Namespace(batch_norm=False, batch_size=256, crop_ratio=0.875, data_dir='~/.mxnet/datasets/imagenet', dtype='float16', hard_weight=0.5, input_size=224, label_smoothing=True, last_gamma=True, log_interval=50, logging_file='mobilenetv3_small.log', lr=2.6, lr_decay=0.1, lr_decay_epoch='40,60', lr_decay_period=0, lr_mode='cosine', mixup=False, mixup_alpha=0.2, mixup_off_epoch=0, mode='hybrid', model='mobilenetv3_small', momentum=0.9, no_wd=True, num_epochs=360, num_gpus=8, num_workers=96, rec_train='/media/ramdisk/rec/train.rec', rec_train_idx='/media/ramdisk/rec/train.idx', rec_val='/media/ramdisk/rec/val.rec', rec_val_idx='/media/ramdisk/rec/val.idx', resume_epoch=0, resume_params='', resume_states='', save_dir='params_mobilenetv3_small', save_frequency=10, teacher=None, temperature=20, use_gn=False, use_pretrained=False, use_rec=True, use_se=False, warmup_epochs=5, warmup_lr=0.0, wd=3e-05)
Epoch[0] Batch [49]	Speed: 5851.651635 samples/sec	accuracy=0.001104	lr=0.041613
Epoch[0] Batch [99]	Speed: 11089.693156 samples/sec	accuracy=0.001543	lr=0.083227
Epoch[0] Batch [149]	Speed: 11265.380433 samples/sec	accuracy=0.002484	lr=0.124840
Epoch[0] Batch [199]	Speed: 10725.145456 samples/sec	accuracy=0.004094	lr=0.166453
Epoch[0] Batch [249]	Speed: 10628.237807 samples/sec	accuracy=0.006664	lr=0.208067
Epoch[0] Batch [299]	Speed: 10363.926428 samples/sec	accuracy=0.009657	lr=0.249680
Epoch[0] Batch [349]	Speed: 10338.577377 samples/sec	accuracy=0.013588	lr=0.291293
Epoch[0] Batch [399]	Speed: 11342.749893 samples/sec	accuracy=0.018079	lr=0.332907
Epoch[0] Batch [449]	Speed: 10660.920803 samples/sec	accuracy=0.023287	lr=0.374520
Epoch[0] Batch [499]	Speed: 11151.251498 samples/sec	accuracy=0.028911	lr=0.416133
Epoch[0] Batch [549]	Speed: 10559.614397 samples/sec	accuracy=0.035224	lr=0.457746
Epoch[0] Batch [599]	Speed: 10392.240995 samples/sec	accuracy=0.041589	lr=0.499360
[Epoch 0] training: accuracy=0.045006
[Epoch 0] speed: 10078 samples/sec	time cost: 132.151719
[Epoch 0] validation: err-top1=0.858145 err-top5=0.669941
Epoch[1] Batch [49]	Speed: 10754.361737 samples/sec	accuracy=0.135322	lr=0.562612
Epoch[1] Batch [99]	Speed: 11162.884900 samples/sec	accuracy=0.141245	lr=0.604225
Epoch[1] Batch [149]	Speed: 10305.634566 samples/sec	accuracy=0.147464	lr=0.645839
Epoch[1] Batch [199]	Speed: 11272.474115 samples/sec	accuracy=0.152512	lr=0.687452
Epoch[1] Batch [249]	Speed: 11312.785671 samples/sec	accuracy=0.157857	lr=0.729065
Epoch[1] Batch [299]	Speed: 11132.134796 samples/sec	accuracy=0.163547	lr=0.770679
Epoch[1] Batch [349]	Speed: 11423.814416 samples/sec	accuracy=0.168743	lr=0.812292
Epoch[1] Batch [399]	Speed: 11225.612259 samples/sec	accuracy=0.174080	lr=0.853905
Epoch[1] Batch [449]	Speed: 11576.384901 samples/sec	accuracy=0.179026	lr=0.895519
Epoch[1] Batch [499]	Speed: 11361.945468 samples/sec	accuracy=0.183813	lr=0.937132
Epoch[1] Batch [549]	Speed: 11399.656490 samples/sec	accuracy=0.188280	lr=0.978745
Epoch[1] Batch [599]	Speed: 11588.415252 samples/sec	accuracy=0.192828	lr=1.020359
[Epoch 1] training: accuracy=0.195199
[Epoch 1] speed: 11191 samples/sec	time cost: 118.952978
[Epoch 1] validation: err-top1=0.730632 err-top5=0.483358
Epoch[2] Batch [49]	Speed: 11102.493771 samples/sec	accuracy=0.256816	lr=1.083611
Epoch[2] Batch [99]	Speed: 11245.095801 samples/sec	accuracy=0.258999	lr=1.125224
Epoch[2] Batch [149]	Speed: 10737.086196 samples/sec	accuracy=0.260807	lr=1.166837
Epoch[2] Batch [199]	Speed: 11321.997005 samples/sec	accuracy=0.263733	lr=1.208451
Epoch[2] Batch [249]	Speed: 11195.788954 samples/sec	accuracy=0.266547	lr=1.250064
Epoch[2] Batch [299]	Speed: 11303.205564 samples/sec	accuracy=0.269100	lr=1.291677
Epoch[2] Batch [349]	Speed: 10858.409150 samples/sec	accuracy=0.271961	lr=1.333291
Epoch[2] Batch [399]	Speed: 10596.632964 samples/sec	accuracy=0.274341	lr=1.374904
Epoch[2] Batch [449]	Speed: 11141.358209 samples/sec	accuracy=0.277043	lr=1.416517
Epoch[2] Batch [499]	Speed: 11199.943716 samples/sec	accuracy=0.279478	lr=1.458131
Epoch[2] Batch [549]	Speed: 11721.166355 samples/sec	accuracy=0.281368	lr=1.499744
Epoch[2] Batch [599]	Speed: 10964.249932 samples/sec	accuracy=0.282921	lr=1.541357
[Epoch 2] training: accuracy=0.283949
[Epoch 2] speed: 11093 samples/sec	time cost: 120.043854
[Epoch 2] validation: err-top1=0.665508 err-top5=0.407910
Epoch[3] Batch [49]	Speed: 10865.240185 samples/sec	accuracy=0.320527	lr=1.603777
Epoch[3] Batch [99]	Speed: 11181.432942 samples/sec	accuracy=0.320137	lr=1.645391
Epoch[3] Batch [149]	Speed: 11281.912679 samples/sec	accuracy=0.319961	lr=1.687004
Epoch[3] Batch [199]	Speed: 11357.345905 samples/sec	accuracy=0.320212	lr=1.728617
Epoch[3] Batch [249]	Speed: 11154.331746 samples/sec	accuracy=0.321523	lr=1.770230
Epoch[3] Batch [299]	Speed: 11422.914781 samples/sec	accuracy=0.322235	lr=1.811844
Epoch[3] Batch [349]	Speed: 11098.958489 samples/sec	accuracy=0.323711	lr=1.853457
Epoch[3] Batch [399]	Speed: 11045.595036 samples/sec	accuracy=0.324956	lr=1.895070
Epoch[3] Batch [449]	Speed: 11043.879547 samples/sec	accuracy=0.326089	lr=1.936684
Epoch[3] Batch [499]	Speed: 11546.635619 samples/sec	accuracy=0.327186	lr=1.978297
Epoch[3] Batch [549]	Speed: 11637.233894 samples/sec	accuracy=0.328105	lr=2.019910
Epoch[3] Batch [599]	Speed: 9767.072526 samples/sec	accuracy=0.329068	lr=2.061524
[Epoch 3] training: accuracy=0.329362
[Epoch 3] speed: 11078 samples/sec	time cost: 120.002222
[Epoch 3] validation: err-top1=0.625102 err-top5=0.363281
Epoch[4] Batch [49]	Speed: 10684.883340 samples/sec	accuracy=0.348867	lr=2.124776
Epoch[4] Batch [99]	Speed: 11538.275211 samples/sec	accuracy=0.349028	lr=2.166389
Epoch[4] Batch [149]	Speed: 11087.930738 samples/sec	accuracy=0.348561	lr=2.208003
Epoch[4] Batch [199]	Speed: 11696.534349 samples/sec	accuracy=0.348467	lr=2.249616
Epoch[4] Batch [249]	Speed: 11328.413650 samples/sec	accuracy=0.348291	lr=2.291229
Epoch[4] Batch [299]	Speed: 11590.507402 samples/sec	accuracy=0.348223	lr=2.332843
Epoch[4] Batch [349]	Speed: 11417.295647 samples/sec	accuracy=0.348996	lr=2.374456
Epoch[4] Batch [399]	Speed: 11083.040360 samples/sec	accuracy=0.349080	lr=2.416069
Epoch[4] Batch [449]	Speed: 10684.815292 samples/sec	accuracy=0.349435	lr=2.457682
Epoch[4] Batch [499]	Speed: 10950.768504 samples/sec	accuracy=0.349494	lr=2.499296
Epoch[4] Batch [549]	Speed: 11255.945900 samples/sec	accuracy=0.349827	lr=2.540909
Epoch[4] Batch [599]	Speed: 11326.025260 samples/sec	accuracy=0.350070	lr=2.582522
[Epoch 4] training: accuracy=0.350120
[Epoch 4] speed: 11202 samples/sec	time cost: 118.846133
[Epoch 4] validation: err-top1=0.645625 err-top5=0.387500
Epoch[5] Batch [49]	Speed: 11062.992712 samples/sec	accuracy=0.360850	lr=2.600000
Epoch[5] Batch [99]	Speed: 11186.573135 samples/sec	accuracy=0.361313	lr=2.599999
Epoch[5] Batch [149]	Speed: 11296.985318 samples/sec	accuracy=0.361094	lr=2.599997
Epoch[5] Batch [199]	Speed: 10509.037672 samples/sec	accuracy=0.361372	lr=2.599995
Epoch[5] Batch [249]	Speed: 10773.521117 samples/sec	accuracy=0.361797	lr=2.599992
Epoch[5] Batch [299]	Speed: 11009.709551 samples/sec	accuracy=0.361820	lr=2.599988
Epoch[5] Batch [349]	Speed: 11249.334093 samples/sec	accuracy=0.362243	lr=2.599984
Epoch[5] Batch [399]	Speed: 11058.839541 samples/sec	accuracy=0.362761	lr=2.599979
Epoch[5] Batch [449]	Speed: 11048.745936 samples/sec	accuracy=0.363265	lr=2.599973
Epoch[5] Batch [499]	Speed: 11135.487995 samples/sec	accuracy=0.363771	lr=2.599967
Epoch[5] Batch [549]	Speed: 11190.230935 samples/sec	accuracy=0.364277	lr=2.599960
Epoch[5] Batch [599]	Speed: 10294.925608 samples/sec	accuracy=0.364159	lr=2.599953
[Epoch 5] training: accuracy=0.364189
[Epoch 5] speed: 10921 samples/sec	time cost: 121.764293
[Epoch 5] validation: err-top1=0.630005 err-top5=0.373617
Epoch[6] Batch [49]	Speed: 10394.514385 samples/sec	accuracy=0.375332	lr=2.599940
Epoch[6] Batch [99]	Speed: 10893.065439 samples/sec	accuracy=0.375479	lr=2.599931
Epoch[6] Batch [149]	Speed: 11301.982796 samples/sec	accuracy=0.375208	lr=2.599921
Epoch[6] Batch [199]	Speed: 11651.017642 samples/sec	accuracy=0.375410	lr=2.599910
Epoch[6] Batch [249]	Speed: 10856.947258 samples/sec	accuracy=0.376332	lr=2.599899
Epoch[6] Batch [299]	Speed: 11122.564278 samples/sec	accuracy=0.376670	lr=2.599888
Epoch[6] Batch [349]	Speed: 11210.124237 samples/sec	accuracy=0.376173	lr=2.599875
Epoch[6] Batch [399]	Speed: 11103.229400 samples/sec	accuracy=0.376238	lr=2.599862
Epoch[6] Batch [449]	Speed: 10885.306007 samples/sec	accuracy=0.376582	lr=2.599848
Epoch[6] Batch [499]	Speed: 11041.708684 samples/sec	accuracy=0.376980	lr=2.599834
Epoch[6] Batch [549]	Speed: 11141.157060 samples/sec	accuracy=0.377295	lr=2.599819
Epoch[6] Batch [599]	Speed: 11408.718351 samples/sec	accuracy=0.377796	lr=2.599803
[Epoch 6] training: accuracy=0.377991
[Epoch 6] speed: 11029 samples/sec	time cost: 120.500383
[Epoch 6] validation: err-top1=0.617371 err-top5=0.358846
Epoch[7] Batch [49]	Speed: 10413.351109 samples/sec	accuracy=0.392334	lr=2.599778
Epoch[7] Batch [99]	Speed: 11250.858185 samples/sec	accuracy=0.391328	lr=2.599761
Epoch[7] Batch [149]	Speed: 11242.047326 samples/sec	accuracy=0.391077	lr=2.599743
Epoch[7] Batch [199]	Speed: 11534.511524 samples/sec	accuracy=0.390684	lr=2.599725
Epoch[7] Batch [249]	Speed: 11165.750092 samples/sec	accuracy=0.390766	lr=2.599705
Epoch[7] Batch [299]	Speed: 10950.358640 samples/sec	accuracy=0.390654	lr=2.599685
Epoch[7] Batch [349]	Speed: 11558.840601 samples/sec	accuracy=0.390988	lr=2.599665
Epoch[7] Batch [399]	Speed: 11743.349764 samples/sec	accuracy=0.390677	lr=2.599644
Epoch[7] Batch [449]	Speed: 10907.641530 samples/sec	accuracy=0.390355	lr=2.599622
Epoch[7] Batch [499]	Speed: 11041.539503 samples/sec	accuracy=0.390404	lr=2.599599
Epoch[7] Batch [549]	Speed: 11154.891736 samples/sec	accuracy=0.390365	lr=2.599576
Epoch[7] Batch [599]	Speed: 10949.005025 samples/sec	accuracy=0.390438	lr=2.599552
[Epoch 7] training: accuracy=0.390678
[Epoch 7] speed: 11120 samples/sec	time cost: 119.995014
[Epoch 7] validation: err-top1=0.586289 err-top5=0.329922
Epoch[8] Batch [49]	Speed: 10246.047270 samples/sec	accuracy=0.403047	lr=2.599515
Epoch[8] Batch [99]	Speed: 11304.484532 samples/sec	accuracy=0.401318	lr=2.599489
Epoch[8] Batch [149]	Speed: 11029.406615 samples/sec	accuracy=0.400433	lr=2.599463
Epoch[8] Batch [199]	Speed: 10884.726413 samples/sec	accuracy=0.399878	lr=2.599436
Epoch[8] Batch [249]	Speed: 11181.751700 samples/sec	accuracy=0.399141	lr=2.599409
Epoch[8] Batch [299]	Speed: 10941.205301 samples/sec	accuracy=0.398555	lr=2.599381
Epoch[8] Batch [349]	Speed: 11265.922374 samples/sec	accuracy=0.398258	lr=2.599352
Epoch[8] Batch [399]	Speed: 11372.374589 samples/sec	accuracy=0.398274	lr=2.599323
Epoch[8] Batch [449]	Speed: 11516.448700 samples/sec	accuracy=0.397914	lr=2.599293
Epoch[8] Batch [499]	Speed: 11383.642887 samples/sec	accuracy=0.397986	lr=2.599262
Epoch[8] Batch [549]	Speed: 10698.281253 samples/sec	accuracy=0.398199	lr=2.599231
Epoch[8] Batch [599]	Speed: 10933.597873 samples/sec	accuracy=0.398062	lr=2.599199
[Epoch 8] training: accuracy=0.398186
[Epoch 8] speed: 11011 samples/sec	time cost: 120.783459
[Epoch 8] validation: err-top1=0.578532 err-top5=0.324036
Epoch[9] Batch [49]	Speed: 10839.876755 samples/sec	accuracy=0.412031	lr=2.599149
Epoch[9] Batch [99]	Speed: 11341.765041 samples/sec	accuracy=0.408398	lr=2.599115
Epoch[9] Batch [149]	Speed: 11471.331365 samples/sec	accuracy=0.406201	lr=2.599081
Epoch[9] Batch [199]	Speed: 11199.994242 samples/sec	accuracy=0.404995	lr=2.599046
Epoch[9] Batch [249]	Speed: 11151.116291 samples/sec	accuracy=0.404346	lr=2.599010
