253]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[300]	rmse=0.018406	lr=0.026528
[1,0]<stderr>:INFO:root:Epoch[226] Batch[400] Loss[2.580]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[400]	rmse=0.018403	lr=0.026434
[1,0]<stderr>:INFO:root:Epoch[226] Batch[500] Loss[3.933]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[500]	rmse=0.018439	lr=0.026340
[1,0]<stderr>:INFO:root:Epoch[226] Batch[600] Loss[4.948]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[600]	rmse=0.018454	lr=0.026246
[1,0]<stderr>:INFO:root:Epoch[226] Batch[700] Loss[2.245]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[700]	rmse=0.018464	lr=0.026152
[1,0]<stderr>:INFO:root:Epoch[226] Batch[800] Loss[4.338]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[800]	rmse=0.018477	lr=0.026059
[1,0]<stderr>:INFO:root:Epoch[226] Batch[900] Loss[2.425]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[900]	rmse=0.018471	lr=0.025965
[1,0]<stderr>:INFO:root:Epoch[226] Batch[1000] Loss[2.371]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[1000]	rmse=0.018464	lr=0.025872
[1,0]<stderr>:INFO:root:Epoch[226] Batch[1100] Loss[4.439]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[1100]	rmse=0.018472	lr=0.025779
[1,0]<stderr>:INFO:root:Epoch[226] Batch[1200] Loss[2.609]
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[1200]	rmse=0.018468	lr=0.025686
[1,0]<stderr>:INFO:root:Epoch[226] Rank[0] Batch[1251]	Time cost=402.85	Train-metric=0.018475
[1,0]<stderr>:INFO:root:Epoch[226]	Speed: 3179.88 samples/sec
[1,0]<stderr>:INFO:root:Epoch[227] Batch[100] Loss[2.304]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[100]	rmse=0.018276	lr=0.025546
[1,0]<stderr>:INFO:root:Epoch[227] Batch[200] Loss[2.310]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[200]	rmse=0.018415	lr=0.025453
[1,0]<stderr>:INFO:root:Epoch[227] Batch[300] Loss[2.264]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[300]	rmse=0.018370	lr=0.025361
[1,0]<stderr>:INFO:root:Epoch[227] Batch[400] Loss[2.144]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[400]	rmse=0.018370	lr=0.025269
[1,0]<stderr>:INFO:root:Epoch[227] Batch[500] Loss[2.189]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[500]	rmse=0.018370	lr=0.025177
[1,0]<stderr>:INFO:root:Epoch[227] Batch[600] Loss[4.051]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[600]	rmse=0.018421	lr=0.025085
[1,0]<stderr>:INFO:root:Epoch[227] Batch[700] Loss[4.592]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[700]	rmse=0.018436	lr=0.024993
[1,0]<stderr>:INFO:root:Epoch[227] Batch[800] Loss[2.027]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[800]	rmse=0.018445	lr=0.024901
[1,0]<stderr>:INFO:root:Epoch[227] Batch[900] Loss[2.229]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[900]	rmse=0.018439	lr=0.024810
[1,0]<stderr>:INFO:root:Epoch[227] Batch[1000] Loss[4.676]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[1000]	rmse=0.018443	lr=0.024719
[1,0]<stderr>:INFO:root:Epoch[227] Batch[1100] Loss[2.543]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[1100]	rmse=0.018456	lr=0.024628
[1,0]<stderr>:INFO:root:Epoch[227] Batch[1200] Loss[2.212]
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[1200]	rmse=0.018470	lr=0.024537
[1,0]<stderr>:INFO:root:Epoch[227] Rank[0] Batch[1251]	Time cost=404.24	Train-metric=0.018459
[1,0]<stderr>:INFO:root:Epoch[227]	Speed: 3169.00 samples/sec
[1,0]<stderr>:INFO:root:Epoch[228] Batch[100] Loss[2.242]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[100]	rmse=0.018395	lr=0.024400
[1,0]<stderr>:INFO:root:Epoch[228] Batch[200] Loss[2.337]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[200]	rmse=0.018345	lr=0.024309
[1,0]<stderr>:INFO:root:Epoch[228] Batch[300] Loss[2.328]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[300]	rmse=0.018365	lr=0.024219
[1,0]<stderr>:INFO:root:Epoch[228] Batch[400] Loss[2.125]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[400]	rmse=0.018391	lr=0.024128
[1,0]<stderr>:INFO:root:Epoch[228] Batch[500] Loss[2.572]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[500]	rmse=0.018401	lr=0.024038
[1,0]<stderr>:INFO:root:Epoch[228] Batch[600] Loss[3.847]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[600]	rmse=0.018392	lr=0.023948
[1,0]<stderr>:INFO:root:Epoch[228] Batch[700] Loss[2.307]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[700]	rmse=0.018397	lr=0.023859
[1,0]<stderr>:INFO:root:Epoch[228] Batch[800] Loss[2.427]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[800]	rmse=0.018411	lr=0.023769
[1,0]<stderr>:INFO:root:Epoch[228] Batch[900] Loss[2.795]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[900]	rmse=0.018414	lr=0.023680
[1,0]<stderr>:INFO:root:Epoch[228] Batch[1000] Loss[4.395]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[1000]	rmse=0.018433	lr=0.023590
[1,0]<stderr>:INFO:root:Epoch[228] Batch[1100] Loss[2.368]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[1100]	rmse=0.018426	lr=0.023501
[1,0]<stderr>:INFO:root:Epoch[228] Batch[1200] Loss[2.377]
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[1200]	rmse=0.018434	lr=0.023412
[1,0]<stderr>:INFO:root:Epoch[228] Rank[0] Batch[1251]	Time cost=401.70	Train-metric=0.018435
[1,0]<stderr>:INFO:root:Epoch[228]	Speed: 3189.03 samples/sec
[1,0]<stderr>:INFO:root:Epoch[229] Batch[100] Loss[4.314]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[100]	rmse=0.018317	lr=0.023278
[1,0]<stderr>:INFO:root:Epoch[229] Batch[200] Loss[4.020]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[200]	rmse=0.018333	lr=0.023189
[1,0]<stderr>:INFO:root:Epoch[229] Batch[300] Loss[2.749]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[300]	rmse=0.018316	lr=0.023101
[1,0]<stderr>:INFO:root:Epoch[229] Batch[400] Loss[2.383]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[400]	rmse=0.018332	lr=0.023013
[1,0]<stderr>:INFO:root:Epoch[229] Batch[500] Loss[2.034]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[500]	rmse=0.018324	lr=0.022925
[1,0]<stderr>:INFO:root:Epoch[229] Batch[600] Loss[2.349]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[600]	rmse=0.018306	lr=0.022837
[1,0]<stderr>:INFO:root:Epoch[229] Batch[700] Loss[2.553]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[700]	rmse=0.018327	lr=0.022749
[1,0]<stderr>:INFO:root:Epoch[229] Batch[800] Loss[3.021]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[800]	rmse=0.018345	lr=0.022661
[1,0]<stderr>:INFO:root:Epoch[229] Batch[900] Loss[4.713]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[900]	rmse=0.018338	lr=0.022574
[1,0]<stderr>:INFO:root:Epoch[229] Batch[1000] Loss[2.282]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[1000]	rmse=0.018348	lr=0.022486
[1,0]<stderr>:INFO:root:Epoch[229] Batch[1100] Loss[3.603]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[1100]	rmse=0.018363	lr=0.022399
[1,0]<stderr>:INFO:root:Epoch[229] Batch[1200] Loss[2.377]
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[1200]	rmse=0.018362	lr=0.022312
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0] Batch[1251]	Time cost=399.79	Train-metric=0.018359
[1,0]<stderr>:INFO:root:Epoch[229]	Speed: 3204.24 samples/sec
[1,0]<stderr>:INFO:root:Epoch[229] Rank[0]	Validation-accuracy=0.732360	Validation-top_k_accuracy_5=0.910980
[1,0]<stderr>:INFO:root:Epoch[230] Batch[100] Loss[2.357]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[100]	rmse=0.018225	lr=0.022181
[1,0]<stderr>:INFO:root:Epoch[230] Batch[200] Loss[2.338]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[200]	rmse=0.018280	lr=0.022095
[1,0]<stderr>:INFO:root:Epoch[230] Batch[300] Loss[2.516]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[300]	rmse=0.018285	lr=0.022008
[1,0]<stderr>:INFO:root:Epoch[230] Batch[400] Loss[2.174]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[400]	rmse=0.018304	lr=0.021922
[1,0]<stderr>:INFO:root:Epoch[230] Batch[500] Loss[2.237]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[500]	rmse=0.018330	lr=0.021836
[1,0]<stderr>:INFO:root:Epoch[230] Batch[600] Loss[2.116]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[600]	rmse=0.018307	lr=0.021750
[1,0]<stderr>:INFO:root:Epoch[230] Batch[700] Loss[3.200]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[700]	rmse=0.018309	lr=0.021664
[1,0]<stderr>:INFO:root:Epoch[230] Batch[800] Loss[4.145]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[800]	rmse=0.018298	lr=0.021578
[1,0]<stderr>:INFO:root:Epoch[230] Batch[900] Loss[4.928]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[900]	rmse=0.018304	lr=0.021493
[1,0]<stderr>:INFO:root:Epoch[230] Batch[1000] Loss[2.424]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[1000]	rmse=0.018309	lr=0.021407
[1,0]<stderr>:INFO:root:Epoch[230] Batch[1100] Loss[2.494]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[1100]	rmse=0.018323	lr=0.021322
[1,0]<stderr>:INFO:root:Epoch[230] Batch[1200] Loss[2.390]
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[1200]	rmse=0.018337	lr=0.021237
[1,0]<stderr>:INFO:root:Epoch[230] Rank[0] Batch[1251]	Time cost=399.09	Train-metric=0.018334
[1,0]<stderr>:INFO:root:Epoch[230]	Speed: 3209.90 samples/sec
[1,0]<stderr>:INFO:root:Epoch[231] Batch[100] Loss[2.309]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[100]	rmse=0.018237	lr=0.021109
[1,0]<stderr>:INFO:root:Epoch[231] Batch[200] Loss[2.860]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[200]	rmse=0.018275	lr=0.021025
[1,0]<stderr>:INFO:root:Epoch[231] Batch[300] Loss[2.761]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[300]	rmse=0.018242	lr=0.020940
[1,0]<stderr>:INFO:root:Epoch[231] Batch[400] Loss[2.371]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[400]	rmse=0.018253	lr=0.020856
[1,0]<stderr>:INFO:root:Epoch[231] Batch[500] Loss[3.828]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[500]	rmse=0.018246	lr=0.020772
[1,0]<stderr>:INFO:root:Epoch[231] Batch[600] Loss[2.403]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[600]	rmse=0.018259	lr=0.020688
[1,0]<stderr>:INFO:root:Epoch[231] Batch[700] Loss[4.090]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[700]	rmse=0.018283	lr=0.020604
[1,0]<stderr>:INFO:root:Epoch[231] Batch[800] Loss[4.892]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[800]	rmse=0.018285	lr=0.020520
[1,0]<stderr>:INFO:root:Epoch[231] Batch[900] Loss[4.623]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[900]	rmse=0.018289	lr=0.020437