Epoch[9] Batch [299]	Speed: 10910.529335 samples/sec	accuracy=0.403900	lr=2.598974
Epoch[9] Batch [349]	Speed: 10832.559527 samples/sec	accuracy=0.403571	lr=2.598937
Epoch[9] Batch [399]	Speed: 10247.114559 samples/sec	accuracy=0.403162	lr=2.598900
Epoch[9] Batch [449]	Speed: 10783.559473 samples/sec	accuracy=0.403008	lr=2.598861
Epoch[9] Batch [499]	Speed: 10555.170035 samples/sec	accuracy=0.403069	lr=2.598823
Epoch[9] Batch [549]	Speed: 10574.467285 samples/sec	accuracy=0.402609	lr=2.598783
Epoch[9] Batch [599]	Speed: 10880.919892 samples/sec	accuracy=0.402249	lr=2.598743
[Epoch 9] training: accuracy=0.402246
[Epoch 9] speed: 10832 samples/sec	time cost: 122.663609
[Epoch 9] validation: err-top1=0.572148 err-top5=0.312500
Epoch[10] Batch [49]	Speed: 10368.417911 samples/sec	accuracy=0.415039	lr=2.598682
Epoch[10] Batch [99]	Speed: 10628.996891 samples/sec	accuracy=0.412690	lr=2.598640
Epoch[10] Batch [149]	Speed: 11013.815163 samples/sec	accuracy=0.411387	lr=2.598597
Epoch[10] Batch [199]	Speed: 10930.688387 samples/sec	accuracy=0.411052	lr=2.598554
Epoch[10] Batch [249]	Speed: 10671.051768 samples/sec	accuracy=0.410104	lr=2.598511
Epoch[10] Batch [299]	Speed: 10256.687551 samples/sec	accuracy=0.409852	lr=2.598466
Epoch[10] Batch [349]	Speed: 10850.791103 samples/sec	accuracy=0.409372	lr=2.598421
Epoch[10] Batch [399]	Speed: 11346.882863 samples/sec	accuracy=0.409099	lr=2.598376
Epoch[10] Batch [449]	Speed: 10835.975501 samples/sec	accuracy=0.408031	lr=2.598329
Epoch[10] Batch [499]	Speed: 11048.280107 samples/sec	accuracy=0.407144	lr=2.598282
Epoch[10] Batch [549]	Speed: 11105.961808 samples/sec	accuracy=0.406681	lr=2.598235
Epoch[10] Batch [599]	Speed: 10959.720274 samples/sec	accuracy=0.406293	lr=2.598186
[Epoch 10] training: accuracy=0.406323
[Epoch 10] speed: 10817 samples/sec	time cost: 122.822823
[Epoch 10] validation: err-top1=0.573079 err-top5=0.322184
Epoch[11] Batch [49]	Speed: 10863.247507 samples/sec	accuracy=0.413457	lr=2.598112
Epoch[11] Batch [99]	Speed: 10870.907274 samples/sec	accuracy=0.411899	lr=2.598062
Epoch[11] Batch [149]	Speed: 11008.181809 samples/sec	accuracy=0.410361	lr=2.598011
Epoch[11] Batch [199]	Speed: 10775.363143 samples/sec	accuracy=0.409785	lr=2.597960
Epoch[11] Batch [249]	Speed: 11155.901774 samples/sec	accuracy=0.409189	lr=2.597908
Epoch[11] Batch [299]	Speed: 11119.682628 samples/sec	accuracy=0.408939	lr=2.597856
Epoch[11] Batch [349]	Speed: 11109.126847 samples/sec	accuracy=0.408919	lr=2.597803
Epoch[11] Batch [399]	Speed: 10969.692987 samples/sec	accuracy=0.408827	lr=2.597749
Epoch[11] Batch [449]	Speed: 10427.339659 samples/sec	accuracy=0.408784	lr=2.597694
Epoch[11] Batch [499]	Speed: 10787.853613 samples/sec	accuracy=0.408509	lr=2.597639
Epoch[11] Batch [549]	Speed: 11250.217792 samples/sec	accuracy=0.408564	lr=2.597583
Epoch[11] Batch [599]	Speed: 10704.850605 samples/sec	accuracy=0.408598	lr=2.597527
[Epoch 11] training: accuracy=0.408666
[Epoch 11] speed: 10905 samples/sec	time cost: 121.708831
[Epoch 11] validation: err-top1=0.614583 err-top5=0.356445
Epoch[12] Batch [49]	Speed: 10607.524069 samples/sec	accuracy=0.416201	lr=2.597441
Epoch[12] Batch [99]	Speed: 11317.368290 samples/sec	accuracy=0.415171	lr=2.597383
Epoch[12] Batch [149]	Speed: 11390.252559 samples/sec	accuracy=0.413796	lr=2.597324
Epoch[12] Batch [199]	Speed: 11021.889629 samples/sec	accuracy=0.413279	lr=2.597265
Epoch[12] Batch [249]	Speed: 11230.473949 samples/sec	accuracy=0.412906	lr=2.597205
Epoch[12] Batch [299]	Speed: 11413.526139 samples/sec	accuracy=0.412744	lr=2.597144
Epoch[12] Batch [349]	Speed: 10370.548423 samples/sec	accuracy=0.412981	lr=2.597083
Epoch[12] Batch [399]	Speed: 11275.224756 samples/sec	accuracy=0.412600	lr=2.597021
Epoch[12] Batch [449]	Speed: 10869.350690 samples/sec	accuracy=0.412123	lr=2.596959
Epoch[12] Batch [499]	Speed: 11242.886617 samples/sec	accuracy=0.411872	lr=2.596895
Epoch[12] Batch [549]	Speed: 11338.226317 samples/sec	accuracy=0.411760	lr=2.596831
Epoch[12] Batch [599]	Speed: 11309.609855 samples/sec	accuracy=0.411819	lr=2.596767
[Epoch 12] training: accuracy=0.411909
[Epoch 12] speed: 11076 samples/sec	time cost: 120.346170
[Epoch 12] validation: err-top1=0.578262 err-top5=0.320293
Epoch[13] Batch [49]	Speed: 10998.370528 samples/sec	accuracy=0.418818	lr=2.596667
Epoch[13] Batch [99]	Speed: 10806.743228 samples/sec	accuracy=0.417656	lr=2.596601
Epoch[13] Batch [149]	Speed: 11182.657714 samples/sec	accuracy=0.417454	lr=2.596534
Epoch[13] Batch [199]	Speed: 11306.457848 samples/sec	accuracy=0.417017	lr=2.596467
Epoch[13] Batch [249]	Speed: 10856.300155 samples/sec	accuracy=0.417369	lr=2.596399
Epoch[13] Batch [299]	Speed: 11475.867046 samples/sec	accuracy=0.417259	lr=2.596330
Epoch[13] Batch [349]	Speed: 11522.141153 samples/sec	accuracy=0.416860	lr=2.596261
Epoch[13] Batch [399]	Speed: 11198.967737 samples/sec	accuracy=0.416357	lr=2.596191
Epoch[13] Batch [449]	Speed: 11141.374972 samples/sec	accuracy=0.415477	lr=2.596120
Epoch[13] Batch [499]	Speed: 11231.885429 samples/sec	accuracy=0.415134	lr=2.596048
Epoch[13] Batch [549]	Speed: 11764.786775 samples/sec	accuracy=0.415188	lr=2.595976
Epoch[13] Batch [599]	Speed: 11387.540923 samples/sec	accuracy=0.415145	lr=2.595904
[Epoch 13] training: accuracy=0.415132
[Epoch 13] speed: 11213 samples/sec	time cost: 118.486706
[Epoch 13] validation: err-top1=0.540955 err-top5=0.283569
Epoch[14] Batch [49]	Speed: 10117.232400 samples/sec	accuracy=0.415010	lr=2.595793
Epoch[14] Batch [99]	Speed: 10362.854671 samples/sec	accuracy=0.416650	lr=2.595719
Epoch[14] Batch [149]	Speed: 11690.335797 samples/sec	accuracy=0.416605	lr=2.595644
Epoch[14] Batch [199]	Speed: 11474.681440 samples/sec	accuracy=0.416689	lr=2.595569
Epoch[14] Batch [249]	Speed: 11385.748974 samples/sec	accuracy=0.416650	lr=2.595492
Epoch[14] Batch [299]	Speed: 10730.775085 samples/sec	accuracy=0.416453	lr=2.595415
Epoch[14] Batch [349]	Speed: 10980.599819 samples/sec	accuracy=0.416522	lr=2.595338
Epoch[14] Batch [399]	Speed: 11169.586578 samples/sec	accuracy=0.416857	lr=2.595260
Epoch[14] Batch [449]	Speed: 10906.852651 samples/sec	accuracy=0.416909	lr=2.595181
Epoch[14] Batch [499]	Speed: 11618.534493 samples/sec	accuracy=0.416843	lr=2.595101
Epoch[14] Batch [549]	Speed: 10610.310435 samples/sec	accuracy=0.416892	lr=2.595021
Epoch[14] Batch [599]	Speed: 10462.213379 samples/sec	accuracy=0.416878	lr=2.594940
[Epoch 14] training: accuracy=0.416892
[Epoch 14] speed: 10909 samples/sec	time cost: 122.214237
[Epoch 14] validation: err-top1=0.612520 err-top5=0.355332
Epoch[15] Batch [49]	Speed: 10990.997309 samples/sec	accuracy=0.426953	lr=2.594816
Epoch[15] Batch [99]	Speed: 11328.000128 samples/sec	accuracy=0.424224	lr=2.594734
Epoch[15] Batch [149]	Speed: 11211.259312 samples/sec	accuracy=0.422793	lr=2.594651
Epoch[15] Batch [199]	Speed: 11043.667704 samples/sec	accuracy=0.421794	lr=2.594567
Epoch[15] Batch [249]	Speed: 11388.889482 samples/sec	accuracy=0.421406	lr=2.594483
Epoch[15] Batch [299]	Speed: 11230.763206 samples/sec	accuracy=0.420947	lr=2.594398
Epoch[15] Batch [349]	Speed: 10901.410336 samples/sec	accuracy=0.420628	lr=2.594312
Epoch[15] Batch [399]	Speed: 11516.376133 samples/sec	accuracy=0.420477	lr=2.594226
Epoch[15] Batch [449]	Speed: 10941.569044 samples/sec	accuracy=0.420552	lr=2.594139
Epoch[15] Batch [499]	Speed: 11160.575651 samples/sec	accuracy=0.420516	lr=2.594051
Epoch[15] Batch [549]	Speed: 10593.146993 samples/sec	accuracy=0.420100	lr=2.593963
Epoch[15] Batch [599]	Speed: 11422.296876 samples/sec	accuracy=0.420240	lr=2.593874
[Epoch 15] training: accuracy=0.420134
[Epoch 15] speed: 11120 samples/sec	time cost: 119.712960
[Epoch 15] validation: err-top1=0.558167 err-top5=0.304484
Epoch[16] Batch [49]	Speed: 10921.181313 samples/sec	accuracy=0.428711	lr=2.593738
Epoch[16] Batch [99]	Speed: 10706.022294 samples/sec	accuracy=0.427324	lr=2.593647
Epoch[16] Batch [149]	Speed: 11205.837955 samples/sec	accuracy=0.426048	lr=2.593556
Epoch[16] Batch [199]	Speed: 10941.206694 samples/sec	accuracy=0.425315	lr=2.593464
Epoch[16] Batch [249]	Speed: 11208.769411 samples/sec	accuracy=0.424719	lr=2.593371
Epoch[16] Batch [299]	Speed: 10857.834612 samples/sec	accuracy=0.424020	lr=2.593278
Epoch[16] Batch [349]	Speed: 11026.743731 samples/sec	accuracy=0.423687	lr=2.593185
Epoch[16] Batch [399]	Speed: 11392.826470 samples/sec	accuracy=0.423590	lr=2.593090
Epoch[16] Batch [449]	Speed: 10405.138971 samples/sec	accuracy=0.423524	lr=2.592995
Epoch[16] Batch [499]	Speed: 11335.041888 samples/sec	accuracy=0.423292	lr=2.592899
Epoch[16] Batch [549]	Speed: 11560.469010 samples/sec	accuracy=0.423424	lr=2.592803
Epoch[16] Batch [599]	Speed: 11269.355189 samples/sec	accuracy=0.423395	lr=2.592706
[Epoch 16] training: accuracy=0.423293
[Epoch 16] speed: 11020 samples/sec	time cost: 120.828138
[Epoch 16] validation: err-top1=0.598008 err-top5=0.337285
Epoch[17] Batch [49]	Speed: 11155.759500 samples/sec	accuracy=0.432256	lr=2.592559
Epoch[17] Batch [99]	Speed: 11326.505247 samples/sec	accuracy=0.429365	lr=2.592460
Epoch[17] Batch [149]	Speed: 11264.234373 samples/sec	accuracy=0.427666	lr=2.592361
Epoch[17] Batch [199]	Speed: 10324.444586 samples/sec	accuracy=0.427812	lr=2.592261
Epoch[17] Batch [249]	Speed: 10859.475487 samples/sec	accuracy=0.426912	lr=2.592161
Epoch[17] Batch [299]	Speed: 10911.201492 samples/sec	accuracy=0.426924	lr=2.592059
Epoch[17] Batch [349]	Speed: 10616.336533 samples/sec	accuracy=0.426353	lr=2.591957
Epoch[17] Batch [399]	Speed: 11268.043360 samples/sec	accuracy=0.425535	lr=2.591855
Epoch[17] Batch [449]	Speed: 11174.521668 samples/sec	accuracy=0.424488	lr=2.591752
Epoch[17] Batch [499]	Speed: 10843.450104 samples/sec	accuracy=0.424036	lr=2.591648
Epoch[17] Batch [549]	Speed: 10144.947953 samples/sec	accuracy=0.423555	lr=2.591543
Epoch[17] Batch [599]	Speed: 10681.437095 samples/sec	accuracy=0.423464	lr=2.591438
[Epoch 17] training: accuracy=0.423241
[Epoch 17] speed: 10858 samples/sec	time cost: 122.561087
[Epoch 17] validation: err-top1=0.548096 err-top5=0.291687
Epoch[18] Batch [49]	Speed: 11064.170011 samples/sec	accuracy=0.429297	lr=2.591277
Epoch[18] Batch [99]	Speed: 10839.805350 samples/sec	accuracy=0.426851	lr=2.591170
Epoch[18] Batch [149]	Speed: 10901.283887 samples/sec	accuracy=0.425573	lr=2.591063
Epoch[18] Batch [199]	Speed: 11174.881610 samples/sec	accuracy=0.425757	lr=2.590955
Epoch[18] Batch [249]	Speed: 11276.934411 samples/sec	accuracy=0.425475	lr=2.590846
Epoch[18] Batch [299]	Speed: 10582.205331 samples/sec	accuracy=0.424977	lr=2.590737
Epoch[18] Batch [349]	Speed: 11102.855115 samples/sec	accuracy=0.423977	lr=2.590627
Epoch[18] Batch [399]	Speed: 11437.608929 samples/sec	accuracy=0.423734	lr=2.590516
Epoch[18] Batch [449]	Speed: 10722.972519 samples/sec	accuracy=0.423366	lr=2.590405
Epoch[18] Batch [499]	Speed: 11016.622134 samples/sec	accuracy=0.423494	lr=2.590293
Epoch[18] Batch [549]	Speed: 11349.114816 samples/sec	accuracy=0.423182	lr=2.590180
Epoch[18] Batch [599]	Speed: 11496.783676 samples/sec	accuracy=0.423101	lr=2.590067
[Epoch 18] training: accuracy=0.423141
[Epoch 18] speed: 11080 samples/sec	time cost: 120.119272
[Epoch 18] validation: err-top1=0.561239 err-top5=0.306966
Epoch[19] Batch [49]	Speed: 11154.544379 samples/sec	accuracy=0.431240	lr=2.589896
Epoch[19] Batch [99]	Speed: 11405.157720 samples/sec	accuracy=0.430854	lr=2.589781
Epoch[19] Batch [149]	Speed: 11386.401267 samples/sec	accuracy=0.429144	lr=2.589666
Epoch[19] Batch [199]	Speed: 11073.543186 samples/sec	accuracy=0.428596	lr=2.589550
Epoch[19] Batch [249]	Speed: 10329.707276 samples/sec	accuracy=0.427906	lr=2.589433