[1,0]<stderr>:INFO:root:Epoch[231] Batch[1000] Loss[2.218]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[1000]	rmse=0.018299	lr=0.020354
[1,0]<stderr>:INFO:root:Epoch[231] Batch[1100] Loss[2.169]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[1100]	rmse=0.018295	lr=0.020270
[1,0]<stderr>:INFO:root:Epoch[231] Batch[1200] Loss[2.286]
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[1200]	rmse=0.018296	lr=0.020187
[1,0]<stderr>:INFO:root:Epoch[231] Rank[0] Batch[1251]	Time cost=399.34	Train-metric=0.018305
[1,0]<stderr>:INFO:root:Epoch[231]	Speed: 3207.84 samples/sec
[1,0]<stderr>:INFO:root:Epoch[232] Batch[100] Loss[2.446]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[100]	rmse=0.018154	lr=0.020062
[1,0]<stderr>:INFO:root:Epoch[232] Batch[200] Loss[2.968]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[200]	rmse=0.018203	lr=0.019980
[1,0]<stderr>:INFO:root:Epoch[232] Batch[300] Loss[2.861]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[300]	rmse=0.018201	lr=0.019897
[1,0]<stderr>:INFO:root:Epoch[232] Batch[400] Loss[2.220]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[400]	rmse=0.018222	lr=0.019815
[1,0]<stderr>:INFO:root:Epoch[232] Batch[500] Loss[3.471]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[500]	rmse=0.018249	lr=0.019733
[1,0]<stderr>:INFO:root:Epoch[232] Batch[600] Loss[2.147]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[600]	rmse=0.018263	lr=0.019651
[1,0]<stderr>:INFO:root:Epoch[232] Batch[700] Loss[3.073]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[700]	rmse=0.018269	lr=0.019569
[1,0]<stderr>:INFO:root:Epoch[232] Batch[800] Loss[2.364]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[800]	rmse=0.018280	lr=0.019488
[1,0]<stderr>:INFO:root:Epoch[232] Batch[900] Loss[2.583]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[900]	rmse=0.018273	lr=0.019406
[1,0]<stderr>:INFO:root:Epoch[232] Batch[1000] Loss[4.621]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[1000]	rmse=0.018275	lr=0.019325
[1,0]<stderr>:INFO:root:Epoch[232] Batch[1100] Loss[2.644]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[1100]	rmse=0.018277	lr=0.019244
[1,0]<stderr>:INFO:root:Epoch[232] Batch[1200] Loss[2.341]
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[1200]	rmse=0.018271	lr=0.019163
[1,0]<stderr>:INFO:root:Epoch[232] Rank[0] Batch[1251]	Time cost=399.78	Train-metric=0.018269
[1,0]<stderr>:INFO:root:Epoch[232]	Speed: 3204.33 samples/sec
[1,0]<stderr>:INFO:root:Epoch[233] Batch[100] Loss[2.814]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[100]	rmse=0.018117	lr=0.019041
[1,0]<stderr>:INFO:root:Epoch[233] Batch[200] Loss[2.364]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[200]	rmse=0.018136	lr=0.018960
[1,0]<stderr>:INFO:root:Epoch[233] Batch[300] Loss[2.449]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[300]	rmse=0.018157	lr=0.018880
[1,0]<stderr>:INFO:root:Epoch[233] Batch[400] Loss[2.231]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[400]	rmse=0.018189	lr=0.018800
[1,0]<stderr>:INFO:root:Epoch[233] Batch[500] Loss[2.304]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[500]	rmse=0.018205	lr=0.018720
[1,0]<stderr>:INFO:root:Epoch[233] Batch[600] Loss[3.311]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[600]	rmse=0.018195	lr=0.018640
[1,0]<stderr>:INFO:root:Epoch[233] Batch[700] Loss[2.046]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[700]	rmse=0.018184	lr=0.018560
[1,0]<stderr>:INFO:root:Epoch[233] Batch[800] Loss[2.617]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[800]	rmse=0.018187	lr=0.018480
[1,0]<stderr>:INFO:root:Epoch[233] Batch[900] Loss[2.027]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[900]	rmse=0.018189	lr=0.018401
[1,0]<stderr>:INFO:root:Epoch[233] Batch[1000] Loss[2.361]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[1000]	rmse=0.018185	lr=0.018322
[1,0]<stderr>:INFO:root:Epoch[233] Batch[1100] Loss[2.139]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[1100]	rmse=0.018200	lr=0.018243
[1,0]<stderr>:INFO:root:Epoch[233] Batch[1200] Loss[2.361]
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[1200]	rmse=0.018204	lr=0.018164
[1,0]<stderr>:INFO:root:Epoch[233] Rank[0] Batch[1251]	Time cost=399.16	Train-metric=0.018199
[1,0]<stderr>:INFO:root:Epoch[233]	Speed: 3209.29 samples/sec
[1,0]<stderr>:INFO:root:Epoch[234] Batch[100] Loss[2.418]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[100]	rmse=0.018117	lr=0.018045
[1,0]<stderr>:INFO:root:Epoch[234] Batch[200] Loss[3.945]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[200]	rmse=0.018101	lr=0.017966
[1,0]<stderr>:INFO:root:Epoch[234] Batch[300] Loss[4.423]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[300]	rmse=0.018143	lr=0.017888
[1,0]<stderr>:INFO:root:Epoch[234] Batch[400] Loss[3.033]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[400]	rmse=0.018147	lr=0.017810
[1,0]<stderr>:INFO:root:Epoch[234] Batch[500] Loss[2.512]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[500]	rmse=0.018134	lr=0.017732
[1,0]<stderr>:INFO:root:Epoch[234] Batch[600] Loss[2.789]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[600]	rmse=0.018154	lr=0.017654
[1,0]<stderr>:INFO:root:Epoch[234] Batch[700] Loss[2.322]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[700]	rmse=0.018172	lr=0.017576
[1,0]<stderr>:INFO:root:Epoch[234] Batch[800] Loss[2.351]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[800]	rmse=0.018176	lr=0.017498
[1,0]<stderr>:INFO:root:Epoch[234] Batch[900] Loss[4.293]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[900]	rmse=0.018197	lr=0.017421
[1,0]<stderr>:INFO:root:Epoch[234] Batch[1000] Loss[2.262]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[1000]	rmse=0.018206	lr=0.017344
[1,0]<stderr>:INFO:root:Epoch[234] Batch[1100] Loss[2.305]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[1100]	rmse=0.018201	lr=0.017267
[1,0]<stderr>:INFO:root:Epoch[234] Batch[1200] Loss[2.811]
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[1200]	rmse=0.018208	lr=0.017190
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0] Batch[1251]	Time cost=399.51	Train-metric=0.018198
[1,0]<stderr>:INFO:root:Epoch[234]	Speed: 3206.48 samples/sec
[1,0]<stderr>:INFO:root:Epoch[234] Rank[0]	Validation-accuracy=0.734900	Validation-top_k_accuracy_5=0.915180
[1,0]<stderr>:INFO:root:Epoch[235] Batch[100] Loss[2.991]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[100]	rmse=0.018036	lr=0.017074
[1,0]<stderr>:INFO:root:Epoch[235] Batch[200] Loss[2.623]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[200]	rmse=0.018189	lr=0.016998
[1,0]<stderr>:INFO:root:Epoch[235] Batch[300] Loss[2.968]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[300]	rmse=0.018169	lr=0.016921
[1,0]<stderr>:INFO:root:Epoch[235] Batch[400] Loss[2.585]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[400]	rmse=0.018138	lr=0.016845
[1,0]<stderr>:INFO:root:Epoch[235] Batch[500] Loss[2.554]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[500]	rmse=0.018133	lr=0.016769
[1,0]<stderr>:INFO:root:Epoch[235] Batch[600] Loss[2.602]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[600]	rmse=0.018111	lr=0.016693
[1,0]<stderr>:INFO:root:Epoch[235] Batch[700] Loss[2.927]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[700]	rmse=0.018119	lr=0.016618
[1,0]<stderr>:INFO:root:Epoch[235] Batch[800] Loss[2.169]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[800]	rmse=0.018133	lr=0.016542
[1,0]<stderr>:INFO:root:Epoch[235] Batch[900] Loss[2.480]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[900]	rmse=0.018129	lr=0.016467
[1,0]<stderr>:INFO:root:Epoch[235] Batch[1000] Loss[4.923]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[1000]	rmse=0.018134	lr=0.016392
[1,0]<stderr>:INFO:root:Epoch[235] Batch[1100] Loss[2.419]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[1100]	rmse=0.018126	lr=0.016317
[1,0]<stderr>:INFO:root:Epoch[235] Batch[1200] Loss[4.569]
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[1200]	rmse=0.018140	lr=0.016242
[1,0]<stderr>:INFO:root:Epoch[235] Rank[0] Batch[1251]	Time cost=398.90	Train-metric=0.018144
[1,0]<stderr>:INFO:root:Epoch[235]	Speed: 3211.42 samples/sec
[1,0]<stderr>:INFO:root:Epoch[236] Batch[100] Loss[2.374]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[100]	rmse=0.018061	lr=0.016129
[1,0]<stderr>:INFO:root:Epoch[236] Batch[200] Loss[2.193]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[200]	rmse=0.018095	lr=0.016055
[1,0]<stderr>:INFO:root:Epoch[236] Batch[300] Loss[4.856]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[300]	rmse=0.018139	lr=0.015980
[1,0]<stderr>:INFO:root:Epoch[236] Batch[400] Loss[2.194]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[400]	rmse=0.018119	lr=0.015906
[1,0]<stderr>:INFO:root:Epoch[236] Batch[500] Loss[3.484]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[500]	rmse=0.018105	lr=0.015832
[1,0]<stderr>:INFO:root:Epoch[236] Batch[600] Loss[2.265]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[600]	rmse=0.018107	lr=0.015759
[1,0]<stderr>:INFO:root:Epoch[236] Batch[700] Loss[3.747]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[700]	rmse=0.018105	lr=0.015685
[1,0]<stderr>:INFO:root:Epoch[236] Batch[800] Loss[2.126]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[800]	rmse=0.018108	lr=0.015612
[1,0]<stderr>:INFO:root:Epoch[236] Batch[900] Loss[3.078]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[900]	rmse=0.018113	lr=0.015538
[1,0]<stderr>:INFO:root:Epoch[236] Batch[1000] Loss[2.314]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[1000]	rmse=0.018120	lr=0.015465
[1,0]<stderr>:INFO:root:Epoch[236] Batch[1100] Loss[3.123]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[1100]	rmse=0.018124	lr=0.015392
[1,0]<stderr>:INFO:root:Epoch[236] Batch[1200] Loss[4.294]
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[1200]	rmse=0.018121	lr=0.015320
[1,0]<stderr>:INFO:root:Epoch[236] Rank[0] Batch[1251]	Time cost=399.14	Train-metric=0.018127
[1,0]<stderr>:INFO:root:Epoch[236]	Speed: 3209.44 samples/sec
[1,0]<stderr>:INFO:root:Epoch[237] Batch[100] Loss[2.633]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[100]	rmse=0.017955	lr=0.015210
[1,0]<stderr>:INFO:root:Epoch[237] Batch[200] Loss[2.295]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[200]	rmse=0.018030	lr=0.015138
[1,0]<stderr>:INFO:root:Epoch[237] Batch[300] Loss[3.551]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[300]	rmse=0.018050	lr=0.015065
[1,0]<stderr>:INFO:root:Epoch[237] Batch[400] Loss[2.163]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[400]	rmse=0.018036	lr=0.014993
[1,0]<stderr>:INFO:root:Epoch[237] Batch[500] Loss[2.257]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[500]	rmse=0.018077	lr=0.014922
[1,0]<stderr>:INFO:root:Epoch[237] Batch[600] Loss[2.241]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[600]	rmse=0.018081	lr=0.014850
[1,0]<stderr>:INFO:root:Epoch[237] Batch[700] Loss[3.879]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[700]	rmse=0.018073	lr=0.014778
[1,0]<stderr>:INFO:root:Epoch[237] Batch[800] Loss[2.259]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[800]	rmse=0.018081	lr=0.014707
[1,0]<stderr>:INFO:root:Epoch[237] Batch[900] Loss[4.666]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[900]	rmse=0.018077	lr=0.014636
[1,0]<stderr>:INFO:root:Epoch[237] Batch[1000] Loss[2.930]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[1000]	rmse=0.018084	lr=0.014565
[1,0]<stderr>:INFO:root:Epoch[237] Batch[1100] Loss[4.466]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[1100]	rmse=0.018080	lr=0.014494
[1,0]<stderr>:INFO:root:Epoch[237] Batch[1200] Loss[2.644]
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[1200]	rmse=0.018075	lr=0.014423
[1,0]<stderr>:INFO:root:Epoch[237] Rank[0] Batch[1251]	Time cost=400.20	Train-metric=0.018082
[1,0]<stderr>:INFO:root:Epoch[237]	Speed: 3200.99 samples/sec
[1,0]<stderr>:INFO:root:Epoch[238] Batch[100] Loss[2.153]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[100]	rmse=0.017962	lr=0.014317
[1,0]<stderr>:INFO:root:Epoch[238] Batch[200] Loss[2.582]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[200]	rmse=0.017922	lr=0.014246
[1,0]<stderr>:INFO:root:Epoch[238] Batch[300] Loss[3.679]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[300]	rmse=0.017980	lr=0.014176
[1,0]<stderr>:INFO:root:Epoch[238] Batch[400] Loss[1.962]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[400]	rmse=0.017984	lr=0.014106
[1,0]<stderr>:INFO:root:Epoch[238] Batch[500] Loss[2.659]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[500]	rmse=0.018000	lr=0.014037
[1,0]<stderr>:INFO:root:Epoch[238] Batch[600] Loss[4.465]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[600]	rmse=0.018002	lr=0.013967
[1,0]<stderr>:INFO:root:Epoch[238] Batch[700] Loss[2.143]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[700]	rmse=0.018031	lr=0.013898
[1,0]<stderr>:INFO:root:Epoch[238] Batch[800] Loss[3.870]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[800]	rmse=0.018039	lr=0.013828
[1,0]<stderr>:INFO:root:Epoch[238] Batch[900] Loss[2.426]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[900]	rmse=0.018035	lr=0.013759
[1,0]<stderr>:INFO:root:Epoch[238] Batch[1000] Loss[2.663]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[1000]	rmse=0.018038	lr=0.013690
[1,0]<stderr>:INFO:root:Epoch[238] Batch[1100] Loss[2.316]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[1100]	rmse=0.018023	lr=0.013621
[1,0]<stderr>:INFO:root:Epoch[238] Batch[1200] Loss[2.081]
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[1200]	rmse=0.018033	lr=0.013553
[1,0]<stderr>:INFO:root:Epoch[238] Rank[0] Batch[1251]	Time cost=399.42	Train-metric=0.018037
[1,0]<stderr>:INFO:root:Epoch[238]	Speed: 3207.20 samples/sec
[1,0]<stderr>:INFO:root:Epoch[239] Batch[100] Loss[2.695]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[100]	rmse=0.017806	lr=0.013450
[1,0]<stderr>:INFO:root:Epoch[239] Batch[200] Loss[3.960]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[200]	rmse=0.017864	lr=0.013381
[1,0]<stderr>:INFO:root:Epoch[239] Batch[300] Loss[2.395]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[300]	rmse=0.017905	lr=0.013313
[1,0]<stderr>:INFO:root:Epoch[239] Batch[400] Loss[3.643]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[400]	rmse=0.017921	lr=0.013245
[1,0]<stderr>:INFO:root:Epoch[239] Batch[500] Loss[2.201]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[500]	rmse=0.017956	lr=0.013178
[1,0]<stderr>:INFO:root:Epoch[239] Batch[600] Loss[4.658]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[600]	rmse=0.017982	lr=0.013110
[1,0]<stderr>:INFO:root:Epoch[239] Batch[700] Loss[2.289]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[700]	rmse=0.017991	lr=0.013043
[1,0]<stderr>:INFO:root:Epoch[239] Batch[800] Loss[2.548]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[800]	rmse=0.017986	lr=0.012976
[1,0]<stderr>:INFO:root:Epoch[239] Batch[900] Loss[3.650]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[900]	rmse=0.017989	lr=0.012909
[1,0]<stderr>:INFO:root:Epoch[239] Batch[1000] Loss[2.307]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[1000]	rmse=0.017974	lr=0.012842
[1,0]<stderr>:INFO:root:Epoch[239] Batch[1100] Loss[4.391]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[1100]	rmse=0.017976	lr=0.012775
[1,0]<stderr>:INFO:root:Epoch[239] Batch[1200] Loss[2.130]
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[1200]	rmse=0.017992	lr=0.012709
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0] Batch[1251]	Time cost=398.72	Train-metric=0.017997
[1,0]<stderr>:INFO:root:Epoch[239]	Speed: 3212.81 samples/sec
[1,0]<stderr>:INFO:root:Epoch[239] Rank[0]	Validation-accuracy=0.741840	Validation-top_k_accuracy_5=0.918000
[1,0]<stderr>:INFO:root:Epoch[240] Batch[100] Loss[2.392]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[100]	rmse=0.017917	lr=0.012609
[1,0]<stderr>:INFO:root:Epoch[240] Batch[200] Loss[2.570]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[200]	rmse=0.017938	lr=0.012542
[1,0]<stderr>:INFO:root:Epoch[240] Batch[300] Loss[2.310]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[300]	rmse=0.017920	lr=0.012477
[1,0]<stderr>:INFO:root:Epoch[240] Batch[400] Loss[2.254]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[400]	rmse=0.017909	lr=0.012411
[1,0]<stderr>:INFO:root:Epoch[240] Batch[500] Loss[2.271]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[500]	rmse=0.017942	lr=0.012345
[1,0]<stderr>:INFO:root:Epoch[240] Batch[600] Loss[2.336]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[600]	rmse=0.017938	lr=0.012280
[1,0]<stderr>:INFO:root:Epoch[240] Batch[700] Loss[2.321]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[700]	rmse=0.017962	lr=0.012215
[1,0]<stderr>:INFO:root:Epoch[240] Batch[800] Loss[2.175]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[800]	rmse=0.017963	lr=0.012149
[1,0]<stderr>:INFO:root:Epoch[240] Batch[900] Loss[2.155]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[900]	rmse=0.017975	lr=0.012085
[1,0]<stderr>:INFO:root:Epoch[240] Batch[1000] Loss[3.330]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[1000]	rmse=0.017988	lr=0.012020
[1,0]<stderr>:INFO:root:Epoch[240] Batch[1100] Loss[4.544]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[1100]	rmse=0.017981	lr=0.011955
[1,0]<stderr>:INFO:root:Epoch[240] Batch[1200] Loss[2.203]
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[1200]	rmse=0.017977	lr=0.011891
[1,0]<stderr>:INFO:root:Epoch[240] Rank[0] Batch[1251]	Time cost=397.51	Train-metric=0.017970
[1,0]<stderr>:INFO:root:Epoch[240]	Speed: 3222.64 samples/sec
[1,0]<stderr>:INFO:root:Epoch[241] Batch[100] Loss[4.880]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[100]	rmse=0.017921	lr=0.011794
[1,0]<stderr>:INFO:root:Epoch[241] Batch[200] Loss[1.931]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[200]	rmse=0.017940	lr=0.011730
[1,0]<stderr>:INFO:root:Epoch[241] Batch[300] Loss[2.959]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[300]	rmse=0.017949	lr=0.011666
[1,0]<stderr>:INFO:root:Epoch[241] Batch[400] Loss[4.788]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[400]	rmse=0.017967	lr=0.011602
[1,0]<stderr>:INFO:root:Epoch[241] Batch[500] Loss[2.046]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[500]	rmse=0.017943	lr=0.011539
[1,0]<stderr>:INFO:root:Epoch[241] Batch[600] Loss[2.458]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[600]	rmse=0.017948	lr=0.011476
[1,0]<stderr>:INFO:root:Epoch[241] Batch[700] Loss[2.283]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[700]	rmse=0.017929	lr=0.011412
[1,0]<stderr>:INFO:root:Epoch[241] Batch[800] Loss[2.146]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[800]	rmse=0.017937	lr=0.011349
[1,0]<stderr>:INFO:root:Epoch[241] Batch[900] Loss[2.445]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[900]	rmse=0.017927	lr=0.011287
[1,0]<stderr>:INFO:root:Epoch[241] Batch[1000] Loss[2.792]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[1000]	rmse=0.017926	lr=0.011224
[1,0]<stderr>:INFO:root:Epoch[241] Batch[1100] Loss[3.178]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[1100]	rmse=0.017929	lr=0.011162
[1,0]<stderr>:INFO:root:Epoch[241] Batch[1200] Loss[2.021]
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[1200]	rmse=0.017925	lr=0.011099
[1,0]<stderr>:INFO:root:Epoch[241] Rank[0] Batch[1251]	Time cost=399.30	Train-metric=0.017927
[1,0]<stderr>:INFO:root:Epoch[241]	Speed: 3208.19 samples/sec
[1,0]<stderr>:INFO:root:Epoch[242] Batch[100] Loss[2.023]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[100]	rmse=0.017846	lr=0.011006
[1,0]<stderr>:INFO:root:Epoch[242] Batch[200] Loss[2.332]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[200]	rmse=0.017845	lr=0.010944
[1,0]<stderr>:INFO:root:Epoch[242] Batch[300] Loss[4.760]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[300]	rmse=0.017855	lr=0.010882
[1,0]<stderr>:INFO:root:Epoch[242] Batch[400] Loss[2.234]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[400]	rmse=0.017844	lr=0.010820
[1,0]<stderr>:INFO:root:Epoch[242] Batch[500] Loss[4.475]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[500]	rmse=0.017858	lr=0.010759
[1,0]<stderr>:INFO:root:Epoch[242] Batch[600] Loss[3.502]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[600]	rmse=0.017867	lr=0.010698
[1,0]<stderr>:INFO:root:Epoch[242] Batch[700] Loss[4.662]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[700]	rmse=0.017884	lr=0.010637
[1,0]<stderr>:INFO:root:Epoch[242] Batch[800] Loss[2.128]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[800]	rmse=0.017883	lr=0.010576
[1,0]<stderr>:INFO:root:Epoch[242] Batch[900] Loss[2.270]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[900]	rmse=0.017894	lr=0.010515