Epoch[19] Batch [299]	Speed: 10644.428776 samples/sec	accuracy=0.427552	lr=2.589315
Epoch[19] Batch [349]	Speed: 10855.938766 samples/sec	accuracy=0.426949	lr=2.589197
Epoch[19] Batch [399]	Speed: 11039.515975 samples/sec	accuracy=0.425493	lr=2.589079
Epoch[19] Batch [449]	Speed: 10626.817248 samples/sec	accuracy=0.425267	lr=2.588959
Epoch[19] Batch [499]	Speed: 10731.195756 samples/sec	accuracy=0.425099	lr=2.588839
Epoch[19] Batch [549]	Speed: 11175.800178 samples/sec	accuracy=0.424877	lr=2.588719
Epoch[19] Batch [599]	Speed: 11172.107922 samples/sec	accuracy=0.424729	lr=2.588597
[Epoch 19] training: accuracy=0.424684
[Epoch 19] speed: 10871 samples/sec	time cost: 122.579328
[Epoch 19] validation: err-top1=0.530215 err-top5=0.274766
Epoch[20] Batch [49]	Speed: 10528.377567 samples/sec	accuracy=0.415283	lr=2.588412
Epoch[20] Batch [99]	Speed: 11196.456147 samples/sec	accuracy=0.420405	lr=2.588289
Epoch[20] Batch [149]	Speed: 11124.443469 samples/sec	accuracy=0.422572	lr=2.588165
Epoch[20] Batch [199]	Speed: 10695.988132 samples/sec	accuracy=0.423210	lr=2.588041
Epoch[20] Batch [249]	Speed: 11211.879472 samples/sec	accuracy=0.423961	lr=2.587916
Epoch[20] Batch [299]	Speed: 10922.005040 samples/sec	accuracy=0.424128	lr=2.587790
Epoch[20] Batch [349]	Speed: 11169.303950 samples/sec	accuracy=0.424242	lr=2.587664
Epoch[20] Batch [399]	Speed: 10756.657328 samples/sec	accuracy=0.424341	lr=2.587538
Epoch[20] Batch [449]	Speed: 11148.736954 samples/sec	accuracy=0.424484	lr=2.587410
Epoch[20] Batch [499]	Speed: 11292.423641 samples/sec	accuracy=0.424360	lr=2.587282
Epoch[20] Batch [549]	Speed: 10995.288543 samples/sec	accuracy=0.424291	lr=2.587153
Epoch[20] Batch [599]	Speed: 10658.268870 samples/sec	accuracy=0.423958	lr=2.587024
[Epoch 20] training: accuracy=0.423650
[Epoch 20] speed: 10939 samples/sec	time cost: 121.449939
[Epoch 20] validation: err-top1=0.550720 err-top5=0.291626
Epoch[21] Batch [49]	Speed: 11020.372362 samples/sec	accuracy=0.431494	lr=2.586829
Epoch[21] Batch [99]	Speed: 11204.449967 samples/sec	accuracy=0.431050	lr=2.586698
Epoch[21] Batch [149]	Speed: 11087.420383 samples/sec	accuracy=0.430267	lr=2.586566
Epoch[21] Batch [199]	Speed: 11237.872447 samples/sec	accuracy=0.429954	lr=2.586434
Epoch[21] Batch [249]	Speed: 11295.395233 samples/sec	accuracy=0.429934	lr=2.586301
Epoch[21] Batch [299]	Speed: 11233.591955 samples/sec	accuracy=0.429652	lr=2.586167
Epoch[21] Batch [349]	Speed: 11093.611359 samples/sec	accuracy=0.429621	lr=2.586033
Epoch[21] Batch [399]	Speed: 11324.502534 samples/sec	accuracy=0.429248	lr=2.585898
Epoch[21] Batch [449]	Speed: 11369.692827 samples/sec	accuracy=0.429027	lr=2.585762
Epoch[21] Batch [499]	Speed: 11267.216270 samples/sec	accuracy=0.428626	lr=2.585626
Epoch[21] Batch [549]	Speed: 11128.982012 samples/sec	accuracy=0.428503	lr=2.585489
Epoch[21] Batch [599]	Speed: 11187.671972 samples/sec	accuracy=0.428331	lr=2.585352
[Epoch 21] training: accuracy=0.428308
[Epoch 21] speed: 11174 samples/sec	time cost: 119.316677
[Epoch 21] validation: err-top1=0.566465 err-top5=0.304219
Epoch[22] Batch [49]	Speed: 10594.047669 samples/sec	accuracy=0.436943	lr=2.585142
Epoch[22] Batch [99]	Speed: 11097.715850 samples/sec	accuracy=0.435449	lr=2.585003
Epoch[22] Batch [149]	Speed: 11172.627266 samples/sec	accuracy=0.433812	lr=2.584863
Epoch[22] Batch [199]	Speed: 11066.962217 samples/sec	accuracy=0.433159	lr=2.584723
Epoch[22] Batch [249]	Speed: 11334.995520 samples/sec	accuracy=0.432656	lr=2.584582
Epoch[22] Batch [299]	Speed: 10965.818978 samples/sec	accuracy=0.432293	lr=2.584440
Epoch[22] Batch [349]	Speed: 11490.455896 samples/sec	accuracy=0.431882	lr=2.584298
Epoch[22] Batch [399]	Speed: 11384.432541 samples/sec	accuracy=0.431536	lr=2.584155
Epoch[22] Batch [449]	Speed: 11016.289268 samples/sec	accuracy=0.430641	lr=2.584011
Epoch[22] Batch [499]	Speed: 10799.898080 samples/sec	accuracy=0.430311	lr=2.583867
Epoch[22] Batch [549]	Speed: 11214.017048 samples/sec	accuracy=0.429951	lr=2.583722
Epoch[22] Batch [599]	Speed: 10932.815811 samples/sec	accuracy=0.429945	lr=2.583577
[Epoch 22] training: accuracy=0.429895
[Epoch 22] speed: 11072 samples/sec	time cost: 120.108131
[Epoch 22] validation: err-top1=0.541809 err-top5=0.284261
Epoch[23] Batch [49]	Speed: 10603.912595 samples/sec	accuracy=0.441152	lr=2.583354
Epoch[23] Batch [99]	Speed: 11250.970770 samples/sec	accuracy=0.438066	lr=2.583207
Epoch[23] Batch [149]	Speed: 10202.813756 samples/sec	accuracy=0.436973	lr=2.583059
Epoch[23] Batch [199]	Speed: 10920.767830 samples/sec	accuracy=0.435972	lr=2.582911
Epoch[23] Batch [249]	Speed: 11215.850533 samples/sec	accuracy=0.435293	lr=2.582762
Epoch[23] Batch [299]	Speed: 10248.122649 samples/sec	accuracy=0.434723	lr=2.582612
Epoch[23] Batch [349]	Speed: 11343.882324 samples/sec	accuracy=0.433684	lr=2.582462
Epoch[23] Batch [399]	Speed: 11396.972734 samples/sec	accuracy=0.432970	lr=2.582311
Epoch[23] Batch [449]	Speed: 10908.583737 samples/sec	accuracy=0.432565	lr=2.582159
Epoch[23] Batch [499]	Speed: 10070.726101 samples/sec	accuracy=0.432229	lr=2.582007
Epoch[23] Batch [549]	Speed: 10861.354166 samples/sec	accuracy=0.432096	lr=2.581854
Epoch[23] Batch [599]	Speed: 10238.849250 samples/sec	accuracy=0.431826	lr=2.581700
[Epoch 23] training: accuracy=0.431804
[Epoch 23] speed: 10759 samples/sec	time cost: 123.488572
[Epoch 23] validation: err-top1=0.555827 err-top5=0.298787
Epoch[24] Batch [49]	Speed: 10812.321135 samples/sec	accuracy=0.437910	lr=2.581469
Epoch[24] Batch [99]	Speed: 10819.498744 samples/sec	accuracy=0.437393	lr=2.581314
Epoch[24] Batch [149]	Speed: 10651.471780 samples/sec	accuracy=0.435635	lr=2.581158
Epoch[24] Batch [199]	Speed: 10018.828300 samples/sec	accuracy=0.434983	lr=2.581001
Epoch[24] Batch [249]	Speed: 11429.385224 samples/sec	accuracy=0.434047	lr=2.580844
Epoch[24] Batch [299]	Speed: 11366.647425 samples/sec	accuracy=0.434028	lr=2.580687
Epoch[24] Batch [349]	Speed: 11155.814265 samples/sec	accuracy=0.433903	lr=2.580528
Epoch[24] Batch [399]	Speed: 11034.450474 samples/sec	accuracy=0.433851	lr=2.580369
Epoch[24] Batch [449]	Speed: 10711.087442 samples/sec	accuracy=0.433147	lr=2.580210
Epoch[24] Batch [499]	Speed: 10604.025432 samples/sec	accuracy=0.432925	lr=2.580049
Epoch[24] Batch [549]	Speed: 11717.183013 samples/sec	accuracy=0.432801	lr=2.579888
Epoch[24] Batch [599]	Speed: 9952.031990 samples/sec	accuracy=0.432699	lr=2.579727
[Epoch 24] training: accuracy=0.432648
[Epoch 24] speed: 10808 samples/sec	time cost: 123.301472
[Epoch 24] validation: err-top1=0.524043 err-top5=0.262832
Epoch[25] Batch [49]	Speed: 11006.683827 samples/sec	accuracy=0.441846	lr=2.579480
Epoch[25] Batch [99]	Speed: 11319.018262 samples/sec	accuracy=0.439399	lr=2.579317
Epoch[25] Batch [149]	Speed: 10460.210374 samples/sec	accuracy=0.438844	lr=2.579153
Epoch[25] Batch [199]	Speed: 10759.246600 samples/sec	accuracy=0.436792	lr=2.578988
Epoch[25] Batch [249]	Speed: 10934.084701 samples/sec	accuracy=0.434646	lr=2.578823
Epoch[25] Batch [299]	Speed: 11329.949085 samples/sec	accuracy=0.433810	lr=2.578657
Epoch[25] Batch [349]	Speed: 11042.670789 samples/sec	accuracy=0.433391	lr=2.578491
Epoch[25] Batch [399]	Speed: 11060.951628 samples/sec	accuracy=0.433214	lr=2.578324
Epoch[25] Batch [449]	Speed: 11064.647158 samples/sec	accuracy=0.432983	lr=2.578156
Epoch[25] Batch [499]	Speed: 11286.358205 samples/sec	accuracy=0.432796	lr=2.577988
Epoch[25] Batch [549]	Speed: 10417.488066 samples/sec	accuracy=0.432613	lr=2.577819
Epoch[25] Batch [599]	Speed: 10327.337235 samples/sec	accuracy=0.432433	lr=2.577649
[Epoch 25] training: accuracy=0.432334
[Epoch 25] speed: 10903 samples/sec	time cost: 121.793328
[Epoch 25] validation: err-top1=0.555135 err-top5=0.297953
Epoch[26] Batch [49]	Speed: 10448.905923 samples/sec	accuracy=0.440977	lr=2.577394
Epoch[26] Batch [99]	Speed: 9868.843555 samples/sec	accuracy=0.439854	lr=2.577223
Epoch[26] Batch [149]	Speed: 11239.612853 samples/sec	accuracy=0.438874	lr=2.577051
Epoch[26] Batch [199]	Speed: 10936.006828 samples/sec	accuracy=0.438135	lr=2.576878
Epoch[26] Batch [249]	Speed: 10653.073067 samples/sec	accuracy=0.438174	lr=2.576705
Epoch[26] Batch [299]	Speed: 11105.095744 samples/sec	accuracy=0.437747	lr=2.576531
Epoch[26] Batch [349]	Speed: 11216.980031 samples/sec	accuracy=0.437208	lr=2.576357
Epoch[26] Batch [399]	Speed: 10392.430846 samples/sec	accuracy=0.436659	lr=2.576182
Epoch[26] Batch [449]	Speed: 10929.386354 samples/sec	accuracy=0.436540	lr=2.576006
Epoch[26] Batch [499]	Speed: 11127.858346 samples/sec	accuracy=0.436185	lr=2.575830
Epoch[26] Batch [549]	Speed: 11156.636093 samples/sec	accuracy=0.435566	lr=2.575653
Epoch[26] Batch [599]	Speed: 11192.149977 samples/sec	accuracy=0.435306	lr=2.575475
[Epoch 26] training: accuracy=0.435232
[Epoch 26] speed: 10815 samples/sec	time cost: 123.217197
[Epoch 26] validation: err-top1=0.520762 err-top5=0.263340
Epoch[27] Batch [49]	Speed: 11283.285245 samples/sec	accuracy=0.442031	lr=2.575204
Epoch[27] Batch [99]	Speed: 11048.148806 samples/sec	accuracy=0.433706	lr=2.575025
Epoch[27] Batch [149]	Speed: 10779.675917 samples/sec	accuracy=0.434912	lr=2.574845
Epoch[27] Batch [199]	Speed: 11098.754852 samples/sec	accuracy=0.435642	lr=2.574665
Epoch[27] Batch [249]	Speed: 10983.684248 samples/sec	accuracy=0.435695	lr=2.574483
Epoch[27] Batch [299]	Speed: 11616.495679 samples/sec	accuracy=0.435988	lr=2.574302
Epoch[27] Batch [349]	Speed: 11383.361692 samples/sec	accuracy=0.435940	lr=2.574119
Epoch[27] Batch [399]	Speed: 10910.101693 samples/sec	accuracy=0.435691	lr=2.573936
Epoch[27] Batch [449]	Speed: 10968.509657 samples/sec	accuracy=0.435527	lr=2.573752
Epoch[27] Batch [499]	Speed: 11265.045956 samples/sec	accuracy=0.435495	lr=2.573568
Epoch[27] Batch [549]	Speed: 11224.926625 samples/sec	accuracy=0.435589	lr=2.573383
Epoch[27] Batch [599]	Speed: 11063.444679 samples/sec	accuracy=0.435378	lr=2.573198
[Epoch 27] training: accuracy=0.435547
[Epoch 27] speed: 11109 samples/sec	time cost: 119.542582
[Epoch 27] validation: err-top1=0.583089 err-top5=0.325684
Epoch[28] Batch [49]	Speed: 11204.125236 samples/sec	accuracy=0.443516	lr=2.572918
Epoch[28] Batch [99]	Speed: 11468.205862 samples/sec	accuracy=0.442227	lr=2.572731
Epoch[28] Batch [149]	Speed: 11134.489155 samples/sec	accuracy=0.439971	lr=2.572543
Epoch[28] Batch [199]	Speed: 10429.673007 samples/sec	accuracy=0.439111	lr=2.572354
Epoch[28] Batch [249]	Speed: 11086.202075 samples/sec	accuracy=0.438123	lr=2.572165
Epoch[28] Batch [299]	Speed: 11599.518213 samples/sec	accuracy=0.437246	lr=2.571976
Epoch[28] Batch [349]	Speed: 10557.191930 samples/sec	accuracy=0.436434	lr=2.571785
Epoch[28] Batch [399]	Speed: 11415.461554 samples/sec	accuracy=0.435859	lr=2.571594
Epoch[28] Batch [449]	Speed: 11088.179206 samples/sec	accuracy=0.435303	lr=2.571403
Epoch[28] Batch [499]	Speed: 11303.483705 samples/sec	accuracy=0.434846	lr=2.571210
Epoch[28] Batch [549]	Speed: 11638.822334 samples/sec	accuracy=0.434900	lr=2.571017
Epoch[28] Batch [599]	Speed: 11072.506329 samples/sec	accuracy=0.434374	lr=2.570824
[Epoch 28] training: accuracy=0.434272
[Epoch 28] speed: 11151 samples/sec	time cost: 119.610992
[Epoch 28] validation: err-top1=0.740684 err-top5=0.513418
Epoch[29] Batch [49]	Speed: 10134.216021 samples/sec	accuracy=0.436465	lr=2.570528
Epoch[29] Batch [99]	Speed: 11382.965870 samples/sec	accuracy=0.438022	lr=2.570333
Epoch[29] Batch [149]	Speed: 11451.368098 samples/sec	accuracy=0.437171	lr=2.570137
Epoch[29] Batch [199]	Speed: 11007.269430 samples/sec	accuracy=0.437380	lr=2.569941
Epoch[29] Batch [249]	Speed: 11708.367389 samples/sec	accuracy=0.436756	lr=2.569744
Epoch[29] Batch [299]	Speed: 11075.658609 samples/sec	accuracy=0.436841	lr=2.569546