[1,0]<stderr>:INFO:root:Epoch[242] Batch[1000] Loss[2.407]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[1000]	rmse=0.017910	lr=0.010455
[1,0]<stderr>:INFO:root:Epoch[242] Batch[1100] Loss[3.488]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[1100]	rmse=0.017904	lr=0.010394
[1,0]<stderr>:INFO:root:Epoch[242] Batch[1200] Loss[2.221]
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[1200]	rmse=0.017905	lr=0.010334
[1,0]<stderr>:INFO:root:Epoch[242] Rank[0] Batch[1251]	Time cost=399.94	Train-metric=0.017897
[1,0]<stderr>:INFO:root:Epoch[242]	Speed: 3203.08 samples/sec
[1,0]<stderr>:INFO:root:Epoch[243] Batch[100] Loss[2.283]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[100]	rmse=0.017951	lr=0.010244
[1,0]<stderr>:INFO:root:Epoch[243] Batch[200] Loss[2.223]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[200]	rmse=0.017888	lr=0.010184
[1,0]<stderr>:INFO:root:Epoch[243] Batch[300] Loss[4.650]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[300]	rmse=0.017893	lr=0.010124
[1,0]<stderr>:INFO:root:Epoch[243] Batch[400] Loss[2.169]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[400]	rmse=0.017905	lr=0.010065
[1,0]<stderr>:INFO:root:Epoch[243] Batch[500] Loss[4.900]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[500]	rmse=0.017869	lr=0.010006
[1,0]<stderr>:INFO:root:Epoch[243] Batch[600] Loss[3.333]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[600]	rmse=0.017852	lr=0.009947
[1,0]<stderr>:INFO:root:Epoch[243] Batch[700] Loss[3.512]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[700]	rmse=0.017829	lr=0.009888
[1,0]<stderr>:INFO:root:Epoch[243] Batch[800] Loss[2.269]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[800]	rmse=0.017821	lr=0.009829
[1,0]<stderr>:INFO:root:Epoch[243] Batch[900] Loss[2.306]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[900]	rmse=0.017837	lr=0.009771
[1,0]<stderr>:INFO:root:Epoch[243] Batch[1000] Loss[4.639]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[1000]	rmse=0.017844	lr=0.009712
[1,0]<stderr>:INFO:root:Epoch[243] Batch[1100] Loss[2.575]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[1100]	rmse=0.017833	lr=0.009654
[1,0]<stderr>:INFO:root:Epoch[243] Batch[1200] Loss[2.644]
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[1200]	rmse=0.017834	lr=0.009596
[1,0]<stderr>:INFO:root:Epoch[243] Rank[0] Batch[1251]	Time cost=398.95	Train-metric=0.017839
[1,0]<stderr>:INFO:root:Epoch[243]	Speed: 3211.01 samples/sec
[1,0]<stderr>:INFO:root:Epoch[244] Batch[100] Loss[2.411]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[100]	rmse=0.017812	lr=0.009509
[1,0]<stderr>:INFO:root:Epoch[244] Batch[200] Loss[4.755]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[200]	rmse=0.017844	lr=0.009451
[1,0]<stderr>:INFO:root:Epoch[244] Batch[300] Loss[2.132]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[300]	rmse=0.017815	lr=0.009394
[1,0]<stderr>:INFO:root:Epoch[244] Batch[400] Loss[4.749]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[400]	rmse=0.017839	lr=0.009336
[1,0]<stderr>:INFO:root:Epoch[244] Batch[500] Loss[4.510]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[500]	rmse=0.017833	lr=0.009279
[1,0]<stderr>:INFO:root:Epoch[244] Batch[600] Loss[2.358]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[600]	rmse=0.017859	lr=0.009222
[1,0]<stderr>:INFO:root:Epoch[244] Batch[700] Loss[2.114]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[700]	rmse=0.017841	lr=0.009165
[1,0]<stderr>:INFO:root:Epoch[244] Batch[800] Loss[4.755]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[800]	rmse=0.017839	lr=0.009109
[1,0]<stderr>:INFO:root:Epoch[244] Batch[900] Loss[2.244]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[900]	rmse=0.017844	lr=0.009052
[1,0]<stderr>:INFO:root:Epoch[244] Batch[1000] Loss[2.814]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[1000]	rmse=0.017856	lr=0.008996
[1,0]<stderr>:INFO:root:Epoch[244] Batch[1100] Loss[2.736]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[1100]	rmse=0.017847	lr=0.008940
[1,0]<stderr>:INFO:root:Epoch[244] Batch[1200] Loss[2.093]
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[1200]	rmse=0.017842	lr=0.008884
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0] Batch[1251]	Time cost=400.08	Train-metric=0.017843
[1,0]<stderr>:INFO:root:Epoch[244]	Speed: 3201.95 samples/sec
[1,0]<stderr>:INFO:root:Epoch[244] Rank[0]	Validation-accuracy=0.747260	Validation-top_k_accuracy_5=0.920760
[1,0]<stderr>:INFO:root:Epoch[245] Batch[100] Loss[2.231]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[100]	rmse=0.017810	lr=0.008800
[1,0]<stderr>:INFO:root:Epoch[245] Batch[200] Loss[3.150]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[200]	rmse=0.017774	lr=0.008745
[1,0]<stderr>:INFO:root:Epoch[245] Batch[300] Loss[4.761]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[300]	rmse=0.017747	lr=0.008689
[1,0]<stderr>:INFO:root:Epoch[245] Batch[400] Loss[3.325]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[400]	rmse=0.017724	lr=0.008634
[1,0]<stderr>:INFO:root:Epoch[245] Batch[500] Loss[2.245]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[500]	rmse=0.017713	lr=0.008579
[1,0]<stderr>:INFO:root:Epoch[245] Batch[600] Loss[2.958]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[600]	rmse=0.017735	lr=0.008525
[1,0]<stderr>:INFO:root:Epoch[245] Batch[700] Loss[2.488]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[700]	rmse=0.017746	lr=0.008470
[1,0]<stderr>:INFO:root:Epoch[245] Batch[800] Loss[2.159]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[800]	rmse=0.017759	lr=0.008415
[1,0]<stderr>:INFO:root:Epoch[245] Batch[900] Loss[2.176]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[900]	rmse=0.017742	lr=0.008361
[1,0]<stderr>:INFO:root:Epoch[245] Batch[1000] Loss[2.207]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[1000]	rmse=0.017756	lr=0.008307
[1,0]<stderr>:INFO:root:Epoch[245] Batch[1100] Loss[2.223]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[1100]	rmse=0.017766	lr=0.008253
[1,0]<stderr>:INFO:root:Epoch[245] Batch[1200] Loss[4.611]
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[1200]	rmse=0.017770	lr=0.008199
[1,0]<stderr>:INFO:root:Epoch[245] Rank[0] Batch[1251]	Time cost=398.41	Train-metric=0.017779
[1,0]<stderr>:INFO:root:Epoch[245]	Speed: 3215.33 samples/sec
[1,0]<stderr>:INFO:root:Epoch[246] Batch[100] Loss[2.218]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[100]	rmse=0.017727	lr=0.008119
[1,0]<stderr>:INFO:root:Epoch[246] Batch[200] Loss[3.928]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[200]	rmse=0.017739	lr=0.008065
[1,0]<stderr>:INFO:root:Epoch[246] Batch[300] Loss[2.328]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[300]	rmse=0.017768	lr=0.008012
[1,0]<stderr>:INFO:root:Epoch[246] Batch[400] Loss[2.820]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[400]	rmse=0.017732	lr=0.007959
[1,0]<stderr>:INFO:root:Epoch[246] Batch[500] Loss[4.676]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[500]	rmse=0.017719	lr=0.007906
[1,0]<stderr>:INFO:root:Epoch[246] Batch[600] Loss[4.174]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[600]	rmse=0.017723	lr=0.007854
[1,0]<stderr>:INFO:root:Epoch[246] Batch[700] Loss[2.102]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[700]	rmse=0.017735	lr=0.007801
[1,0]<stderr>:INFO:root:Epoch[246] Batch[800] Loss[2.479]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[800]	rmse=0.017733	lr=0.007749
[1,0]<stderr>:INFO:root:Epoch[246] Batch[900] Loss[4.796]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[900]	rmse=0.017713	lr=0.007697
[1,0]<stderr>:INFO:root:Epoch[246] Batch[1000] Loss[4.075]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[1000]	rmse=0.017721	lr=0.007645
[1,0]<stderr>:INFO:root:Epoch[246] Batch[1100] Loss[2.530]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[1100]	rmse=0.017725	lr=0.007593
[1,0]<stderr>:INFO:root:Epoch[246] Batch[1200] Loss[4.610]
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[1200]	rmse=0.017736	lr=0.007542
[1,0]<stderr>:INFO:root:Epoch[246] Rank[0] Batch[1251]	Time cost=398.93	Train-metric=0.017733
[1,0]<stderr>:INFO:root:Epoch[246]	Speed: 3211.16 samples/sec
[1,0]<stderr>:INFO:root:Epoch[247] Batch[100] Loss[2.247]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[100]	rmse=0.017661	lr=0.007464
[1,0]<stderr>:INFO:root:Epoch[247] Batch[200] Loss[1.992]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[200]	rmse=0.017692	lr=0.007413
[1,0]<stderr>:INFO:root:Epoch[247] Batch[300] Loss[2.150]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[300]	rmse=0.017697	lr=0.007362
[1,0]<stderr>:INFO:root:Epoch[247] Batch[400] Loss[2.466]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[400]	rmse=0.017682	lr=0.007311
[1,0]<stderr>:INFO:root:Epoch[247] Batch[500] Loss[4.202]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[500]	rmse=0.017677	lr=0.007260
[1,0]<stderr>:INFO:root:Epoch[247] Batch[600] Loss[2.360]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[600]	rmse=0.017685	lr=0.007210
[1,0]<stderr>:INFO:root:Epoch[247] Batch[700] Loss[4.814]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[700]	rmse=0.017687	lr=0.007159
[1,0]<stderr>:INFO:root:Epoch[247] Batch[800] Loss[2.605]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[800]	rmse=0.017693	lr=0.007109
[1,0]<stderr>:INFO:root:Epoch[247] Batch[900] Loss[4.484]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[900]	rmse=0.017699	lr=0.007059
[1,0]<stderr>:INFO:root:Epoch[247] Batch[1000] Loss[2.270]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[1000]	rmse=0.017705	lr=0.007010
[1,0]<stderr>:INFO:root:Epoch[247] Batch[1100] Loss[3.118]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[1100]	rmse=0.017699	lr=0.006960
[1,0]<stderr>:INFO:root:Epoch[247] Batch[1200] Loss[2.338]
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[1200]	rmse=0.017707	lr=0.006911
[1,0]<stderr>:INFO:root:Epoch[247] Rank[0] Batch[1251]	Time cost=399.13	Train-metric=0.017713
[1,0]<stderr>:INFO:root:Epoch[247]	Speed: 3209.56 samples/sec
[1,0]<stderr>:INFO:root:Epoch[248] Batch[100] Loss[2.172]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[100]	rmse=0.017550	lr=0.006836
[1,0]<stderr>:INFO:root:Epoch[248] Batch[200] Loss[4.440]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[200]	rmse=0.017566	lr=0.006787
[1,0]<stderr>:INFO:root:Epoch[248] Batch[300] Loss[3.036]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[300]	rmse=0.017614	lr=0.006738
[1,0]<stderr>:INFO:root:Epoch[248] Batch[400] Loss[3.230]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[400]	rmse=0.017630	lr=0.006690
[1,0]<stderr>:INFO:root:Epoch[248] Batch[500] Loss[2.205]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[500]	rmse=0.017631	lr=0.006641
[1,0]<stderr>:INFO:root:Epoch[248] Batch[600] Loss[3.973]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[600]	rmse=0.017636	lr=0.006593
[1,0]<stderr>:INFO:root:Epoch[248] Batch[700] Loss[2.153]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[700]	rmse=0.017637	lr=0.006545
[1,0]<stderr>:INFO:root:Epoch[248] Batch[800] Loss[4.294]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[800]	rmse=0.017645	lr=0.006497
[1,0]<stderr>:INFO:root:Epoch[248] Batch[900] Loss[3.439]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[900]	rmse=0.017639	lr=0.006449
[1,0]<stderr>:INFO:root:Epoch[248] Batch[1000] Loss[2.111]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[1000]	rmse=0.017647	lr=0.006401
[1,0]<stderr>:INFO:root:Epoch[248] Batch[1100] Loss[2.080]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[1100]	rmse=0.017664	lr=0.006354
[1,0]<stderr>:INFO:root:Epoch[248] Batch[1200] Loss[2.493]
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[1200]	rmse=0.017659	lr=0.006307
[1,0]<stderr>:INFO:root:Epoch[248] Rank[0] Batch[1251]	Time cost=399.93	Train-metric=0.017663
[1,0]<stderr>:INFO:root:Epoch[248]	Speed: 3203.10 samples/sec
[1,0]<stderr>:INFO:root:Epoch[249] Batch[100] Loss[4.404]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[100]	rmse=0.017533	lr=0.006236
[1,0]<stderr>:INFO:root:Epoch[249] Batch[200] Loss[2.216]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[200]	rmse=0.017626	lr=0.006189
[1,0]<stderr>:INFO:root:Epoch[249] Batch[300] Loss[2.214]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[300]	rmse=0.017640	lr=0.006142
[1,0]<stderr>:INFO:root:Epoch[249] Batch[400] Loss[2.097]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[400]	rmse=0.017609	lr=0.006096
[1,0]<stderr>:INFO:root:Epoch[249] Batch[500] Loss[2.685]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[500]	rmse=0.017590	lr=0.006049
[1,0]<stderr>:INFO:root:Epoch[249] Batch[600] Loss[2.119]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[600]	rmse=0.017597	lr=0.006003
[1,0]<stderr>:INFO:root:Epoch[249] Batch[700] Loss[2.135]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[700]	rmse=0.017579	lr=0.005957
[1,0]<stderr>:INFO:root:Epoch[249] Batch[800] Loss[2.133]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[800]	rmse=0.017610	lr=0.005912
[1,0]<stderr>:INFO:root:Epoch[249] Batch[900] Loss[2.271]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[900]	rmse=0.017622	lr=0.005866
[1,0]<stderr>:INFO:root:Epoch[249] Batch[1000] Loss[4.339]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[1000]	rmse=0.017630	lr=0.005820
[1,0]<stderr>:INFO:root:Epoch[249] Batch[1100] Loss[2.167]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[1100]	rmse=0.017632	lr=0.005775
[1,0]<stderr>:INFO:root:Epoch[249] Batch[1200] Loss[2.891]
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[1200]	rmse=0.017637	lr=0.005730
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0] Batch[1251]	Time cost=399.26	Train-metric=0.017637
[1,0]<stderr>:INFO:root:Epoch[249]	Speed: 3208.50 samples/sec
[1,0]<stderr>:INFO:root:Epoch[249] Rank[0]	Validation-accuracy=0.751920	Validation-top_k_accuracy_5=0.923380
[1,0]<stderr>:INFO:root:Epoch[250] Batch[100] Loss[3.357]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[100]	rmse=0.017636	lr=0.005662
[1,0]<stderr>:INFO:root:Epoch[250] Batch[200] Loss[4.455]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[200]	rmse=0.017598	lr=0.005618
[1,0]<stderr>:INFO:root:Epoch[250] Batch[300] Loss[2.265]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[300]	rmse=0.017598	lr=0.005573
[1,0]<stderr>:INFO:root:Epoch[250] Batch[400] Loss[3.628]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[400]	rmse=0.017599	lr=0.005529
[1,0]<stderr>:INFO:root:Epoch[250] Batch[500] Loss[4.510]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[500]	rmse=0.017622	lr=0.005485
[1,0]<stderr>:INFO:root:Epoch[250] Batch[600] Loss[1.937]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[600]	rmse=0.017586	lr=0.005441
[1,0]<stderr>:INFO:root:Epoch[250] Batch[700] Loss[2.180]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[700]	rmse=0.017593	lr=0.005397
[1,0]<stderr>:INFO:root:Epoch[250] Batch[800] Loss[2.234]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[800]	rmse=0.017585	lr=0.005353
[1,0]<stderr>:INFO:root:Epoch[250] Batch[900] Loss[2.269]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[900]	rmse=0.017566	lr=0.005310
[1,0]<stderr>:INFO:root:Epoch[250] Batch[1000] Loss[2.422]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[1000]	rmse=0.017573	lr=0.005267
[1,0]<stderr>:INFO:root:Epoch[250] Batch[1100] Loss[2.445]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[1100]	rmse=0.017562	lr=0.005224
[1,0]<stderr>:INFO:root:Epoch[250] Batch[1200] Loss[4.724]
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[1200]	rmse=0.017559	lr=0.005181
[1,0]<stderr>:INFO:root:Epoch[250] Rank[0] Batch[1251]	Time cost=400.41	Train-metric=0.017564
[1,0]<stderr>:INFO:root:Epoch[250]	Speed: 3199.26 samples/sec
[1,0]<stderr>:INFO:root:Epoch[251] Batch[100] Loss[2.687]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[100]	rmse=0.017485	lr=0.005116
[1,0]<stderr>:INFO:root:Epoch[251] Batch[200] Loss[2.708]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[200]	rmse=0.017532	lr=0.005074
[1,0]<stderr>:INFO:root:Epoch[251] Batch[300] Loss[2.173]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[300]	rmse=0.017520	lr=0.005031
[1,0]<stderr>:INFO:root:Epoch[251] Batch[400] Loss[2.181]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[400]	rmse=0.017557	lr=0.004989
[1,0]<stderr>:INFO:root:Epoch[251] Batch[500] Loss[4.670]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[500]	rmse=0.017557	lr=0.004947
[1,0]<stderr>:INFO:root:Epoch[251] Batch[600] Loss[1.892]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[600]	rmse=0.017559	lr=0.004906
[1,0]<stderr>:INFO:root:Epoch[251] Batch[700] Loss[2.078]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[700]	rmse=0.017559	lr=0.004864
[1,0]<stderr>:INFO:root:Epoch[251] Batch[800] Loss[4.783]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[800]	rmse=0.017536	lr=0.004823
[1,0]<stderr>:INFO:root:Epoch[251] Batch[900] Loss[2.080]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[900]	rmse=0.017534	lr=0.004781
[1,0]<stderr>:INFO:root:Epoch[251] Batch[1000] Loss[3.056]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[1000]	rmse=0.017531	lr=0.004740
[1,0]<stderr>:INFO:root:Epoch[251] Batch[1100] Loss[2.172]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[1100]	rmse=0.017533	lr=0.004699
[1,0]<stderr>:INFO:root:Epoch[251] Batch[1200] Loss[3.791]
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[1200]	rmse=0.017537	lr=0.004659
[1,0]<stderr>:INFO:root:Epoch[251] Rank[0] Batch[1251]	Time cost=400.06	Train-metric=0.017547
[1,0]<stderr>:INFO:root:Epoch[251]	Speed: 3202.05 samples/sec
[1,0]<stderr>:INFO:root:Epoch[252] Batch[100] Loss[1.998]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[100]	rmse=0.017619	lr=0.004597
[1,0]<stderr>:INFO:root:Epoch[252] Batch[200] Loss[4.450]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[200]	rmse=0.017434	lr=0.004557
[1,0]<stderr>:INFO:root:Epoch[252] Batch[300] Loss[2.533]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[300]	rmse=0.017484	lr=0.004517
[1,0]<stderr>:INFO:root:Epoch[252] Batch[400] Loss[2.207]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[400]	rmse=0.017498	lr=0.004477
[1,0]<stderr>:INFO:root:Epoch[252] Batch[500] Loss[2.190]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[500]	rmse=0.017517	lr=0.004437
[1,0]<stderr>:INFO:root:Epoch[252] Batch[600] Loss[4.264]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[600]	rmse=0.017533	lr=0.004398
[1,0]<stderr>:INFO:root:Epoch[252] Batch[700] Loss[2.595]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[700]	rmse=0.017531	lr=0.004358
[1,0]<stderr>:INFO:root:Epoch[252] Batch[800] Loss[2.702]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[800]	rmse=0.017518	lr=0.004319
[1,0]<stderr>:INFO:root:Epoch[252] Batch[900] Loss[2.370]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[900]	rmse=0.017522	lr=0.004280
[1,0]<stderr>:INFO:root:Epoch[252] Batch[1000] Loss[2.947]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[1000]	rmse=0.017529	lr=0.004241
[1,0]<stderr>:INFO:root:Epoch[252] Batch[1100] Loss[2.239]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[1100]	rmse=0.017533	lr=0.004203
[1,0]<stderr>:INFO:root:Epoch[252] Batch[1200] Loss[3.157]
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[1200]	rmse=0.017532	lr=0.004164
[1,0]<stderr>:INFO:root:Epoch[252] Rank[0] Batch[1251]	Time cost=401.33	Train-metric=0.017525
[1,0]<stderr>:INFO:root:Epoch[252]	Speed: 3191.91 samples/sec
[1,0]<stderr>:INFO:root:Epoch[253] Batch[100] Loss[2.616]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[100]	rmse=0.017512	lr=0.004106
[1,0]<stderr>:INFO:root:Epoch[253] Batch[200] Loss[1.867]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[200]	rmse=0.017495	lr=0.004068
[1,0]<stderr>:INFO:root:Epoch[253] Batch[300] Loss[2.218]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[300]	rmse=0.017515	lr=0.004030
[1,0]<stderr>:INFO:root:Epoch[253] Batch[400] Loss[2.543]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[400]	rmse=0.017521	lr=0.003992
[1,0]<stderr>:INFO:root:Epoch[253] Batch[500] Loss[2.130]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[500]	rmse=0.017537	lr=0.003955
[1,0]<stderr>:INFO:root:Epoch[253] Batch[600] Loss[2.085]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[600]	rmse=0.017512	lr=0.003917
[1,0]<stderr>:INFO:root:Epoch[253] Batch[700] Loss[4.058]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[700]	rmse=0.017501	lr=0.003880
[1,0]<stderr>:INFO:root:Epoch[253] Batch[800] Loss[2.433]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[800]	rmse=0.017507	lr=0.003843
[1,0]<stderr>:INFO:root:Epoch[253] Batch[900] Loss[2.212]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[900]	rmse=0.017505	lr=0.003806