Epoch[29] Batch [349]	Speed: 10536.325617 samples/sec	accuracy=0.436380	lr=2.569348
Epoch[29] Batch [399]	Speed: 10997.084422 samples/sec	accuracy=0.436260	lr=2.569149
Epoch[29] Batch [449]	Speed: 11050.398682 samples/sec	accuracy=0.435928	lr=2.568949
Epoch[29] Batch [499]	Speed: 10822.823303 samples/sec	accuracy=0.435512	lr=2.568749
Epoch[29] Batch [549]	Speed: 11430.543843 samples/sec	accuracy=0.435134	lr=2.568548
Epoch[29] Batch [599]	Speed: 11406.373532 samples/sec	accuracy=0.435030	lr=2.568346
[Epoch 29] training: accuracy=0.434933
[Epoch 29] speed: 11039 samples/sec	time cost: 120.617333
[Epoch 29] validation: err-top1=0.600159 err-top5=0.339498
Epoch[30] Batch [49]	Speed: 10495.365976 samples/sec	accuracy=0.442900	lr=2.568039
Epoch[30] Batch [99]	Speed: 11102.777908 samples/sec	accuracy=0.441738	lr=2.567836
Epoch[30] Batch [149]	Speed: 11392.471691 samples/sec	accuracy=0.440811	lr=2.567632
Epoch[30] Batch [199]	Speed: 11258.836933 samples/sec	accuracy=0.440488	lr=2.567427
Epoch[30] Batch [249]	Speed: 11322.696638 samples/sec	accuracy=0.440041	lr=2.567222
Epoch[30] Batch [299]	Speed: 11200.609067 samples/sec	accuracy=0.439762	lr=2.567017
Epoch[30] Batch [349]	Speed: 11390.668221 samples/sec	accuracy=0.439212	lr=2.566810
Epoch[30] Batch [399]	Speed: 11551.625930 samples/sec	accuracy=0.438452	lr=2.566603
Epoch[30] Batch [449]	Speed: 11133.424116 samples/sec	accuracy=0.438641	lr=2.566396
Epoch[30] Batch [499]	Speed: 11060.511828 samples/sec	accuracy=0.438326	lr=2.566188
Epoch[30] Batch [549]	Speed: 11031.738968 samples/sec	accuracy=0.438025	lr=2.565979
Epoch[30] Batch [599]	Speed: 11408.320158 samples/sec	accuracy=0.437891	lr=2.565769
[Epoch 30] training: accuracy=0.437797
[Epoch 30] speed: 11163 samples/sec	time cost: 119.175504
[Epoch 30] validation: err-top1=0.527018 err-top5=0.271016
Epoch[31] Batch [49]	Speed: 11096.218630 samples/sec	accuracy=0.446045	lr=2.565454
Epoch[31] Batch [99]	Speed: 11025.593629 samples/sec	accuracy=0.442329	lr=2.565243
Epoch[31] Batch [149]	Speed: 11294.484820 samples/sec	accuracy=0.441943	lr=2.565031
Epoch[31] Batch [199]	Speed: 11553.769774 samples/sec	accuracy=0.441426	lr=2.564819
Epoch[31] Batch [249]	Speed: 11178.550969 samples/sec	accuracy=0.440516	lr=2.564606
Epoch[31] Batch [299]	Speed: 11362.082830 samples/sec	accuracy=0.439715	lr=2.564392
Epoch[31] Batch [349]	Speed: 11577.002738 samples/sec	accuracy=0.439456	lr=2.564178
Epoch[31] Batch [399]	Speed: 10889.560095 samples/sec	accuracy=0.439420	lr=2.563963
Epoch[31] Batch [449]	Speed: 10917.616225 samples/sec	accuracy=0.439551	lr=2.563747
Epoch[31] Batch [499]	Speed: 10485.902594 samples/sec	accuracy=0.439305	lr=2.563531
Epoch[31] Batch [549]	Speed: 11200.261194 samples/sec	accuracy=0.439137	lr=2.563314
Epoch[31] Batch [599]	Speed: 11199.148785 samples/sec	accuracy=0.439060	lr=2.563097
[Epoch 31] training: accuracy=0.438650
[Epoch 31] speed: 11110 samples/sec	time cost: 120.040187
[Epoch 31] validation: err-top1=0.550898 err-top5=0.294492
Epoch[32] Batch [49]	Speed: 11059.497915 samples/sec	accuracy=0.444453	lr=2.562765
Epoch[32] Batch [99]	Speed: 10876.991248 samples/sec	accuracy=0.444150	lr=2.562546
Epoch[32] Batch [149]	Speed: 11469.020154 samples/sec	accuracy=0.443499	lr=2.562327
Epoch[32] Batch [199]	Speed: 11565.759995 samples/sec	accuracy=0.442410	lr=2.562106
Epoch[32] Batch [249]	Speed: 11285.784937 samples/sec	accuracy=0.441156	lr=2.561885
Epoch[32] Batch [299]	Speed: 11300.677038 samples/sec	accuracy=0.441025	lr=2.561664
Epoch[32] Batch [349]	Speed: 10765.508682 samples/sec	accuracy=0.440386	lr=2.561442
Epoch[32] Batch [399]	Speed: 10570.108745 samples/sec	accuracy=0.440049	lr=2.561219
Epoch[32] Batch [449]	Speed: 11295.948682 samples/sec	accuracy=0.440047	lr=2.560995
Epoch[32] Batch [499]	Speed: 11083.951042 samples/sec	accuracy=0.439960	lr=2.560771
Epoch[32] Batch [549]	Speed: 11489.290324 samples/sec	accuracy=0.439617	lr=2.560547
Epoch[32] Batch [599]	Speed: 10255.860710 samples/sec	accuracy=0.439329	lr=2.560321
[Epoch 32] training: accuracy=0.439158
[Epoch 32] speed: 11047 samples/sec	time cost: 120.307713
[Epoch 32] validation: err-top1=0.572550 err-top5=0.317220
Epoch[33] Batch [49]	Speed: 11229.904876 samples/sec	accuracy=0.447236	lr=2.559982
Epoch[33] Batch [99]	Speed: 11147.426147 samples/sec	accuracy=0.444707	lr=2.559755
Epoch[33] Batch [149]	Speed: 11187.715977 samples/sec	accuracy=0.443910	lr=2.559528
Epoch[33] Batch [199]	Speed: 10908.738339 samples/sec	accuracy=0.443774	lr=2.559300
Epoch[33] Batch [249]	Speed: 11338.060498 samples/sec	accuracy=0.443193	lr=2.559071
Epoch[33] Batch [299]	Speed: 10660.588974 samples/sec	accuracy=0.442622	lr=2.558841
Epoch[33] Batch [349]	Speed: 11435.836206 samples/sec	accuracy=0.442236	lr=2.558611
Epoch[33] Batch [399]	Speed: 10428.352885 samples/sec	accuracy=0.442115	lr=2.558381
Epoch[33] Batch [449]	Speed: 10884.977443 samples/sec	accuracy=0.440821	lr=2.558149
Epoch[33] Batch [499]	Speed: 10849.564218 samples/sec	accuracy=0.439709	lr=2.557917
Epoch[33] Batch [549]	Speed: 10536.258414 samples/sec	accuracy=0.439189	lr=2.557685
Epoch[33] Batch [599]	Speed: 11057.405456 samples/sec	accuracy=0.438888	lr=2.557451
[Epoch 33] training: accuracy=0.438912
[Epoch 33] speed: 10948 samples/sec	time cost: 121.682966
[Epoch 33] validation: err-top1=0.566426 err-top5=0.308516
Epoch[34] Batch [49]	Speed: 10970.762519 samples/sec	accuracy=0.446982	lr=2.557096
Epoch[34] Batch [99]	Speed: 10659.220863 samples/sec	accuracy=0.444937	lr=2.556861
Epoch[34] Batch [149]	Speed: 11112.369880 samples/sec	accuracy=0.443356	lr=2.556625
Epoch[34] Batch [199]	Speed: 11436.663875 samples/sec	accuracy=0.441746	lr=2.556389
Epoch[34] Batch [249]	Speed: 11279.776397 samples/sec	accuracy=0.438930	lr=2.556153
Epoch[34] Batch [299]	Speed: 10479.231459 samples/sec	accuracy=0.436803	lr=2.555915
Epoch[34] Batch [349]	Speed: 10975.155473 samples/sec	accuracy=0.436352	lr=2.555677
Epoch[34] Batch [399]	Speed: 11029.441169 samples/sec	accuracy=0.436243	lr=2.555439
Epoch[34] Batch [449]	Speed: 11501.282533 samples/sec	accuracy=0.436182	lr=2.555200
Epoch[34] Batch [499]	Speed: 11163.313149 samples/sec	accuracy=0.436146	lr=2.554960
Epoch[34] Batch [549]	Speed: 10659.723513 samples/sec	accuracy=0.436217	lr=2.554719
Epoch[34] Batch [599]	Speed: 11036.170408 samples/sec	accuracy=0.436281	lr=2.554478
[Epoch 34] training: accuracy=0.436216
[Epoch 34] speed: 10998 samples/sec	time cost: 120.770975
[Epoch 34] validation: err-top1=0.570475 err-top5=0.310750
Epoch[35] Batch [49]	Speed: 11310.285620 samples/sec	accuracy=0.443965	lr=2.554115
Epoch[35] Batch [99]	Speed: 11670.401484 samples/sec	accuracy=0.442354	lr=2.553873
Epoch[35] Batch [149]	Speed: 10975.160240 samples/sec	accuracy=0.441025	lr=2.553629
Epoch[35] Batch [199]	Speed: 11348.772950 samples/sec	accuracy=0.441174	lr=2.553385
Epoch[35] Batch [249]	Speed: 10284.964893 samples/sec	accuracy=0.440809	lr=2.553141
Epoch[35] Batch [299]	Speed: 11149.319537 samples/sec	accuracy=0.440317	lr=2.552896
Epoch[35] Batch [349]	Speed: 11132.305611 samples/sec	accuracy=0.439681	lr=2.552650
Epoch[35] Batch [399]	Speed: 11125.875686 samples/sec	accuracy=0.439297	lr=2.552403
Epoch[35] Batch [449]	Speed: 11455.894114 samples/sec	accuracy=0.438917	lr=2.552156
Epoch[35] Batch [499]	Speed: 10735.063504 samples/sec	accuracy=0.438552	lr=2.551909
Epoch[35] Batch [549]	Speed: 11208.999044 samples/sec	accuracy=0.438601	lr=2.551660
Epoch[35] Batch [599]	Speed: 11338.224821 samples/sec	accuracy=0.438096	lr=2.551411
[Epoch 35] training: accuracy=0.438094
[Epoch 35] speed: 11105 samples/sec	time cost: 120.018265
[Epoch 35] validation: err-top1=0.539408 err-top5=0.282206
Epoch[36] Batch [49]	Speed: 10951.268311 samples/sec	accuracy=0.445303	lr=2.551032
Epoch[36] Batch [99]	Speed: 11339.859922 samples/sec	accuracy=0.444888	lr=2.550781
Epoch[36] Batch [149]	Speed: 11480.958401 samples/sec	accuracy=0.443356	lr=2.550530
Epoch[36] Batch [199]	Speed: 11462.803656 samples/sec	accuracy=0.441743	lr=2.550278
Epoch[36] Batch [249]	Speed: 11084.603541 samples/sec	accuracy=0.441529	lr=2.550026
Epoch[36] Batch [299]	Speed: 11108.685794 samples/sec	accuracy=0.440986	lr=2.549773
Epoch[36] Batch [349]	Speed: 10942.347899 samples/sec	accuracy=0.440872	lr=2.549519
Epoch[36] Batch [399]	Speed: 11247.774773 samples/sec	accuracy=0.440573	lr=2.549265
Epoch[36] Batch [449]	Speed: 11537.388450 samples/sec	accuracy=0.440312	lr=2.549010
Epoch[36] Batch [499]	Speed: 10994.515080 samples/sec	accuracy=0.440090	lr=2.548754
Epoch[36] Batch [549]	Speed: 11183.661720 samples/sec	accuracy=0.439933	lr=2.548498
Epoch[36] Batch [599]	Speed: 11137.595672 samples/sec	accuracy=0.439655	lr=2.548241
[Epoch 36] training: accuracy=0.439696
[Epoch 36] speed: 11169 samples/sec	time cost: 119.549999
[Epoch 36] validation: err-top1=0.563887 err-top5=0.308574
Epoch[37] Batch [49]	Speed: 11269.089960 samples/sec	accuracy=0.445459	lr=2.547850
Epoch[37] Batch [99]	Speed: 10879.878278 samples/sec	accuracy=0.445557	lr=2.547591
Epoch[37] Batch [149]	Speed: 10992.513251 samples/sec	accuracy=0.444567	lr=2.547332
Epoch[37] Batch [199]	Speed: 11281.828220 samples/sec	accuracy=0.443591	lr=2.547073
Epoch[37] Batch [249]	Speed: 10963.562268 samples/sec	accuracy=0.442713	lr=2.546812
Epoch[37] Batch [299]	Speed: 11090.584598 samples/sec	accuracy=0.442043	lr=2.546552
Epoch[37] Batch [349]	Speed: 11099.142341 samples/sec	accuracy=0.441888	lr=2.546290
Epoch[37] Batch [399]	Speed: 11074.944620 samples/sec	accuracy=0.441703	lr=2.546028
Epoch[37] Batch [449]	Speed: 11606.682127 samples/sec	accuracy=0.441525	lr=2.545765
Epoch[37] Batch [499]	Speed: 11254.826237 samples/sec	accuracy=0.441214	lr=2.545502
Epoch[37] Batch [549]	Speed: 11556.007379 samples/sec	accuracy=0.440925	lr=2.545238
Epoch[37] Batch [599]	Speed: 11654.053862 samples/sec	accuracy=0.440742	lr=2.544973
[Epoch 37] training: accuracy=0.440901
[Epoch 37] speed: 11184 samples/sec	time cost: 118.961980
[Epoch 37] validation: err-top1=0.520610 err-top5=0.263245
Epoch[38] Batch [49]	Speed: 10182.678636 samples/sec	accuracy=0.445723	lr=2.544575
Epoch[38] Batch [99]	Speed: 11023.037827 samples/sec	accuracy=0.444888	lr=2.544309
Epoch[38] Batch [149]	Speed: 11321.655278 samples/sec	accuracy=0.444463	lr=2.544042
Epoch[38] Batch [199]	Speed: 11200.973029 samples/sec	accuracy=0.443533	lr=2.543775
Epoch[38] Batch [249]	Speed: 11220.641623 samples/sec	accuracy=0.443518	lr=2.543507
Epoch[38] Batch [299]	Speed: 11017.382037 samples/sec	accuracy=0.442287	lr=2.543238
Epoch[38] Batch [349]	Speed: 11783.219958 samples/sec	accuracy=0.442024	lr=2.542969
Epoch[38] Batch [399]	Speed: 11133.521952 samples/sec	accuracy=0.441984	lr=2.542699
Epoch[38] Batch [449]	Speed: 11676.693192 samples/sec	accuracy=0.441149	lr=2.542428
Epoch[38] Batch [499]	Speed: 11400.260448 samples/sec	accuracy=0.440725	lr=2.542157
Epoch[38] Batch [549]	Speed: 11249.277228 samples/sec	accuracy=0.440679	lr=2.541885
Epoch[38] Batch [599]	Speed: 11405.724400 samples/sec	accuracy=0.440504	lr=2.541613
[Epoch 38] training: accuracy=0.440428
[Epoch 38] speed: 11186 samples/sec	time cost: 119.349768
[Epoch 38] validation: err-top1=0.575898 err-top5=0.317969
Epoch[39] Batch [49]	Speed: 11182.324347 samples/sec	accuracy=0.448457	lr=2.541198
Epoch[39] Batch [99]	Speed: 11307.319287 samples/sec	accuracy=0.445752	lr=2.540924
Epoch[39] Batch [149]	Speed: 11452.598974 samples/sec	accuracy=0.444346	lr=2.540649
Epoch[39] Batch [199]	Speed: 10851.262360 samples/sec	accuracy=0.444124	lr=2.540374
Epoch[39] Batch [249]	Speed: 11525.646855 samples/sec	accuracy=0.443385	lr=2.540098
Epoch[39] Batch [299]	Speed: 11375.773164 samples/sec	accuracy=0.443127	lr=2.539821