[1,0]<stderr>:INFO:root:Epoch[253] Batch[1000] Loss[1.956]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[1000]	rmse=0.017506	lr=0.003770
[1,0]<stderr>:INFO:root:Epoch[253] Batch[1100] Loss[2.147]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[1100]	rmse=0.017506	lr=0.003733
[1,0]<stderr>:INFO:root:Epoch[253] Batch[1200] Loss[2.018]
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[1200]	rmse=0.017508	lr=0.003697
[1,0]<stderr>:INFO:root:Epoch[253] Rank[0] Batch[1251]	Time cost=401.63	Train-metric=0.017511
[1,0]<stderr>:INFO:root:Epoch[253]	Speed: 3189.55 samples/sec
[1,0]<stderr>:INFO:root:Epoch[254] Batch[100] Loss[4.449]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[100]	rmse=0.017384	lr=0.003642
[1,0]<stderr>:INFO:root:Epoch[254] Batch[200] Loss[1.913]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[200]	rmse=0.017402	lr=0.003606
[1,0]<stderr>:INFO:root:Epoch[254] Batch[300] Loss[2.217]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[300]	rmse=0.017387	lr=0.003571
[1,0]<stderr>:INFO:root:Epoch[254] Batch[400] Loss[2.424]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[400]	rmse=0.017376	lr=0.003535
[1,0]<stderr>:INFO:root:Epoch[254] Batch[500] Loss[2.025]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[500]	rmse=0.017368	lr=0.003500
[1,0]<stderr>:INFO:root:Epoch[254] Batch[600] Loss[1.999]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[600]	rmse=0.017364	lr=0.003465
[1,0]<stderr>:INFO:root:Epoch[254] Batch[700] Loss[2.673]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[700]	rmse=0.017388	lr=0.003430
[1,0]<stderr>:INFO:root:Epoch[254] Batch[800] Loss[2.709]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[800]	rmse=0.017401	lr=0.003395
[1,0]<stderr>:INFO:root:Epoch[254] Batch[900] Loss[2.377]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[900]	rmse=0.017408	lr=0.003360
[1,0]<stderr>:INFO:root:Epoch[254] Batch[1000] Loss[4.696]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[1000]	rmse=0.017404	lr=0.003326
[1,0]<stderr>:INFO:root:Epoch[254] Batch[1100] Loss[2.360]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[1100]	rmse=0.017409	lr=0.003291
[1,0]<stderr>:INFO:root:Epoch[254] Batch[1200] Loss[4.644]
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[1200]	rmse=0.017422	lr=0.003257
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0] Batch[1251]	Time cost=400.29	Train-metric=0.017430
[1,0]<stderr>:INFO:root:Epoch[254]	Speed: 3200.23 samples/sec
[1,0]<stderr>:INFO:root:Epoch[254] Rank[0]	Validation-accuracy=0.754500	Validation-top_k_accuracy_5=0.925100
[1,0]<stderr>:INFO:root:Epoch[255] Batch[100] Loss[2.094]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[100]	rmse=0.017456	lr=0.003206
[1,0]<stderr>:INFO:root:Epoch[255] Batch[200] Loss[2.480]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[200]	rmse=0.017391	lr=0.003172
[1,0]<stderr>:INFO:root:Epoch[255] Batch[300] Loss[4.364]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[300]	rmse=0.017423	lr=0.003139
[1,0]<stderr>:INFO:root:Epoch[255] Batch[400] Loss[3.275]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[400]	rmse=0.017402	lr=0.003105
[1,0]<stderr>:INFO:root:Epoch[255] Batch[500] Loss[4.836]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[500]	rmse=0.017395	lr=0.003072
[1,0]<stderr>:INFO:root:Epoch[255] Batch[600] Loss[2.108]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[600]	rmse=0.017406	lr=0.003039
[1,0]<stderr>:INFO:root:Epoch[255] Batch[700] Loss[4.073]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[700]	rmse=0.017420	lr=0.003007
[1,0]<stderr>:INFO:root:Epoch[255] Batch[800] Loss[2.939]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[800]	rmse=0.017419	lr=0.002974
[1,0]<stderr>:INFO:root:Epoch[255] Batch[900] Loss[2.079]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[900]	rmse=0.017411	lr=0.002941
[1,0]<stderr>:INFO:root:Epoch[255] Batch[1000] Loss[2.255]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[1000]	rmse=0.017413	lr=0.002909
[1,0]<stderr>:INFO:root:Epoch[255] Batch[1100] Loss[2.188]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[1100]	rmse=0.017412	lr=0.002877
[1,0]<stderr>:INFO:root:Epoch[255] Batch[1200] Loss[4.341]
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[1200]	rmse=0.017428	lr=0.002845
[1,0]<stderr>:INFO:root:Epoch[255] Rank[0] Batch[1251]	Time cost=399.17	Train-metric=0.017422
[1,0]<stderr>:INFO:root:Epoch[255]	Speed: 3209.23 samples/sec
[1,0]<stderr>:INFO:root:Epoch[256] Batch[100] Loss[3.863]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[100]	rmse=0.017428	lr=0.002797
[1,0]<stderr>:INFO:root:Epoch[256] Batch[200] Loss[2.447]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[200]	rmse=0.017327	lr=0.002766
[1,0]<stderr>:INFO:root:Epoch[256] Batch[300] Loss[2.247]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[300]	rmse=0.017368	lr=0.002735
[1,0]<stderr>:INFO:root:Epoch[256] Batch[400] Loss[3.791]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[400]	rmse=0.017392	lr=0.002703
[1,0]<stderr>:INFO:root:Epoch[256] Batch[500] Loss[2.245]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[500]	rmse=0.017377	lr=0.002673
[1,0]<stderr>:INFO:root:Epoch[256] Batch[600] Loss[2.011]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[600]	rmse=0.017365	lr=0.002642
[1,0]<stderr>:INFO:root:Epoch[256] Batch[700] Loss[3.894]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[700]	rmse=0.017387	lr=0.002611
[1,0]<stderr>:INFO:root:Epoch[256] Batch[800] Loss[2.020]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[800]	rmse=0.017387	lr=0.002581
[1,0]<stderr>:INFO:root:Epoch[256] Batch[900] Loss[4.103]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[900]	rmse=0.017381	lr=0.002550
[1,0]<stderr>:INFO:root:Epoch[256] Batch[1000] Loss[2.391]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[1000]	rmse=0.017384	lr=0.002520
[1,0]<stderr>:INFO:root:Epoch[256] Batch[1100] Loss[4.746]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[1100]	rmse=0.017403	lr=0.002491
[1,0]<stderr>:INFO:root:Epoch[256] Batch[1200] Loss[2.301]
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[1200]	rmse=0.017408	lr=0.002461
[1,0]<stderr>:INFO:root:Epoch[256] Rank[0] Batch[1251]	Time cost=399.15	Train-metric=0.017400
[1,0]<stderr>:INFO:root:Epoch[256]	Speed: 3209.37 samples/sec
[1,0]<stderr>:INFO:root:Epoch[257] Batch[100] Loss[4.748]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[100]	rmse=0.017307	lr=0.002416
[1,0]<stderr>:INFO:root:Epoch[257] Batch[200] Loss[2.383]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[200]	rmse=0.017301	lr=0.002387
[1,0]<stderr>:INFO:root:Epoch[257] Batch[300] Loss[4.270]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[300]	rmse=0.017301	lr=0.002358
[1,0]<stderr>:INFO:root:Epoch[257] Batch[400] Loss[2.145]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[400]	rmse=0.017305	lr=0.002329
[1,0]<stderr>:INFO:root:Epoch[257] Batch[500] Loss[4.488]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[500]	rmse=0.017320	lr=0.002300
[1,0]<stderr>:INFO:root:Epoch[257] Batch[600] Loss[2.165]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[600]	rmse=0.017335	lr=0.002272
[1,0]<stderr>:INFO:root:Epoch[257] Batch[700] Loss[2.474]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[700]	rmse=0.017326	lr=0.002243
[1,0]<stderr>:INFO:root:Epoch[257] Batch[800] Loss[3.655]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[800]	rmse=0.017341	lr=0.002215
[1,0]<stderr>:INFO:root:Epoch[257] Batch[900] Loss[1.894]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[900]	rmse=0.017335	lr=0.002187
[1,0]<stderr>:INFO:root:Epoch[257] Batch[1000] Loss[2.141]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[1000]	rmse=0.017346	lr=0.002159
[1,0]<stderr>:INFO:root:Epoch[257] Batch[1100] Loss[2.954]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[1100]	rmse=0.017352	lr=0.002132
[1,0]<stderr>:INFO:root:Epoch[257] Batch[1200] Loss[2.627]
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[1200]	rmse=0.017363	lr=0.002104
[1,0]<stderr>:INFO:root:Epoch[257] Rank[0] Batch[1251]	Time cost=399.89	Train-metric=0.017368
[1,0]<stderr>:INFO:root:Epoch[257]	Speed: 3203.48 samples/sec
[1,0]<stderr>:INFO:root:Epoch[258] Batch[100] Loss[2.312]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[100]	rmse=0.017211	lr=0.002063
[1,0]<stderr>:INFO:root:Epoch[258] Batch[200] Loss[2.771]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[200]	rmse=0.017263	lr=0.002036
[1,0]<stderr>:INFO:root:Epoch[258] Batch[300] Loss[2.075]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[300]	rmse=0.017278	lr=0.002009
[1,0]<stderr>:INFO:root:Epoch[258] Batch[400] Loss[4.227]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[400]	rmse=0.017323	lr=0.001982
[1,0]<stderr>:INFO:root:Epoch[258] Batch[500] Loss[2.260]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[500]	rmse=0.017350	lr=0.001956
[1,0]<stderr>:INFO:root:Epoch[258] Batch[600] Loss[2.383]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[600]	rmse=0.017344	lr=0.001930
[1,0]<stderr>:INFO:root:Epoch[258] Batch[700] Loss[4.011]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[700]	rmse=0.017344	lr=0.001903
[1,0]<stderr>:INFO:root:Epoch[258] Batch[800] Loss[3.939]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[800]	rmse=0.017341	lr=0.001877
[1,0]<stderr>:INFO:root:Epoch[258] Batch[900] Loss[4.151]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[900]	rmse=0.017349	lr=0.001852
[1,0]<stderr>:INFO:root:Epoch[258] Batch[1000] Loss[1.833]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[1000]	rmse=0.017348	lr=0.001826
[1,0]<stderr>:INFO:root:Epoch[258] Batch[1100] Loss[1.869]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[1100]	rmse=0.017333	lr=0.001801
[1,0]<stderr>:INFO:root:Epoch[258] Batch[1200] Loss[2.003]
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[1200]	rmse=0.017336	lr=0.001775
[1,0]<stderr>:INFO:root:Epoch[258] Rank[0] Batch[1251]	Time cost=399.96	Train-metric=0.017339
[1,0]<stderr>:INFO:root:Epoch[258]	Speed: 3202.91 samples/sec
[1,0]<stderr>:INFO:root:Epoch[259] Batch[100] Loss[1.965]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[100]	rmse=0.017234	lr=0.001738
[1,0]<stderr>:INFO:root:Epoch[259] Batch[200] Loss[1.967]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[200]	rmse=0.017191	lr=0.001713
[1,0]<stderr>:INFO:root:Epoch[259] Batch[300] Loss[2.380]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[300]	rmse=0.017305	lr=0.001688
[1,0]<stderr>:INFO:root:Epoch[259] Batch[400] Loss[2.081]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[400]	rmse=0.017319	lr=0.001664
[1,0]<stderr>:INFO:root:Epoch[259] Batch[500] Loss[4.133]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[500]	rmse=0.017342	lr=0.001639
[1,0]<stderr>:INFO:root:Epoch[259] Batch[600] Loss[3.087]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[600]	rmse=0.017341	lr=0.001615
[1,0]<stderr>:INFO:root:Epoch[259] Batch[700] Loss[2.256]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[700]	rmse=0.017358	lr=0.001591
[1,0]<stderr>:INFO:root:Epoch[259] Batch[800] Loss[2.064]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[800]	rmse=0.017346	lr=0.001568
[1,0]<stderr>:INFO:root:Epoch[259] Batch[900] Loss[2.154]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[900]	rmse=0.017342	lr=0.001544
[1,0]<stderr>:INFO:root:Epoch[259] Batch[1000] Loss[1.936]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[1000]	rmse=0.017329	lr=0.001521
[1,0]<stderr>:INFO:root:Epoch[259] Batch[1100] Loss[1.959]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[1100]	rmse=0.017336	lr=0.001497
[1,0]<stderr>:INFO:root:Epoch[259] Batch[1200] Loss[2.141]
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[1200]	rmse=0.017351	lr=0.001474
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0] Batch[1251]	Time cost=399.90	Train-metric=0.017346
[1,0]<stderr>:INFO:root:Epoch[259]	Speed: 3203.33 samples/sec
[1,0]<stderr>:INFO:root:Epoch[259] Rank[0]	Validation-accuracy=0.757360	Validation-top_k_accuracy_5=0.926780
[1,0]<stderr>:INFO:root:Epoch[260] Batch[100] Loss[2.030]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[100]	rmse=0.017196	lr=0.001440
[1,0]<stderr>:INFO:root:Epoch[260] Batch[200] Loss[2.085]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[200]	rmse=0.017232	lr=0.001417
[1,0]<stderr>:INFO:root:Epoch[260] Batch[300] Loss[4.560]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[300]	rmse=0.017251	lr=0.001395
[1,0]<stderr>:INFO:root:Epoch[260] Batch[400] Loss[4.102]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[400]	rmse=0.017235	lr=0.001373
[1,0]<stderr>:INFO:root:Epoch[260] Batch[500] Loss[1.922]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[500]	rmse=0.017240	lr=0.001351
[1,0]<stderr>:INFO:root:Epoch[260] Batch[600] Loss[2.102]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[600]	rmse=0.017260	lr=0.001329
[1,0]<stderr>:INFO:root:Epoch[260] Batch[700] Loss[2.011]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[700]	rmse=0.017252	lr=0.001307
[1,0]<stderr>:INFO:root:Epoch[260] Batch[800] Loss[4.566]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[800]	rmse=0.017265	lr=0.001285
[1,0]<stderr>:INFO:root:Epoch[260] Batch[900] Loss[3.551]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[900]	rmse=0.017264	lr=0.001264
[1,0]<stderr>:INFO:root:Epoch[260] Batch[1000] Loss[2.439]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[1000]	rmse=0.017280	lr=0.001243
[1,0]<stderr>:INFO:root:Epoch[260] Batch[1100] Loss[4.135]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[1100]	rmse=0.017279	lr=0.001222
[1,0]<stderr>:INFO:root:Epoch[260] Batch[1200] Loss[2.281]
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[1200]	rmse=0.017276	lr=0.001201
[1,0]<stderr>:INFO:root:Epoch[260] Rank[0] Batch[1251]	Time cost=397.17	Train-metric=0.017283
[1,0]<stderr>:INFO:root:Epoch[260]	Speed: 3225.39 samples/sec
[1,0]<stderr>:INFO:root:Epoch[261] Batch[100] Loss[2.058]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[100]	rmse=0.017402	lr=0.001170
[1,0]<stderr>:INFO:root:Epoch[261] Batch[200] Loss[2.228]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[200]	rmse=0.017290	lr=0.001150
[1,0]<stderr>:INFO:root:Epoch[261] Batch[300] Loss[3.992]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[300]	rmse=0.017257	lr=0.001129
[1,0]<stderr>:INFO:root:Epoch[261] Batch[400] Loss[4.443]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[400]	rmse=0.017244	lr=0.001109
[1,0]<stderr>:INFO:root:Epoch[261] Batch[500] Loss[2.110]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[500]	rmse=0.017248	lr=0.001090
[1,0]<stderr>:INFO:root:Epoch[261] Batch[600] Loss[2.419]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[600]	rmse=0.017247	lr=0.001070
[1,0]<stderr>:INFO:root:Epoch[261] Batch[700] Loss[1.778]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[700]	rmse=0.017242	lr=0.001050
[1,0]<stderr>:INFO:root:Epoch[261] Batch[800] Loss[4.466]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[800]	rmse=0.017253	lr=0.001031
[1,0]<stderr>:INFO:root:Epoch[261] Batch[900] Loss[2.162]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[900]	rmse=0.017268	lr=0.001012
[1,0]<stderr>:INFO:root:Epoch[261] Batch[1000] Loss[1.834]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[1000]	rmse=0.017274	lr=0.000993
[1,0]<stderr>:INFO:root:Epoch[261] Batch[1100] Loss[2.028]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[1100]	rmse=0.017278	lr=0.000974
[1,0]<stderr>:INFO:root:Epoch[261] Batch[1200] Loss[2.116]
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[1200]	rmse=0.017285	lr=0.000956
[1,0]<stderr>:INFO:root:Epoch[261] Rank[0] Batch[1251]	Time cost=399.11	Train-metric=0.017281
[1,0]<stderr>:INFO:root:Epoch[261]	Speed: 3209.69 samples/sec
[1,0]<stderr>:INFO:root:Epoch[262] Batch[100] Loss[2.061]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[100]	rmse=0.017392	lr=0.000928
[1,0]<stderr>:INFO:root:Epoch[262] Batch[200] Loss[2.189]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[200]	rmse=0.017346	lr=0.000910
[1,0]<stderr>:INFO:root:Epoch[262] Batch[300] Loss[2.586]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[300]	rmse=0.017314	lr=0.000892
[1,0]<stderr>:INFO:root:Epoch[262] Batch[400] Loss[2.187]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[400]	rmse=0.017283	lr=0.000874
[1,0]<stderr>:INFO:root:Epoch[262] Batch[500] Loss[2.230]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[500]	rmse=0.017270	lr=0.000857
[1,0]<stderr>:INFO:root:Epoch[262] Batch[600] Loss[3.939]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[600]	rmse=0.017289	lr=0.000839
[1,0]<stderr>:INFO:root:Epoch[262] Batch[700] Loss[2.251]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[700]	rmse=0.017277	lr=0.000822
[1,0]<stderr>:INFO:root:Epoch[262] Batch[800] Loss[2.814]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[800]	rmse=0.017263	lr=0.000805
[1,0]<stderr>:INFO:root:Epoch[262] Batch[900] Loss[2.174]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[900]	rmse=0.017254	lr=0.000788
[1,0]<stderr>:INFO:root:Epoch[262] Batch[1000] Loss[4.750]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[1000]	rmse=0.017263	lr=0.000771
[1,0]<stderr>:INFO:root:Epoch[262] Batch[1100] Loss[4.676]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[1100]	rmse=0.017274	lr=0.000755
[1,0]<stderr>:INFO:root:Epoch[262] Batch[1200] Loss[2.178]
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[1200]	rmse=0.017275	lr=0.000738
[1,0]<stderr>:INFO:root:Epoch[262] Rank[0] Batch[1251]	Time cost=399.46	Train-metric=0.017265
[1,0]<stderr>:INFO:root:Epoch[262]	Speed: 3206.91 samples/sec
[1,0]<stderr>:INFO:root:Epoch[263] Batch[100] Loss[3.285]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[100]	rmse=0.017157	lr=0.000714
[1,0]<stderr>:INFO:root:Epoch[263] Batch[200] Loss[2.494]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[200]	rmse=0.017251	lr=0.000698
[1,0]<stderr>:INFO:root:Epoch[263] Batch[300] Loss[2.724]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[300]	rmse=0.017232	lr=0.000682
[1,0]<stderr>:INFO:root:Epoch[263] Batch[400] Loss[2.710]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[400]	rmse=0.017264	lr=0.000667
[1,0]<stderr>:INFO:root:Epoch[263] Batch[500] Loss[4.703]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[500]	rmse=0.017247	lr=0.000652
[1,0]<stderr>:INFO:root:Epoch[263] Batch[600] Loss[2.103]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[600]	rmse=0.017217	lr=0.000636
[1,0]<stderr>:INFO:root:Epoch[263] Batch[700] Loss[1.953]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[700]	rmse=0.017233	lr=0.000621
[1,0]<stderr>:INFO:root:Epoch[263] Batch[800] Loss[1.983]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[800]	rmse=0.017246	lr=0.000607
[1,0]<stderr>:INFO:root:Epoch[263] Batch[900] Loss[2.076]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[900]	rmse=0.017235	lr=0.000592
[1,0]<stderr>:INFO:root:Epoch[263] Batch[1000] Loss[2.210]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[1000]	rmse=0.017234	lr=0.000577
[1,0]<stderr>:INFO:root:Epoch[263] Batch[1100] Loss[3.337]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[1100]	rmse=0.017238	lr=0.000563
[1,0]<stderr>:INFO:root:Epoch[263] Batch[1200] Loss[1.998]
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[1200]	rmse=0.017236	lr=0.000549
[1,0]<stderr>:INFO:root:Epoch[263] Rank[0] Batch[1251]	Time cost=399.22	Train-metric=0.017238
[1,0]<stderr>:INFO:root:Epoch[263]	Speed: 3208.81 samples/sec
[1,0]<stderr>:INFO:root:Epoch[264] Batch[100] Loss[1.807]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[100]	rmse=0.017111	lr=0.000528
[1,0]<stderr>:INFO:root:Epoch[264] Batch[200] Loss[1.874]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[200]	rmse=0.017206	lr=0.000514
[1,0]<stderr>:INFO:root:Epoch[264] Batch[300] Loss[2.982]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[300]	rmse=0.017207	lr=0.000501
[1,0]<stderr>:INFO:root:Epoch[264] Batch[400] Loss[2.172]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[400]	rmse=0.017216	lr=0.000488
[1,0]<stderr>:INFO:root:Epoch[264] Batch[500] Loss[3.567]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[500]	rmse=0.017199	lr=0.000474
[1,0]<stderr>:INFO:root:Epoch[264] Batch[600] Loss[4.602]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[600]	rmse=0.017211	lr=0.000462
[1,0]<stderr>:INFO:root:Epoch[264] Batch[700] Loss[2.183]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[700]	rmse=0.017221	lr=0.000449
[1,0]<stderr>:INFO:root:Epoch[264] Batch[800] Loss[3.683]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[800]	rmse=0.017220	lr=0.000436