Epoch[39] Batch [349]	Speed: 11113.285388 samples/sec	accuracy=0.442953	lr=2.539544
Epoch[39] Batch [399]	Speed: 11591.955775 samples/sec	accuracy=0.442524	lr=2.539267
Epoch[39] Batch [449]	Speed: 11350.323808 samples/sec	accuracy=0.442760	lr=2.538988
Epoch[39] Batch [499]	Speed: 11019.681033 samples/sec	accuracy=0.442337	lr=2.538709
Epoch[39] Batch [549]	Speed: 10940.131213 samples/sec	accuracy=0.442139	lr=2.538430
Epoch[39] Batch [599]	Speed: 11042.008738 samples/sec	accuracy=0.441881	lr=2.538150
[Epoch 39] training: accuracy=0.441891
[Epoch 39] speed: 11220 samples/sec	time cost: 118.499798
[Epoch 39] validation: err-top1=0.538289 err-top5=0.282511
Epoch[40] Batch [49]	Speed: 10868.904266 samples/sec	accuracy=0.450449	lr=2.537728
Epoch[40] Batch [99]	Speed: 11134.658599 samples/sec	accuracy=0.448955	lr=2.537446
Epoch[40] Batch [149]	Speed: 11065.331026 samples/sec	accuracy=0.448314	lr=2.537164
Epoch[40] Batch [199]	Speed: 11018.601943 samples/sec	accuracy=0.446904	lr=2.536881
Epoch[40] Batch [249]	Speed: 11083.090695 samples/sec	accuracy=0.446240	lr=2.536597
Epoch[40] Batch [299]	Speed: 10996.541289 samples/sec	accuracy=0.445436	lr=2.536313
Epoch[40] Batch [349]	Speed: 11204.746654 samples/sec	accuracy=0.445086	lr=2.536028
Epoch[40] Batch [399]	Speed: 11587.400097 samples/sec	accuracy=0.444917	lr=2.535743
Epoch[40] Batch [449]	Speed: 10671.864710 samples/sec	accuracy=0.444459	lr=2.535457
Epoch[40] Batch [499]	Speed: 11531.805703 samples/sec	accuracy=0.444174	lr=2.535170
Epoch[40] Batch [549]	Speed: 10759.554140 samples/sec	accuracy=0.444076	lr=2.534883
Epoch[40] Batch [599]	Speed: 10764.072775 samples/sec	accuracy=0.443808	lr=2.534595
[Epoch 40] training: accuracy=0.443564
[Epoch 40] speed: 11013 samples/sec	time cost: 120.811312
[Epoch 40] validation: err-top1=0.545247 err-top5=0.288167
Epoch[41] Batch [49]	Speed: 10633.479171 samples/sec	accuracy=0.448252	lr=2.534156
Epoch[41] Batch [99]	Speed: 10916.443825 samples/sec	accuracy=0.448481	lr=2.533866
Epoch[41] Batch [149]	Speed: 10764.154246 samples/sec	accuracy=0.448223	lr=2.533576
Epoch[41] Batch [199]	Speed: 10495.753259 samples/sec	accuracy=0.447317	lr=2.533286
Epoch[41] Batch [249]	Speed: 10319.060565 samples/sec	accuracy=0.446553	lr=2.532994
Epoch[41] Batch [299]	Speed: 11138.201066 samples/sec	accuracy=0.445991	lr=2.532702
Epoch[41] Batch [349]	Speed: 11291.534190 samples/sec	accuracy=0.445275	lr=2.532410
Epoch[41] Batch [399]	Speed: 11352.455391 samples/sec	accuracy=0.445073	lr=2.532116
Epoch[41] Batch [449]	Speed: 11071.213386 samples/sec	accuracy=0.444780	lr=2.531823
Epoch[41] Batch [499]	Speed: 11069.189523 samples/sec	accuracy=0.444758	lr=2.531528
Epoch[41] Batch [549]	Speed: 11118.033557 samples/sec	accuracy=0.444449	lr=2.531233
Epoch[41] Batch [599]	Speed: 11236.481510 samples/sec	accuracy=0.443806	lr=2.530937
[Epoch 41] training: accuracy=0.443676
[Epoch 41] speed: 10928 samples/sec	time cost: 122.074825
[Epoch 41] validation: err-top1=0.548672 err-top5=0.293633
Epoch[42] Batch [49]	Speed: 10750.069190 samples/sec	accuracy=0.450469	lr=2.530493
Epoch[42] Batch [99]	Speed: 10547.159610 samples/sec	accuracy=0.449912	lr=2.530196
Epoch[42] Batch [149]	Speed: 10843.366059 samples/sec	accuracy=0.448574	lr=2.529898
Epoch[42] Batch [199]	Speed: 11428.711574 samples/sec	accuracy=0.447739	lr=2.529599
Epoch[42] Batch [249]	Speed: 10403.541543 samples/sec	accuracy=0.447105	lr=2.529300
Epoch[42] Batch [299]	Speed: 10954.574327 samples/sec	accuracy=0.446610	lr=2.529000
Epoch[42] Batch [349]	Speed: 11308.232366 samples/sec	accuracy=0.446384	lr=2.528700
Epoch[42] Batch [399]	Speed: 10814.446024 samples/sec	accuracy=0.445953	lr=2.528399
Epoch[42] Batch [449]	Speed: 10830.437346 samples/sec	accuracy=0.445837	lr=2.528098
Epoch[42] Batch [499]	Speed: 11152.165606 samples/sec	accuracy=0.445313	lr=2.527796
Epoch[42] Batch [549]	Speed: 10445.691236 samples/sec	accuracy=0.444998	lr=2.527493
Epoch[42] Batch [599]	Speed: 11452.812748 samples/sec	accuracy=0.444931	lr=2.527189
[Epoch 42] training: accuracy=0.444961
[Epoch 42] speed: 10862 samples/sec	time cost: 122.726501
[Epoch 42] validation: err-top1=0.547750 err-top5=0.291443
Epoch[43] Batch [49]	Speed: 11027.919557 samples/sec	accuracy=0.453447	lr=2.526727
Epoch[43] Batch [99]	Speed: 10833.412021 samples/sec	accuracy=0.452007	lr=2.526422
Epoch[43] Batch [149]	Speed: 10732.117915 samples/sec	accuracy=0.450104	lr=2.526117
Epoch[43] Batch [199]	Speed: 10239.346721 samples/sec	accuracy=0.449080	lr=2.525810
Epoch[43] Batch [249]	Speed: 11380.997125 samples/sec	accuracy=0.448471	lr=2.525504
Epoch[43] Batch [299]	Speed: 11330.706197 samples/sec	accuracy=0.447581	lr=2.525196
Epoch[43] Batch [349]	Speed: 11281.111108 samples/sec	accuracy=0.447549	lr=2.524888
Epoch[43] Batch [399]	Speed: 11070.748799 samples/sec	accuracy=0.447174	lr=2.524580
Epoch[43] Batch [449]	Speed: 10498.238961 samples/sec	accuracy=0.446753	lr=2.524270
Epoch[43] Batch [499]	Speed: 11247.965061 samples/sec	accuracy=0.446242	lr=2.523961
Epoch[43] Batch [549]	Speed: 10648.634980 samples/sec	accuracy=0.445980	lr=2.523650
Epoch[43] Batch [599]	Speed: 10300.266682 samples/sec	accuracy=0.444469	lr=2.523339
[Epoch 43] training: accuracy=0.444038
[Epoch 43] speed: 10833 samples/sec	time cost: 123.320457
[Epoch 43] validation: err-top1=0.547539 err-top5=0.293652
Epoch[44] Batch [49]	Speed: 10332.421666 samples/sec	accuracy=0.444824	lr=2.522865
Epoch[44] Batch [99]	Speed: 11174.103606 samples/sec	accuracy=0.445332	lr=2.522552
Epoch[44] Batch [149]	Speed: 10740.881637 samples/sec	accuracy=0.444574	lr=2.522239
Epoch[44] Batch [199]	Speed: 11331.744439 samples/sec	accuracy=0.444333	lr=2.521925
Epoch[44] Batch [249]	Speed: 11181.328149 samples/sec	accuracy=0.443748	lr=2.521611
Epoch[44] Batch [299]	Speed: 10784.888195 samples/sec	accuracy=0.444157	lr=2.521296
Epoch[44] Batch [349]	Speed: 10756.210954 samples/sec	accuracy=0.443775	lr=2.520980
Epoch[44] Batch [399]	Speed: 10732.872868 samples/sec	accuracy=0.443628	lr=2.520664
Epoch[44] Batch [449]	Speed: 10917.501055 samples/sec	accuracy=0.443464	lr=2.520347
Epoch[44] Batch [499]	Speed: 10688.736656 samples/sec	accuracy=0.443379	lr=2.520029
Epoch[44] Batch [549]	Speed: 10149.664851 samples/sec	accuracy=0.443343	lr=2.519711
Epoch[44] Batch [599]	Speed: 10924.234947 samples/sec	accuracy=0.443260	lr=2.519393
[Epoch 44] training: accuracy=0.443221
[Epoch 44] speed: 10802 samples/sec	time cost: 123.182939
[Epoch 44] validation: err-top1=0.526652 err-top5=0.270976
Epoch[45] Batch [49]	Speed: 10468.729146 samples/sec	accuracy=0.453955	lr=2.518913
Epoch[45] Batch [99]	Speed: 11286.685643 samples/sec	accuracy=0.451655	lr=2.518593
Epoch[45] Batch [149]	Speed: 10863.268115 samples/sec	accuracy=0.449971	lr=2.518272
Epoch[45] Batch [199]	Speed: 10889.246183 samples/sec	accuracy=0.449077	lr=2.517951
Epoch[45] Batch [249]	Speed: 11116.869801 samples/sec	accuracy=0.449205	lr=2.517629
Epoch[45] Batch [299]	Speed: 11007.079018 samples/sec	accuracy=0.448265	lr=2.517306
Epoch[45] Batch [349]	Speed: 11110.258805 samples/sec	accuracy=0.447635	lr=2.516983
Epoch[45] Batch [399]	Speed: 10198.376183 samples/sec	accuracy=0.447479	lr=2.516659
Epoch[45] Batch [449]	Speed: 10942.014489 samples/sec	accuracy=0.447426	lr=2.516334
Epoch[45] Batch [499]	Speed: 11225.919751 samples/sec	accuracy=0.447300	lr=2.516009
Epoch[45] Batch [549]	Speed: 10375.967771 samples/sec	accuracy=0.446920	lr=2.515683
Epoch[45] Batch [599]	Speed: 10541.484459 samples/sec	accuracy=0.446648	lr=2.515357
[Epoch 45] training: accuracy=0.446678
[Epoch 45] speed: 10825 samples/sec	time cost: 123.315152
[Epoch 45] validation: err-top1=0.550488 err-top5=0.293730
Epoch[46] Batch [49]	Speed: 11439.134806 samples/sec	accuracy=0.451738	lr=2.514860
Epoch[46] Batch [99]	Speed: 11350.044556 samples/sec	accuracy=0.450630	lr=2.514532
Epoch[46] Batch [149]	Speed: 10705.882457 samples/sec	accuracy=0.450286	lr=2.514203
Epoch[46] Batch [199]	Speed: 10735.190688 samples/sec	accuracy=0.450125	lr=2.513874
Epoch[46] Batch [249]	Speed: 10514.292597 samples/sec	accuracy=0.450473	lr=2.513544
Epoch[46] Batch [299]	Speed: 10803.890799 samples/sec	accuracy=0.449442	lr=2.513214
Epoch[46] Batch [349]	Speed: 10563.098853 samples/sec	accuracy=0.448892	lr=2.512883
Epoch[46] Batch [399]	Speed: 10629.991019 samples/sec	accuracy=0.448395	lr=2.512552
Epoch[46] Batch [449]	Speed: 10060.119205 samples/sec	accuracy=0.448301	lr=2.512219
Epoch[46] Batch [499]	Speed: 11442.239604 samples/sec	accuracy=0.447762	lr=2.511887
Epoch[46] Batch [549]	Speed: 10224.042241 samples/sec	accuracy=0.447699	lr=2.511553
Epoch[46] Batch [599]	Speed: 11056.075338 samples/sec	accuracy=0.447288	lr=2.511219
[Epoch 46] training: accuracy=0.447248
[Epoch 46] speed: 10767 samples/sec	time cost: 123.593743
[Epoch 46] validation: err-top1=0.540975 err-top5=0.284892
Epoch[47] Batch [49]	Speed: 10456.041714 samples/sec	accuracy=0.457402	lr=2.510717
Epoch[47] Batch [99]	Speed: 11068.835501 samples/sec	accuracy=0.454878	lr=2.510382
Epoch[47] Batch [149]	Speed: 11203.469403 samples/sec	accuracy=0.454137	lr=2.510046
Epoch[47] Batch [199]	Speed: 10392.241246 samples/sec	accuracy=0.452815	lr=2.509709
Epoch[47] Batch [249]	Speed: 10588.741195 samples/sec	accuracy=0.452365	lr=2.509372
Epoch[47] Batch [299]	Speed: 10180.406945 samples/sec	accuracy=0.451123	lr=2.509034
Epoch[47] Batch [349]	Speed: 10451.320399 samples/sec	accuracy=0.450056	lr=2.508695
Epoch[47] Batch [399]	Speed: 10233.536309 samples/sec	accuracy=0.449232	lr=2.508356
Epoch[47] Batch [449]	Speed: 11165.232548 samples/sec	accuracy=0.448827	lr=2.508016
Epoch[47] Batch [499]	Speed: 10877.847719 samples/sec	accuracy=0.448194	lr=2.507676
Epoch[47] Batch [549]	Speed: 11149.284517 samples/sec	accuracy=0.447608	lr=2.507335
Epoch[47] Batch [599]	Speed: 10617.468977 samples/sec	accuracy=0.447164	lr=2.506993
[Epoch 47] training: accuracy=0.446975
[Epoch 47] speed: 10675 samples/sec	time cost: 124.852060
[Epoch 47] validation: err-top1=0.561483 err-top5=0.307495
Epoch[48] Batch [49]	Speed: 10481.073200 samples/sec	accuracy=0.455977	lr=2.506473
Epoch[48] Batch [99]	Speed: 11083.502261 samples/sec	accuracy=0.452354	lr=2.506130
Epoch[48] Batch [149]	Speed: 10967.340864 samples/sec	accuracy=0.450602	lr=2.505786
Epoch[48] Batch [199]	Speed: 10935.011993 samples/sec	accuracy=0.449380	lr=2.505442
Epoch[48] Batch [249]	Speed: 10339.467640 samples/sec	accuracy=0.448654	lr=2.505097
Epoch[48] Batch [299]	Speed: 10785.104038 samples/sec	accuracy=0.448354	lr=2.504752
Epoch[48] Batch [349]	Speed: 10771.180775 samples/sec	accuracy=0.448241	lr=2.504405
Epoch[48] Batch [399]	Speed: 10402.465609 samples/sec	accuracy=0.448025	lr=2.504059
Epoch[48] Batch [449]	Speed: 10369.990049 samples/sec	accuracy=0.447615	lr=2.503711
Epoch[48] Batch [499]	Speed: 10245.748831 samples/sec	accuracy=0.447147	lr=2.503364
Epoch[48] Batch [549]	Speed: 10023.180669 samples/sec	accuracy=0.446871	lr=2.503015
Epoch[48] Batch [599]	Speed: 10680.825353 samples/sec	accuracy=0.446555	lr=2.502666
[Epoch 48] training: accuracy=0.446615
[Epoch 48] speed: 10580 samples/sec	time cost: 125.837350
[Epoch 48] validation: err-top1=0.535371 err-top5=0.276680
Epoch[49] Batch [49]	Speed: 9940.233391 samples/sec	accuracy=0.452715	lr=2.502141
Epoch[49] Batch [99]	Speed: 11185.127870 samples/sec	accuracy=0.453740	lr=2.501790
Epoch[49] Batch [149]	Speed: 11161.946698 samples/sec	accuracy=0.452620	lr=2.501439
Epoch[49] Batch [199]	Speed: 11374.144559 samples/sec	accuracy=0.451440	lr=2.501087
Epoch[49] Batch [249]	Speed: 10690.840120 samples/sec	accuracy=0.450656	lr=2.500735
Epoch[49] Batch [299]	Speed: 10721.104204 samples/sec	accuracy=0.450189	lr=2.500382
Epoch[49] Batch [349]	Speed: 10119.185108 samples/sec	accuracy=0.449678	lr=2.500028
Epoch[49] Batch [399]	Speed: 10346.392128 samples/sec	accuracy=0.449124	lr=2.499674
Epoch[49] Batch [449]	Speed: 9840.503123 samples/sec	accuracy=0.448475	lr=2.499319
Epoch[49] Batch [499]	Speed: 11580.296801 samples/sec	accuracy=0.448226	lr=2.498964
Epoch[49] Batch [549]	Speed: 10898.723157 samples/sec	accuracy=0.447607	lr=2.498608
Epoch[49] Batch [599]	Speed: 10419.675700 samples/sec	accuracy=0.447274	lr=2.498251
[Epoch 49] training: accuracy=0.447106
[Epoch 49] speed: 10635 samples/sec	time cost: 125.229000
[Epoch 49] validation: err-top1=0.549093 err-top5=0.291077
Epoch[50] Batch [49]	Speed: 11044.572779 samples/sec	accuracy=0.455430	lr=2.497708
Epoch[50] Batch [99]	Speed: 10207.006794 samples/sec	accuracy=0.453530	lr=2.497350
Epoch[50] Batch [149]	Speed: 10906.534140 samples/sec	accuracy=0.453652	lr=2.496991
Epoch[50] Batch [199]	Speed: 10372.962642 samples/sec	accuracy=0.452788	lr=2.496632
Epoch[50] Batch [249]	Speed: 10727.580512 samples/sec	accuracy=0.451877	lr=2.496272
Epoch[50] Batch [299]	Speed: 10129.851047 samples/sec	accuracy=0.451121	lr=2.495911
Epoch[50] Batch [349]	Speed: 10078.157105 samples/sec	accuracy=0.450499	lr=2.495550
Epoch[50] Batch [399]	Speed: 10895.185428 samples/sec	accuracy=0.449956	lr=2.495188
Epoch[50] Batch [449]	Speed: 11000.415909 samples/sec	accuracy=0.449844	lr=2.494826
Epoch[50] Batch [499]	Speed: 11319.546283 samples/sec	accuracy=0.449665	lr=2.494463
Epoch[50] Batch [549]	Speed: 11201.207893 samples/sec	accuracy=0.449337	lr=2.494099
Epoch[50] Batch [599]	Speed: 11013.279978 samples/sec	accuracy=0.449106	lr=2.493735
[Epoch 50] training: accuracy=0.449005
[Epoch 50] speed: 10705 samples/sec	time cost: 124.663941
[Epoch 50] validation: err-top1=0.557363 err-top5=0.300625
Epoch[51] Batch [49]	Speed: 11014.622134 samples/sec	accuracy=0.455635	lr=2.493181
Epoch[51] Batch [99]	Speed: 11110.603697 samples/sec	accuracy=0.454292	lr=2.492815
Epoch[51] Batch [149]	Speed: 10632.713919 samples/sec	accuracy=0.452874	lr=2.492449
Epoch[51] Batch [199]	Speed: 10981.222800 samples/sec	accuracy=0.451577	lr=2.492082
Epoch[51] Batch [249]	Speed: 11141.911117 samples/sec	accuracy=0.450953	lr=2.491714
Epoch[51] Batch [299]	Speed: 11413.278040 samples/sec	accuracy=0.450557	lr=2.491346
Epoch[51] Batch [349]	Speed: 11335.850244 samples/sec	accuracy=0.450388	lr=2.490978
Epoch[51] Batch [399]	Speed: 10629.133412 samples/sec	accuracy=0.450104	lr=2.490608
Epoch[51] Batch [449]	Speed: 10622.250210 samples/sec	accuracy=0.450028	lr=2.490239
Epoch[51] Batch [499]	Speed: 11416.827357 samples/sec	accuracy=0.449809	lr=2.489868
Epoch[51] Batch [549]	Speed: 10474.632749 samples/sec	accuracy=0.449220	lr=2.489497
Epoch[51] Batch [599]	Speed: 11295.328098 samples/sec	accuracy=0.449020	lr=2.489126
[Epoch 51] training: accuracy=0.449032
[Epoch 51] speed: 10951 samples/sec	time cost: 121.713512
[Epoch 51] validation: err-top1=0.590637 err-top5=0.338460
Epoch[52] Batch [49]	Speed: 9901.985952 samples/sec	accuracy=0.457969	lr=2.488567
Epoch[52] Batch [99]	Speed: 9988.566890 samples/sec	accuracy=0.454644	lr=2.488194
Epoch[52] Batch [149]	Speed: 10828.705855 samples/sec	accuracy=0.454290	lr=2.487820
Epoch[52] Batch [199]	Speed: 10529.541143 samples/sec	accuracy=0.453701	lr=2.487446
Epoch[52] Batch [249]	Speed: 11250.569955 samples/sec	accuracy=0.453293	lr=2.487071
Epoch[52] Batch [299]	Speed: 11336.672480 samples/sec	accuracy=0.452814	lr=2.486695
Epoch[52] Batch [349]	Speed: 11395.443268 samples/sec	accuracy=0.452358	lr=2.486319
Epoch[52] Batch [399]	Speed: 10108.119329 samples/sec	accuracy=0.451896	lr=2.485943
Epoch[52] Batch [449]	Speed: 11041.757509 samples/sec	accuracy=0.451229	lr=2.485565
Epoch[52] Batch [499]	Speed: 11065.841630 samples/sec	accuracy=0.450820	lr=2.485188
Epoch[52] Batch [549]	Speed: 11140.525338 samples/sec	accuracy=0.450539	lr=2.484809
Epoch[52] Batch [599]	Speed: 11382.124538 samples/sec	accuracy=0.450295	lr=2.484430
[Epoch 52] training: accuracy=0.450184
[Epoch 52] speed: 10803 samples/sec	time cost: 123.434625
[Epoch 52] validation: err-top1=0.594543 err-top5=0.336487
Epoch[53] Batch [49]	Speed: 10209.224365 samples/sec	accuracy=0.454717	lr=2.483853
Epoch[53] Batch [99]	Speed: 10632.707338 samples/sec	accuracy=0.453765	lr=2.483472
Epoch[53] Batch [149]	Speed: 11017.717512 samples/sec	accuracy=0.452751	lr=2.483091
Epoch[53] Batch [199]	Speed: 10952.469150 samples/sec	accuracy=0.452390	lr=2.482709
Epoch[53] Batch [249]	Speed: 10668.824897 samples/sec	accuracy=0.451930	lr=2.482327
Epoch[53] Batch [299]	Speed: 11084.670769 samples/sec	accuracy=0.451603	lr=2.481944
Epoch[53] Batch [349]	Speed: 10848.713565 samples/sec	accuracy=0.451444	lr=2.481560
Epoch[53] Batch [399]	Speed: 11025.262202 samples/sec	accuracy=0.451064	lr=2.481176
Epoch[53] Batch [449]	Speed: 10221.673483 samples/sec	accuracy=0.450948	lr=2.480792
Epoch[53] Batch [499]	Speed: 10763.363058 samples/sec	accuracy=0.450575	lr=2.480406
Epoch[53] Batch [549]	Speed: 10821.825503 samples/sec	accuracy=0.450519	lr=2.480020
Epoch[53] Batch [599]	Speed: 11109.990091 samples/sec	accuracy=0.450545	lr=2.479634
[Epoch 53] training: accuracy=0.450355
[Epoch 53] speed: 10747 samples/sec	time cost: 124.144291
[Epoch 53] validation: err-top1=0.573555 err-top5=0.314688
Epoch[54] Batch [49]	Speed: 10637.901206 samples/sec	accuracy=0.455088	lr=2.479053
Epoch[54] Batch [99]	Speed: 11431.102705 samples/sec	accuracy=0.455532	lr=2.478665
Epoch[54] Batch [149]	Speed: 11081.235170 samples/sec	accuracy=0.455661	lr=2.478277
Epoch[54] Batch [199]	Speed: 10948.726471 samples/sec	accuracy=0.454734	lr=2.477887
Epoch[54] Batch [249]	Speed: 10534.588176 samples/sec	accuracy=0.453867	lr=2.477498
Epoch[54] Batch [299]	Speed: 11219.911165 samples/sec	accuracy=0.452855	lr=2.477107
Epoch[54] Batch [349]	Speed: 10827.295075 samples/sec	accuracy=0.452444	lr=2.476717
Epoch[54] Batch [399]	Speed: 10522.764627 samples/sec	accuracy=0.452073	lr=2.476325
Epoch[54] Batch [449]	Speed: 11051.052355 samples/sec	accuracy=0.451695	lr=2.475933
Epoch[54] Batch [499]	Speed: 11026.637005 samples/sec	accuracy=0.451687	lr=2.475540
Epoch[54] Batch [549]	Speed: 10963.662180 samples/sec	accuracy=0.451556	lr=2.475147
Epoch[54] Batch [599]	Speed: 10557.252653 samples/sec	accuracy=0.451217	lr=2.474753
[Epoch 54] training: accuracy=0.451151
[Epoch 54] speed: 10875 samples/sec	time cost: 122.633994
[Epoch 54] validation: err-top1=0.529928 err-top5=0.274028
Epoch[55] Batch [49]	Speed: 10039.785662 samples/sec	accuracy=0.460430	lr=2.474153
Epoch[55] Batch [99]	Speed: 10273.708008 samples/sec	accuracy=0.457046	lr=2.473758
Epoch[55] Batch [149]	Speed: 10285.136313 samples/sec	accuracy=0.455081	lr=2.473362
Epoch[55] Batch [199]	Speed: 11427.321644 samples/sec	accuracy=0.454417	lr=2.472966
Epoch[55] Batch [249]	Speed: 11318.246306 samples/sec	accuracy=0.454270	lr=2.472568
Epoch[55] Batch [299]	Speed: 11356.606549 samples/sec	accuracy=0.453241	lr=2.472171
Epoch[55] Batch [349]	Speed: 11273.743771 samples/sec	accuracy=0.452263	lr=2.471772
Epoch[55] Batch [399]	Speed: 10620.255055 samples/sec	accuracy=0.451605	lr=2.471374
Epoch[55] Batch [449]	Speed: 10881.752444 samples/sec	accuracy=0.451208	lr=2.470974
Epoch[55] Batch [499]	Speed: 10972.784467 samples/sec	accuracy=0.450898	lr=2.470574
Epoch[55] Batch [549]	Speed: 11254.425443 samples/sec	accuracy=0.450953	lr=2.470173
Epoch[55] Batch [599]	Speed: 10227.957538 samples/sec	accuracy=0.451096	lr=2.469772
[Epoch 55] training: accuracy=0.450813
[Epoch 55] speed: 10772 samples/sec	time cost: 123.762580
[Epoch 55] validation: err-top1=0.529785 err-top5=0.276719
Epoch[56] Batch [49]	Speed: 10748.654077 samples/sec	accuracy=0.458525	lr=2.469169
Epoch[56] Batch [99]	Speed: 10866.779370 samples/sec	accuracy=0.456626	lr=2.468767
Epoch[56] Batch [149]	Speed: 10819.687628 samples/sec	accuracy=0.455898	lr=2.468363
Epoch[56] Batch [199]	Speed: 11164.305269 samples/sec	accuracy=0.454744	lr=2.467960
Epoch[56] Batch [249]	Speed: 11135.498965 samples/sec	accuracy=0.453996	lr=2.467555
Epoch[56] Batch [299]	Speed: 10726.263462 samples/sec	accuracy=0.453854	lr=2.467150
Epoch[56] Batch [349]	Speed: 10653.951984 samples/sec	accuracy=0.453160	lr=2.466744
Epoch[56] Batch [399]	Speed: 11290.656751 samples/sec	accuracy=0.452975	lr=2.466338
Epoch[56] Batch [449]	Speed: 11083.942461 samples/sec	accuracy=0.452074	lr=2.465932
Epoch[56] Batch [499]	Speed: 10950.260367 samples/sec	accuracy=0.451871	lr=2.465524
Epoch[56] Batch [549]	Speed: 10973.859368 samples/sec	accuracy=0.451646	lr=2.465116
Epoch[56] Batch [599]	Speed: 10675.931281 samples/sec	accuracy=0.451258	lr=2.464708
[Epoch 56] training: accuracy=0.451136
[Epoch 56] speed: 10891 samples/sec	time cost: 122.430056
[Epoch 56] validation: err-top1=0.560994 err-top5=0.305359
Epoch[57] Batch [49]	Speed: 10383.973985 samples/sec	accuracy=0.456416	lr=2.464086
Epoch[57] Batch [99]	Speed: 9679.416624 samples/sec	accuracy=0.455088	lr=2.463676
Epoch[57] Batch [149]	Speed: 10848.195401 samples/sec	accuracy=0.455241	lr=2.463265
Epoch[57] Batch [199]	Speed: 10676.207539 samples/sec	accuracy=0.454846	lr=2.462854
Epoch[57] Batch [249]	Speed: 9825.469274 samples/sec	accuracy=0.454244	lr=2.462442
Epoch[57] Batch [299]	Speed: 11070.367000 samples/sec	accuracy=0.453773	lr=2.462030
Epoch[57] Batch [349]	Speed: 10970.637818 samples/sec	accuracy=0.453537	lr=2.461617
Epoch[57] Batch [399]	Speed: 10959.348891 samples/sec	accuracy=0.453518	lr=2.461203
Epoch[57] Batch [449]	Speed: 10970.831736 samples/sec	accuracy=0.453431	lr=2.460789
Epoch[57] Batch [499]	Speed: 10437.446041 samples/sec	accuracy=0.453091	lr=2.460375
Epoch[57] Batch [549]	Speed: 10999.237771 samples/sec	accuracy=0.452854	lr=2.459959
Epoch[57] Batch [599]	Speed: 10799.015283 samples/sec	accuracy=0.452856	lr=2.459544
[Epoch 57] training: accuracy=0.452669
[Epoch 57] speed: 10621 samples/sec	time cost: 125.677746
[Epoch 57] validation: err-top1=0.540000 err-top5=0.283711
Epoch[58] Batch [49]	Speed: 10655.860944 samples/sec	accuracy=0.456738	lr=2.458910
Epoch[58] Batch [99]	Speed: 10695.358209 samples/sec	accuracy=0.456094	lr=2.458493
Epoch[58] Batch [149]	Speed: 11344.787234 samples/sec	accuracy=0.456012	lr=2.458075
Epoch[58] Batch [199]	Speed: 11117.111223 samples/sec	accuracy=0.455813	lr=2.457657
Epoch[58] Batch [249]	Speed: 10533.636613 samples/sec	accuracy=0.455498	lr=2.457238
Epoch[58] Batch [299]	Speed: 11196.931053 samples/sec	accuracy=0.454867	lr=2.456818
Epoch[58] Batch [349]	Speed: 10845.795119 samples/sec	accuracy=0.453577	lr=2.456398
Epoch[58] Batch [399]	Speed: 11371.619425 samples/sec	accuracy=0.453246	lr=2.455977
Epoch[58] Batch [449]	Speed: 11062.048432 samples/sec	accuracy=0.452721	lr=2.455556
Epoch[58] Batch [499]	Speed: 10778.934925 samples/sec	accuracy=0.452322	lr=2.455134
Epoch[58] Batch [549]	Speed: 10502.660155 samples/sec	accuracy=0.452308	lr=2.454711
Epoch[58] Batch [599]	Speed: 10184.335495 samples/sec	accuracy=0.452083	lr=2.454288
[Epoch 58] training: accuracy=0.451916
[Epoch 58] speed: 10821 samples/sec	time cost: 123.250177
[Epoch 58] validation: err-top1=0.535909 err-top5=0.279012
Epoch[59] Batch [49]	Speed: 10369.252737 samples/sec	accuracy=0.464463	lr=2.453653
Epoch[59] Batch [99]	Speed: 11305.601598 samples/sec	accuracy=0.461084	lr=2.453228
Epoch[59] Batch [149]	Speed: 10081.445073 samples/sec	accuracy=0.459258	lr=2.452803
Epoch[59] Batch [199]	Speed: 10846.247864 samples/sec	accuracy=0.457737	lr=2.452377
Epoch[59] Batch [249]	Speed: 11385.804813 samples/sec	accuracy=0.456639	lr=2.451951
Epoch[59] Batch [299]	Speed: 11193.452653 samples/sec	accuracy=0.456317	lr=2.451524