[1,0]<stderr>:INFO:root:Epoch[264] Batch[900] Loss[2.132]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[900]	rmse=0.017213	lr=0.000424
[1,0]<stderr>:INFO:root:Epoch[264] Batch[1000] Loss[1.931]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[1000]	rmse=0.017227	lr=0.000412
[1,0]<stderr>:INFO:root:Epoch[264] Batch[1100] Loss[2.603]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[1100]	rmse=0.017243	lr=0.000399
[1,0]<stderr>:INFO:root:Epoch[264] Batch[1200] Loss[2.060]
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[1200]	rmse=0.017248	lr=0.000388
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0] Batch[1251]	Time cost=398.85	Train-metric=0.017253
[1,0]<stderr>:INFO:root:Epoch[264]	Speed: 3211.76 samples/sec
[1,0]<stderr>:INFO:root:Epoch[264] Rank[0]	Validation-accuracy=0.759560	Validation-top_k_accuracy_5=0.927940
[1,0]<stderr>:INFO:root:Epoch[265] Batch[100] Loss[2.259]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[100]	rmse=0.017186	lr=0.000370
[1,0]<stderr>:INFO:root:Epoch[265] Batch[200] Loss[2.252]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[200]	rmse=0.017275	lr=0.000359
[1,0]<stderr>:INFO:root:Epoch[265] Batch[300] Loss[2.079]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[300]	rmse=0.017283	lr=0.000347
[1,0]<stderr>:INFO:root:Epoch[265] Batch[400] Loss[2.032]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[400]	rmse=0.017280	lr=0.000336
[1,0]<stderr>:INFO:root:Epoch[265] Batch[500] Loss[1.969]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[500]	rmse=0.017274	lr=0.000325
[1,0]<stderr>:INFO:root:Epoch[265] Batch[600] Loss[3.110]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[600]	rmse=0.017254	lr=0.000315
[1,0]<stderr>:INFO:root:Epoch[265] Batch[700] Loss[3.185]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[700]	rmse=0.017246	lr=0.000304
[1,0]<stderr>:INFO:root:Epoch[265] Batch[800] Loss[3.282]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[800]	rmse=0.017252	lr=0.000294
[1,0]<stderr>:INFO:root:Epoch[265] Batch[900] Loss[2.712]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[900]	rmse=0.017244	lr=0.000284
[1,0]<stderr>:INFO:root:Epoch[265] Batch[1000] Loss[4.557]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[1000]	rmse=0.017237	lr=0.000274
[1,0]<stderr>:INFO:root:Epoch[265] Batch[1100] Loss[3.566]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[1100]	rmse=0.017237	lr=0.000264
[1,0]<stderr>:INFO:root:Epoch[265] Batch[1200] Loss[2.197]
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[1200]	rmse=0.017242	lr=0.000254
[1,0]<stderr>:INFO:root:Epoch[265] Rank[0] Batch[1251]	Time cost=398.55	Train-metric=0.017246
[1,0]<stderr>:INFO:root:Epoch[265]	Speed: 3214.21 samples/sec
[1,0]<stderr>:INFO:root:Epoch[266] Batch[100] Loss[2.410]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[100]	rmse=0.017238	lr=0.000240
[1,0]<stderr>:INFO:root:Epoch[266] Batch[200] Loss[3.596]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[200]	rmse=0.017308	lr=0.000231
[1,0]<stderr>:INFO:root:Epoch[266] Batch[300] Loss[2.095]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[300]	rmse=0.017271	lr=0.000222
[1,0]<stderr>:INFO:root:Epoch[266] Batch[400] Loss[3.398]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[400]	rmse=0.017260	lr=0.000213
[1,0]<stderr>:INFO:root:Epoch[266] Batch[500] Loss[2.831]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[500]	rmse=0.017237	lr=0.000204
[1,0]<stderr>:INFO:root:Epoch[266] Batch[600] Loss[4.693]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[600]	rmse=0.017216	lr=0.000196
[1,0]<stderr>:INFO:root:Epoch[266] Batch[700] Loss[3.093]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[700]	rmse=0.017203	lr=0.000188
[1,0]<stderr>:INFO:root:Epoch[266] Batch[800] Loss[2.260]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[800]	rmse=0.017210	lr=0.000179
[1,0]<stderr>:INFO:root:Epoch[266] Batch[900] Loss[2.040]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[900]	rmse=0.017220	lr=0.000172
[1,0]<stderr>:INFO:root:Epoch[266] Batch[1000] Loss[2.088]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[1000]	rmse=0.017223	lr=0.000164
[1,0]<stderr>:INFO:root:Epoch[266] Batch[1100] Loss[2.266]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[1100]	rmse=0.017212	lr=0.000156
[1,0]<stderr>:INFO:root:Epoch[266] Batch[1200] Loss[4.662]
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[1200]	rmse=0.017205	lr=0.000149
[1,0]<stderr>:INFO:root:Epoch[266] Rank[0] Batch[1251]	Time cost=398.88	Train-metric=0.017203
[1,0]<stderr>:INFO:root:Epoch[266]	Speed: 3211.55 samples/sec
[1,0]<stderr>:INFO:root:Epoch[267] Batch[100] Loss[2.142]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[100]	rmse=0.017330	lr=0.000138
[1,0]<stderr>:INFO:root:Epoch[267] Batch[200] Loss[1.857]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[200]	rmse=0.017212	lr=0.000131
[1,0]<stderr>:INFO:root:Epoch[267] Batch[300] Loss[3.076]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[300]	rmse=0.017172	lr=0.000124
[1,0]<stderr>:INFO:root:Epoch[267] Batch[400] Loss[2.034]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[400]	rmse=0.017185	lr=0.000118
[1,0]<stderr>:INFO:root:Epoch[267] Batch[500] Loss[2.032]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[500]	rmse=0.017182	lr=0.000111
[1,0]<stderr>:INFO:root:Epoch[267] Batch[600] Loss[2.754]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[600]	rmse=0.017185	lr=0.000105
[1,0]<stderr>:INFO:root:Epoch[267] Batch[700] Loss[4.571]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[700]	rmse=0.017193	lr=0.000099
[1,0]<stderr>:INFO:root:Epoch[267] Batch[800] Loss[2.124]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[800]	rmse=0.017180	lr=0.000093
[1,0]<stderr>:INFO:root:Epoch[267] Batch[900] Loss[2.436]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[900]	rmse=0.017173	lr=0.000087
[1,0]<stderr>:INFO:root:Epoch[267] Batch[1000] Loss[3.540]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[1000]	rmse=0.017188	lr=0.000082
[1,0]<stderr>:INFO:root:Epoch[267] Batch[1100] Loss[3.839]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[1100]	rmse=0.017187	lr=0.000077
[1,0]<stderr>:INFO:root:Epoch[267] Batch[1200] Loss[2.366]
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[1200]	rmse=0.017203	lr=0.000071
[1,0]<stderr>:INFO:root:Epoch[267] Rank[0] Batch[1251]	Time cost=398.63	Train-metric=0.017211
[1,0]<stderr>:INFO:root:Epoch[267]	Speed: 3213.53 samples/sec
[1,0]<stderr>:INFO:root:Epoch[268] Batch[100] Loss[2.768]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[100]	rmse=0.017178	lr=0.000064
[1,0]<stderr>:INFO:root:Epoch[268] Batch[200] Loss[3.630]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[200]	rmse=0.017172	lr=0.000059
[1,0]<stderr>:INFO:root:Epoch[268] Batch[300] Loss[2.170]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[300]	rmse=0.017263	lr=0.000055
[1,0]<stderr>:INFO:root:Epoch[268] Batch[400] Loss[2.091]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[400]	rmse=0.017265	lr=0.000050
[1,0]<stderr>:INFO:root:Epoch[268] Batch[500] Loss[3.077]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[500]	rmse=0.017262	lr=0.000046
[1,0]<stderr>:INFO:root:Epoch[268] Batch[600] Loss[2.144]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[600]	rmse=0.017264	lr=0.000042
[1,0]<stderr>:INFO:root:Epoch[268] Batch[700] Loss[2.206]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[700]	rmse=0.017242	lr=0.000038
[1,0]<stderr>:INFO:root:Epoch[268] Batch[800] Loss[4.643]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[800]	rmse=0.017227	lr=0.000035
[1,0]<stderr>:INFO:root:Epoch[268] Batch[900] Loss[2.754]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[900]	rmse=0.017204	lr=0.000031
[1,0]<stderr>:INFO:root:Epoch[268] Batch[1000] Loss[2.844]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[1000]	rmse=0.017193	lr=0.000028
[1,0]<stderr>:INFO:root:Epoch[268] Batch[1100] Loss[3.554]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[1100]	rmse=0.017198	lr=0.000025
[1,0]<stderr>:INFO:root:Epoch[268] Batch[1200] Loss[2.405]
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[1200]	rmse=0.017200	lr=0.000022
[1,0]<stderr>:INFO:root:Epoch[268] Rank[0] Batch[1251]	Time cost=398.52	Train-metric=0.017194
[1,0]<stderr>:INFO:root:Epoch[268]	Speed: 3214.46 samples/sec
[1,0]<stderr>:INFO:root:Epoch[269] Batch[100] Loss[2.272]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[100]	rmse=0.017212	lr=0.000018
[1,0]<stderr>:INFO:root:Epoch[269] Batch[200] Loss[4.211]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[200]	rmse=0.017199	lr=0.000016
[1,0]<stderr>:INFO:root:Epoch[269] Batch[300] Loss[2.316]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[300]	rmse=0.017238	lr=0.000013
[1,0]<stderr>:INFO:root:Epoch[269] Batch[400] Loss[2.176]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[400]	rmse=0.017195	lr=0.000011
[1,0]<stderr>:INFO:root:Epoch[269] Batch[500] Loss[4.696]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[500]	rmse=0.017196	lr=0.000009
[1,0]<stderr>:INFO:root:Epoch[269] Batch[600] Loss[2.015]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[600]	rmse=0.017202	lr=0.000008
[1,0]<stderr>:INFO:root:Epoch[269] Batch[700] Loss[4.063]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[700]	rmse=0.017215	lr=0.000006
[1,0]<stderr>:INFO:root:Epoch[269] Batch[800] Loss[2.406]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[800]	rmse=0.017213	lr=0.000005
[1,0]<stderr>:INFO:root:Epoch[269] Batch[900] Loss[4.529]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[900]	rmse=0.017206	lr=0.000003
[1,0]<stderr>:INFO:root:Epoch[269] Batch[1000] Loss[2.900]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[1000]	rmse=0.017191	lr=0.000002
[1,0]<stderr>:INFO:root:Epoch[269] Batch[1100] Loss[2.220]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[1100]	rmse=0.017193	lr=0.000002
[1,0]<stderr>:INFO:root:Epoch[269] Batch[1200] Loss[2.029]
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[1200]	rmse=0.017185	lr=0.000001
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0] Batch[1251]	Time cost=399.55	Train-metric=0.017193
[1,0]<stderr>:INFO:root:Epoch[269]	Speed: 3206.17 samples/sec
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0]	Validation-accuracy=0.758760	Validation-top_k_accuracy_5=0.927580
[1,0]<stderr>:INFO:root:Epoch[269] Rank[0]	Validation-accuracy=0.758760	Validation-top_k_accuracy_5=0.927580