Epoch[59] Batch [349]	Speed: 11089.400239 samples/sec	accuracy=0.455467	lr=2.451097
Epoch[59] Batch [399]	Speed: 10247.583004 samples/sec	accuracy=0.455010	lr=2.450669
Epoch[59] Batch [449]	Speed: 10620.195443 samples/sec	accuracy=0.454708	lr=2.450240
Epoch[59] Batch [499]	Speed: 10856.403609 samples/sec	accuracy=0.454306	lr=2.449811
Epoch[59] Batch [549]	Speed: 11036.583033 samples/sec	accuracy=0.453923	lr=2.449381
Epoch[59] Batch [599]	Speed: 10821.839409 samples/sec	accuracy=0.453600	lr=2.448951
[Epoch 59] training: accuracy=0.453303
[Epoch 59] speed: 10767 samples/sec	time cost: 123.965705
[Epoch 59] validation: err-top1=0.587484 err-top5=0.326253
Epoch[60] Batch [49]	Speed: 10339.809151 samples/sec	accuracy=0.460010	lr=2.448296
Epoch[60] Batch [99]	Speed: 10202.169333 samples/sec	accuracy=0.454893	lr=2.447864
Epoch[60] Batch [149]	Speed: 10730.491708 samples/sec	accuracy=0.453555	lr=2.447432
Epoch[60] Batch [199]	Speed: 11386.445340 samples/sec	accuracy=0.454409	lr=2.446999
Epoch[60] Batch [249]	Speed: 11053.658718 samples/sec	accuracy=0.454500	lr=2.446565
Epoch[60] Batch [299]	Speed: 10830.552052 samples/sec	accuracy=0.454647	lr=2.446131
Epoch[60] Batch [349]	Speed: 11030.253548 samples/sec	accuracy=0.454523	lr=2.445697
Epoch[60] Batch [399]	Speed: 10379.696280 samples/sec	accuracy=0.453901	lr=2.445262
Epoch[60] Batch [449]	Speed: 11038.561513 samples/sec	accuracy=0.453391	lr=2.444826
Epoch[60] Batch [499]	Speed: 11063.683501 samples/sec	accuracy=0.453289	lr=2.444389
Epoch[60] Batch [549]	Speed: 11443.167896 samples/sec	accuracy=0.453175	lr=2.443953
Epoch[60] Batch [599]	Speed: 10535.871496 samples/sec	accuracy=0.453240	lr=2.443515
[Epoch 60] training: accuracy=0.453079
[Epoch 60] speed: 10796 samples/sec	time cost: 123.648243
[Epoch 60] validation: err-top1=0.558945 err-top5=0.301816
Epoch[61] Batch [49]	Speed: 10853.694959 samples/sec	accuracy=0.461602	lr=2.442858
Epoch[61] Batch [99]	Speed: 10459.274492 samples/sec	accuracy=0.459009	lr=2.442419
Epoch[61] Batch [149]	Speed: 10729.380057 samples/sec	accuracy=0.458799	lr=2.441979
Epoch[61] Batch [199]	Speed: 10557.319086 samples/sec	accuracy=0.457449	lr=2.441539
Epoch[61] Batch [249]	Speed: 10087.236817 samples/sec	accuracy=0.456244	lr=2.441099
Epoch[61] Batch [299]	Speed: 10687.964760 samples/sec	accuracy=0.455557	lr=2.440657
Epoch[61] Batch [349]	Speed: 11077.514267 samples/sec	accuracy=0.454944	lr=2.440216
Epoch[61] Batch [399]	Speed: 11303.361440 samples/sec	accuracy=0.454805	lr=2.439773
Epoch[61] Batch [449]	Speed: 10584.133787 samples/sec	accuracy=0.454656	lr=2.439330
Epoch[61] Batch [499]	Speed: 11059.280347 samples/sec	accuracy=0.454396	lr=2.438887
Epoch[61] Batch [549]	Speed: 10738.728096 samples/sec	accuracy=0.454023	lr=2.438443
Epoch[61] Batch [599]	Speed: 11187.114513 samples/sec	accuracy=0.453840	lr=2.437998
[Epoch 61] training: accuracy=0.453617
[Epoch 61] speed: 10739 samples/sec	time cost: 124.203354
[Epoch 61] validation: err-top1=0.544495 err-top5=0.291951
Epoch[62] Batch [49]	Speed: 10025.531795 samples/sec	accuracy=0.462832	lr=2.437321
Epoch[62] Batch [99]	Speed: 11170.320085 samples/sec	accuracy=0.459492	lr=2.436875
Epoch[62] Batch [149]	Speed: 10698.829968 samples/sec	accuracy=0.456950	lr=2.436428
Epoch[62] Batch [199]	Speed: 11106.535047 samples/sec	accuracy=0.456963	lr=2.435981
Epoch[62] Batch [249]	Speed: 10765.069667 samples/sec	accuracy=0.456734	lr=2.435533
Epoch[62] Batch [299]	Speed: 11023.792672 samples/sec	accuracy=0.456763	lr=2.435085
Epoch[62] Batch [349]	Speed: 10925.042739 samples/sec	accuracy=0.456268	lr=2.434636
Epoch[62] Batch [399]	Speed: 10246.213240 samples/sec	accuracy=0.455869	lr=2.434187
Epoch[62] Batch [449]	Speed: 10214.009495 samples/sec	accuracy=0.455521	lr=2.433737
Epoch[62] Batch [499]	Speed: 11175.942673 samples/sec	accuracy=0.454278	lr=2.433286
Epoch[62] Batch [549]	Speed: 11093.939171 samples/sec	accuracy=0.453825	lr=2.432835
Epoch[62] Batch [599]	Speed: 11285.271033 samples/sec	accuracy=0.453608	lr=2.432383
[Epoch 62] training: accuracy=0.453469
[Epoch 62] speed: 10787 samples/sec	time cost: 123.652786
[Epoch 62] validation: err-top1=0.527988 err-top5=0.268145
Epoch[63] Batch [49]	Speed: 10265.072470 samples/sec	accuracy=0.457959	lr=2.431704
Epoch[63] Batch [99]	Speed: 10936.434276 samples/sec	accuracy=0.455562	lr=2.431251
Epoch[63] Batch [149]	Speed: 11389.225011 samples/sec	accuracy=0.454945	lr=2.430797
Epoch[63] Batch [199]	Speed: 10519.521603 samples/sec	accuracy=0.455261	lr=2.430343
Epoch[63] Batch [249]	Speed: 10463.209434 samples/sec	accuracy=0.455027	lr=2.429888
Epoch[63] Batch [299]	Speed: 10239.663097 samples/sec	accuracy=0.454427	lr=2.429433
Epoch[63] Batch [349]	Speed: 10255.560720 samples/sec	accuracy=0.454332	lr=2.428977
Epoch[63] Batch [399]	Speed: 10686.691986 samples/sec	accuracy=0.454282	lr=2.428520
Epoch[63] Batch [449]	Speed: 11161.741034 samples/sec	accuracy=0.453896	lr=2.428063
Epoch[63] Batch [499]	Speed: 10919.964280 samples/sec	accuracy=0.453577	lr=2.427605
Epoch[63] Batch [549]	Speed: 10817.378681 samples/sec	accuracy=0.453353	lr=2.427147
Epoch[63] Batch [599]	Speed: 10950.174659 samples/sec	accuracy=0.453187	lr=2.426688
[Epoch 63] training: accuracy=0.453084
[Epoch 63] speed: 10680 samples/sec	time cost: 124.801001
[Epoch 63] validation: err-top1=0.562948 err-top5=0.306580
Epoch[64] Batch [49]	Speed: 11205.764279 samples/sec	accuracy=0.456191	lr=2.425989
Epoch[64] Batch [99]	Speed: 11049.240798 samples/sec	accuracy=0.457383	lr=2.425529
Epoch[64] Batch [149]	Speed: 10586.528972 samples/sec	accuracy=0.456546	lr=2.425068
Epoch[64] Batch [199]	Speed: 10774.175147 samples/sec	accuracy=0.456235	lr=2.424607
Epoch[64] Batch [249]	Speed: 10714.459831 samples/sec	accuracy=0.456463	lr=2.424145
Epoch[64] Batch [299]	Speed: 10951.436133 samples/sec	accuracy=0.456017	lr=2.423682
Epoch[64] Batch [349]	Speed: 10775.476415 samples/sec	accuracy=0.455741	lr=2.423219
Epoch[64] Batch [399]	Speed: 10328.814471 samples/sec	accuracy=0.455231	lr=2.422756
Epoch[64] Batch [449]	Speed: 10870.684406 samples/sec	accuracy=0.455153	lr=2.422292
Epoch[64] Batch [499]	Speed: 10561.650980 samples/sec	accuracy=0.455141	lr=2.421827
Epoch[64] Batch [549]	Speed: 11053.380503 samples/sec	accuracy=0.454869	lr=2.421361
Epoch[64] Batch [599]	Speed: 10216.321482 samples/sec	accuracy=0.454769	lr=2.420896
[Epoch 64] training: accuracy=0.454629
[Epoch 64] speed: 10725 samples/sec	time cost: 124.488371
[Epoch 64] validation: err-top1=0.566060 err-top5=0.305664
Epoch[65] Batch [49]	Speed: 10425.105029 samples/sec	accuracy=0.462275	lr=2.420186
Epoch[65] Batch [99]	Speed: 9725.558051 samples/sec	accuracy=0.458149	lr=2.419719
Epoch[65] Batch [149]	Speed: 10691.540040 samples/sec	accuracy=0.457604	lr=2.419251
Epoch[65] Batch [199]	Speed: 11351.635064 samples/sec	accuracy=0.456296	lr=2.418783
Epoch[65] Batch [249]	Speed: 11242.177096 samples/sec	accuracy=0.456158	lr=2.418314
Epoch[65] Batch [299]	Speed: 9987.004230 samples/sec	accuracy=0.455469	lr=2.417844
Epoch[65] Batch [349]	Speed: 10518.044697 samples/sec	accuracy=0.455162	lr=2.417374
Epoch[65] Batch [399]	Speed: 10294.036338 samples/sec	accuracy=0.454756	lr=2.416903
Epoch[65] Batch [449]	Speed: 10568.900035 samples/sec	accuracy=0.454572	lr=2.416432
Epoch[65] Batch [499]	Speed: 10569.046200 samples/sec	accuracy=0.454518	lr=2.415960
Epoch[65] Batch [549]	Speed: 10379.220695 samples/sec	accuracy=0.454604	lr=2.415488
Epoch[65] Batch [599]	Speed: 10636.134321 samples/sec	accuracy=0.454550	lr=2.415015
[Epoch 65] training: accuracy=0.454560
[Epoch 65] speed: 10513 samples/sec	time cost: 126.996207
[Epoch 65] validation: err-top1=0.561152 err-top5=0.301992
Epoch[66] Batch [49]	Speed: 10441.361780 samples/sec	accuracy=0.460703	lr=2.414305
Epoch[66] Batch [99]	Speed: 10770.426366 samples/sec	accuracy=0.460015	lr=2.413830
Epoch[66] Batch [149]	Speed: 11145.533401 samples/sec	accuracy=0.459150	lr=2.413355
Epoch[66] Batch [199]	Speed: 10665.454684 samples/sec	accuracy=0.458215	lr=2.412880
Epoch[66] Batch [249]	Speed: 10905.435282 samples/sec	accuracy=0.458061	lr=2.412404
Epoch[66] Batch [299]	Speed: 10796.723291 samples/sec	accuracy=0.457686	lr=2.411927
Epoch[66] Batch [349]	Speed: 11140.059828 samples/sec	accuracy=0.457327	lr=2.411450
Epoch[66] Batch [399]	Speed: 10189.182403 samples/sec	accuracy=0.456818	lr=2.410973
Epoch[66] Batch [449]	Speed: 10965.373833 samples/sec	accuracy=0.456369	lr=2.410494
Epoch[66] Batch [499]	Speed: 11039.436241 samples/sec	accuracy=0.455761	lr=2.410016
Epoch[66] Batch [549]	Speed: 10572.887459 samples/sec	accuracy=0.455565	lr=2.409536
Epoch[66] Batch [599]	Speed: 10815.432932 samples/sec	accuracy=0.455084	lr=2.409056
[Epoch 66] training: accuracy=0.454939
[Epoch 66] speed: 10753 samples/sec	time cost: 124.246473
[Epoch 66] validation: err-top1=0.574137 err-top5=0.316060
Epoch[67] Batch [49]	Speed: 11130.827891 samples/sec	accuracy=0.459961	lr=2.408326
Epoch[67] Batch [99]	Speed: 11074.215019 samples/sec	accuracy=0.460132	lr=2.407845
Epoch[67] Batch [149]	Speed: 10889.827639 samples/sec	accuracy=0.459469	lr=2.407363
Epoch[67] Batch [199]	Speed: 10551.267503 samples/sec	accuracy=0.459341	lr=2.406880
Epoch[67] Batch [249]	Speed: 11477.922738 samples/sec	accuracy=0.458373	lr=2.406397
Epoch[67] Batch [299]	Speed: 11119.485715 samples/sec	accuracy=0.457646	lr=2.405914
Epoch[67] Batch [349]	Speed: 11233.144782 samples/sec	accuracy=0.457143	lr=2.405430
Epoch[67] Batch [399]	Speed: 11458.601720 samples/sec	accuracy=0.456975	lr=2.404945
Epoch[67] Batch [449]	Speed: 10696.780369 samples/sec	accuracy=0.456480	lr=2.404460
Epoch[67] Batch [499]	Speed: 11135.577495 samples/sec	accuracy=0.456328	lr=2.403974
Epoch[67] Batch [549]	Speed: 11098.010353 samples/sec	accuracy=0.456291	lr=2.403488
Epoch[67] Batch [599]	Speed: 11481.101418 samples/sec	accuracy=0.456120	lr=2.403001
[Epoch 67] training: accuracy=0.455910
[Epoch 67] speed: 11074 samples/sec	time cost: 120.131064
[Epoch 67] validation: err-top1=0.558477 err-top5=0.302402
Epoch[68] Batch [49]	Speed: 9959.720441 samples/sec	accuracy=0.463057	lr=2.402270
Epoch[68] Batch [99]	Speed: 10438.091103 samples/sec	accuracy=0.460229	lr=2.401782
Epoch[68] Batch [149]	Speed: 11438.032015 samples/sec	accuracy=0.459626	lr=2.401293
Epoch[68] Batch [199]	Speed: 10890.852656 samples/sec	accuracy=0.459939	lr=2.400804
Epoch[68] Batch [249]	Speed: 10697.074225 samples/sec	accuracy=0.459375	lr=2.400314
Epoch[68] Batch [299]	Speed: 10992.402684 samples/sec	accuracy=0.458555	lr=2.399823
Epoch[68] Batch [349]	Speed: 10823.540339 samples/sec	accuracy=0.458131	lr=2.399332
Epoch[68] Batch [399]	Speed: 11365.818429 samples/sec	accuracy=0.457971	lr=2.398841
Epoch[68] Batch [449]	Speed: 11063.488566 samples/sec	accuracy=0.457641	lr=2.398349
Epoch[68] Batch [499]	Speed: 10967.767404 samples/sec	accuracy=0.457348	lr=2.397856
Epoch[68] Batch [549]	Speed: 11281.912975 samples/sec	accuracy=0.457044	lr=2.397363
Epoch[68] Batch [599]	Speed: 10556.561642 samples/sec	accuracy=0.456811	lr=2.396869
[Epoch 68] training: accuracy=0.456639
[Epoch 68] speed: 10796 samples/sec	time cost: 123.266723
[Epoch 68] validation: err-top1=0.650350 err-top5=0.400085
Epoch[69] Batch [49]	Speed: 10527.103554 samples/sec	accuracy=0.461162	lr=2.396118
Epoch[69] Batch [99]	Speed: 11123.812488 samples/sec	accuracy=0.460537	lr=2.395623
Epoch[69] Batch [149]	Speed: 11343.625860 samples/sec	accuracy=0.460469	lr=2.395127
Epoch[69] Batch [199]	Speed: 10522.644489 samples/sec	accuracy=0.460083	lr=2.394631
Epoch[69] Batch [249]	Speed: 11356.082572 samples/sec	accuracy=0.460066	lr=2.394134
Epoch[69] Batch [299]	Speed: 11288.966375 samples/sec	accuracy=0.459139	lr=2.393637
Epoch[69] Batch [349]	Speed: 11155.235639 samples/sec	accuracy=0.458680	lr=2.393139
Epoch[69] Batch [399]	Speed: 11116.443957 samples/sec	accuracy=0.458154	lr=2.392641
Epoch[69] Batch [449]	Speed: 11113.789212 samples/sec	accuracy=0.457974	lr=2.392142
Epoch[69] Batch [499]	Speed: 10905.855913 samples/sec	accuracy=0.457478	lr=2.391642
Epoch[69] Batch [549]	Speed: 10881.250692 samples/sec	accuracy=0.457330	lr=2.391142
Epoch[69] Batch [599]	Speed: 10121.949555 samples/sec	accuracy=0.456980	lr=2.390642
[Epoch 69] training: accuracy=0.456905
[Epoch 69] speed: 10908 samples/sec	time cost: 121.673322
[Epoch 69] validation: err-top1=0.544495 err-top5=0.284465
Epoch[70] Batch [49]	Speed: 10332.224307 samples/sec	accuracy=0.462090	lr=2.389890
Epoch[70] Batch [99]	Speed: 11380.660875 samples/sec	accuracy=0.461045	lr=2.389388
Epoch[70] Batch [149]	Speed: 11141.120646 samples/sec	accuracy=0.459961	lr=2.388885
Epoch[70] Batch [199]	Speed: 10888.791221 samples/sec	accuracy=0.459578	lr=2.388382
Epoch[70] Batch [249]	Speed: 11053.193612 samples/sec	accuracy=0.458475	lr=2.387879
Epoch[70] Batch [299]	Speed: 11385.047864 samples/sec	accuracy=0.457773	lr=2.387374
Epoch[70] Batch [349]	Speed: 10742.302765 samples/sec	accuracy=0.457303	lr=2.386870
Epoch[70] Batch [399]	Speed: 11415.983136 samples/sec	accuracy=0.457115	lr=2.386364
Epoch[70] Batch [449]	Speed: 11314.343398 samples/sec	accuracy=0.457003	lr=2.385859
Epoch[70] Batch [499]	Speed: 11311.291228 samples/sec	accuracy=0.456733	lr=2.385352
Epoch[70] Batch [549]	Speed: 11098.193314 samples/sec	accuracy=0.456642	lr=2.384846
Epoch[70] Batch [599]	Speed: 10920.895843 samples/sec	accuracy=0.456637	lr=2.384338
[Epoch 70] training: accuracy=0.456706
[Epoch 70] speed: 11065 samples/sec	time cost: 120.629044
[Epoch 70] validation: err-top1=0.583086 err-top5=0.330488
Epoch[71] Batch [49]	Speed: 10350.103398 samples/sec	accuracy=0.462939	lr=2.383566
Epoch[71] Batch [99]	Speed: 10666.234458 samples/sec	accuracy=0.462559	lr=2.383057
Epoch[71] Batch [149]	Speed: 11237.753068 samples/sec	accuracy=0.461615	lr=2.382548
Epoch[71] Batch [199]	Speed: 11120.983174 samples/sec	accuracy=0.459839	lr=2.382038
Epoch[71] Batch [249]	Speed: 10840.661447 samples/sec	accuracy=0.459004	lr=2.381528
Epoch[71] Batch [299]	Speed: 10392.018714 samples/sec	accuracy=0.459150	lr=2.381017
Epoch[71] Batch [349]	Speed: 11397.982624 samples/sec	accuracy=0.458859	lr=2.380505
Epoch[71] Batch [399]	Speed: 11032.546017 samples/sec	accuracy=0.458566	lr=2.379993
Epoch[71] Batch [449]	Speed: 11417.536332 samples/sec	accuracy=0.458181	lr=2.379481
Epoch[71] Batch [499]	Speed: 11023.065269 samples/sec	accuracy=0.457861	lr=2.378967
Epoch[71] Batch [549]	Speed: 11314.628348 samples/sec	accuracy=0.457795	lr=2.378454
Epoch[71] Batch [599]	Speed: 10911.145776 samples/sec	accuracy=0.457567	lr=2.377940
[Epoch 71] training: accuracy=0.457628
[Epoch 71] speed: 10950 samples/sec	time cost: 121.698890
[Epoch 71] validation: err-top1=0.528768 err-top5=0.275675
Epoch[72] Batch [49]	Speed: 10389.629547 samples/sec	accuracy=0.465332	lr=2.377157
Epoch[72] Batch [99]	Speed: 10046.192948 samples/sec	accuracy=0.464067	lr=2.376641
Epoch[72] Batch [149]	Speed: 11135.111532 samples/sec	accuracy=0.463001	lr=2.376125
Epoch[72] Batch [199]	Speed: 10274.137844 samples/sec	accuracy=0.461833	lr=2.375609
Epoch[72] Batch [249]	Speed: 11198.742310 samples/sec	accuracy=0.461084	lr=2.375092
Epoch[72] Batch [299]	Speed: 11143.505407 samples/sec	accuracy=0.460034	lr=2.374574
Epoch[72] Batch [349]	Speed: 11025.284561 samples/sec	accuracy=0.459660	lr=2.374056
Epoch[72] Batch [399]	Speed: 11442.876435 samples/sec	accuracy=0.459236	lr=2.373537
Epoch[72] Batch [449]	Speed: 10512.225606 samples/sec	accuracy=0.458707	lr=2.373017
Epoch[72] Batch [499]	Speed: 11203.510902 samples/sec	accuracy=0.458555	lr=2.372498
Epoch[72] Batch [549]	Speed: 11244.691872 samples/sec	accuracy=0.458362	lr=2.371977
Epoch[72] Batch [599]	Speed: 11442.721565 samples/sec	accuracy=0.457909	lr=2.371456
[Epoch 72] training: accuracy=0.457934
[Epoch 72] speed: 10888 samples/sec	time cost: 122.306650
[Epoch 72] validation: err-top1=0.548594 err-top5=0.298184
Epoch[73] Batch [49]	Speed: 11093.565513 samples/sec	accuracy=0.467588	lr=2.370674
Epoch[73] Batch [99]	Speed: 11386.482168 samples/sec	accuracy=0.465127	lr=2.370152
Epoch[73] Batch [149]	Speed: 11116.651120 samples/sec	accuracy=0.463346	lr=2.369629
Epoch[73] Batch [199]	Speed: 10791.399518 samples/sec	accuracy=0.462288	lr=2.369105
Epoch[73] Batch [249]	Speed: 11313.648373 samples/sec	accuracy=0.461697	lr=2.368582
Epoch[73] Batch [299]	Speed: 10398.736589 samples/sec	accuracy=0.461012	lr=2.368057
Epoch[73] Batch [349]	Speed: 11078.456332 samples/sec	accuracy=0.460608	lr=2.367532
Epoch[73] Batch [399]	Speed: 11364.762506 samples/sec	accuracy=0.460577	lr=2.367007
Epoch[73] Batch [449]	Speed: 11195.096161 samples/sec	accuracy=0.460379	lr=2.366481
Epoch[73] Batch [499]	Speed: 11360.362286 samples/sec	accuracy=0.460169	lr=2.365954
Epoch[73] Batch [549]	Speed: 11529.405997 samples/sec	accuracy=0.459836	lr=2.365427
Epoch[73] Batch [599]	Speed: 11084.495692 samples/sec	accuracy=0.459403	lr=2.364899
[Epoch 73] training: accuracy=0.459341
[Epoch 73] speed: 11116 samples/sec	time cost: 119.675720
[Epoch 73] validation: err-top1=0.571960 err-top5=0.318359
Epoch[74] Batch [49]	Speed: 11198.322142 samples/sec	accuracy=0.466377	lr=2.364096
Epoch[74] Batch [99]	Speed: 11624.824817 samples/sec	accuracy=0.465698	lr=2.363567
Epoch[74] Batch [149]	Speed: 11286.777590 samples/sec	accuracy=0.462881	lr=2.363038
Epoch[74] Batch [199]	Speed: 10916.856981 samples/sec	accuracy=0.461855	lr=2.362508
Epoch[74] Batch [249]	Speed: 11453.744284 samples/sec	accuracy=0.461139	lr=2.361977
Epoch[74] Batch [299]	Speed: 11081.531657 samples/sec	accuracy=0.461366	lr=2.361446
Epoch[74] Batch [349]	Speed: 10969.737535 samples/sec	accuracy=0.460674	lr=2.360915
Epoch[74] Batch [399]	Speed: 11153.787453 samples/sec	accuracy=0.460270	lr=2.360382
Epoch[74] Batch [449]	Speed: 11368.392441 samples/sec	accuracy=0.459448	lr=2.359850
Epoch[74] Batch [499]	Speed: 11068.051942 samples/sec	accuracy=0.459045	lr=2.359316
Epoch[74] Batch [549]	Speed: 11570.375658 samples/sec	accuracy=0.458922	lr=2.358783
Epoch[74] Batch [599]	Speed: 11105.948598 samples/sec	accuracy=0.458651	lr=2.358248
[Epoch 74] training: accuracy=0.458526
[Epoch 74] speed: 11155 samples/sec	time cost: 119.390851
[Epoch 74] validation: err-top1=0.573066 err-top5=0.314922
Epoch[75] Batch [49]	Speed: 11355.578758 samples/sec	accuracy=0.466387	lr=2.357446
Epoch[75] Batch [99]	Speed: 11172.964996 samples/sec	accuracy=0.464897	lr=2.356910
Epoch[75] Batch [149]	Speed: 10657.445832 samples/sec	accuracy=0.462793	lr=2.356374
Epoch[75] Batch [199]	Speed: 11586.574223 samples/sec	accuracy=0.462241	lr=2.355838
Epoch[75] Batch [249]	Speed: 11390.792382 samples/sec	accuracy=0.461375	lr=2.355300
Epoch[75] Batch [299]	Speed: 11472.961259 samples/sec	accuracy=0.460884	lr=2.354763
Epoch[75] Batch [349]	Speed: 9982.125691 samples/sec	accuracy=0.460392	lr=2.354224
Epoch[75] Batch [399]	Speed: 11148.980630 samples/sec	accuracy=0.460204	lr=2.353686
Epoch[75] Batch [449]	Speed: 11105.143983 samples/sec	accuracy=0.459876	lr=2.353146
Epoch[75] Batch [499]	Speed: 10502.675050 samples/sec	accuracy=0.459741	lr=2.352606
Epoch[75] Batch [549]	Speed: 11079.519740 samples/sec	accuracy=0.459869	lr=2.352066
Epoch[75] Batch [599]	Speed: 10414.828559 samples/sec	accuracy=0.459427	lr=2.351525
[Epoch 75] training: accuracy=0.459169
[Epoch 75] speed: 10912 samples/sec	time cost: 122.143978
[Epoch 75] validation: err-top1=0.563578 err-top5=0.302592
Epoch[76] Batch [49]	Speed: 10970.209094 samples/sec	accuracy=0.467598	lr=2.350702
Epoch[76] Batch [99]	Speed: 11056.963661 samples/sec	accuracy=0.465381	lr=2.350160
Epoch[76] Batch [149]	Speed: 11368.326843 samples/sec	accuracy=0.463620	lr=2.349617
Epoch[76] Batch [199]	Speed: 11244.170811 samples/sec	accuracy=0.462275	lr=2.349074
Epoch[76] Batch [249]	Speed: 11223.617787 samples/sec	accuracy=0.461635	lr=2.348530
Epoch[76] Batch [299]	Speed: 11373.416567 samples/sec	accuracy=0.461077	lr=2.347986
Epoch[76] Batch [349]	Speed: 10821.547930 samples/sec	accuracy=0.460891	lr=2.347441
Epoch[76] Batch [399]	Speed: 10054.532256 samples/sec	accuracy=0.460488	lr=2.346895
Epoch[76] Batch [449]	Speed: 10877.003919 samples/sec	accuracy=0.460571	lr=2.346349
Epoch[76] Batch [499]	Speed: 10293.044109 samples/sec	accuracy=0.460162	lr=2.345803
Epoch[76] Batch [549]	Speed: 10688.973141 samples/sec	accuracy=0.459767	lr=2.345256
Epoch[76] Batch [599]	Speed: 11566.520295 samples/sec	accuracy=0.459713	lr=2.344709
[Epoch 76] training: accuracy=0.459384
[Epoch 76] speed: 10923 samples/sec	time cost: 121.841300
[Epoch 76] validation: err-top1=0.510763 err-top5=0.254781
Epoch[77] Batch [49]	Speed: 10889.183789 samples/sec	accuracy=0.467480	lr=2.343886
Epoch[77] Batch [99]	Speed: 10724.642242 samples/sec	accuracy=0.466445	lr=2.343338
Epoch[77] Batch [149]	Speed: 11005.274798 samples/sec	accuracy=0.464788	lr=2.342788
Epoch[77] Batch [199]	Speed: 11040.477697 samples/sec	accuracy=0.463169	lr=2.342238
Epoch[77] Batch [249]	Speed: 11152.865549 samples/sec	accuracy=0.462449	lr=2.341688
Epoch[77] Batch [299]	Speed: 10992.475551 samples/sec	accuracy=0.461917	lr=2.341137
Epoch[77] Batch [349]	Speed: 11585.026260 samples/sec	accuracy=0.461862	lr=2.340586
Epoch[77] Batch [399]	Speed: 11122.521073 samples/sec	accuracy=0.461310	lr=2.340034
Epoch[77] Batch [449]	Speed: 11505.901115 samples/sec	accuracy=0.460909	lr=2.339481
Epoch[77] Batch [499]	Speed: 11499.429980 samples/sec	accuracy=0.460719	lr=2.338928
Epoch[77] Batch [549]	Speed: 11227.425178 samples/sec	accuracy=0.460252	lr=2.338375
Epoch[77] Batch [599]	Speed: 11192.214725 samples/sec	accuracy=0.460365	lr=2.337821
[Epoch 77] training: accuracy=0.460247
[Epoch 77] speed: 11112 samples/sec	time cost: 120.293800
[Epoch 77] validation: err-top1=0.533262 err-top5=0.271465
Epoch[78] Batch [49]	Speed: 11333.267020 samples/sec	accuracy=0.466279	lr=2.336978
Epoch[78] Batch [99]	Speed: 11095.417718 samples/sec	accuracy=0.466079	lr=2.336423
Epoch[78] Batch [149]	Speed: 11476.094875 samples/sec	accuracy=0.464391	lr=2.335867
Epoch[78] Batch [199]	Speed: 10686.614874 samples/sec	accuracy=0.463411	lr=2.335310
Epoch[78] Batch [249]	Speed: 11022.945317 samples/sec	accuracy=0.463254	lr=2.334753
Epoch[78] Batch [299]	Speed: 10705.234558 samples/sec	accuracy=0.462383	lr=2.334196
Epoch[78] Batch [349]	Speed: 10844.172339 samples/sec	accuracy=0.461671	lr=2.333638
Epoch[78] Batch [399]	Speed: 11431.915692 samples/sec	accuracy=0.461373	lr=2.333080
Epoch[78] Batch [449]	Speed: 11019.005058 samples/sec	accuracy=0.460873	lr=2.332521
Epoch[78] Batch [499]	Speed: 10773.063074 samples/sec	accuracy=0.460777	lr=2.331961
Epoch[78] Batch [549]	Speed: 11051.980255 samples/sec	accuracy=0.460371	lr=2.331401
Epoch[78] Batch [599]	Speed: 11476.966717 samples/sec	accuracy=0.460288	lr=2.330841
[Epoch 78] training: accuracy=0.460243
[Epoch 78] speed: 11048 samples/sec	time cost: 120.659547
[Epoch 78] validation: err-top1=0.550008 err-top5=0.298869
Epoch[79] Batch [49]	Speed: 10460.440167 samples/sec	accuracy=0.466328	lr=2.329988
Epoch[79] Batch [99]	Speed: 11117.062880 samples/sec	accuracy=0.464751	lr=2.329426
Epoch[79] Batch [149]	Speed: 10732.849266 samples/sec	accuracy=0.464567	lr=2.328864
Epoch[79] Batch [199]	Speed: 11361.960797 samples/sec	accuracy=0.464360	lr=2.328301
Epoch[79] Batch [249]	Speed: 11339.616213 samples/sec	accuracy=0.463369	lr=2.327738
Epoch[79] Batch [299]	Speed: 11318.777835 samples/sec	accuracy=0.463226	lr=2.327174
Epoch[79] Batch [349]	Speed: 10909.185825 samples/sec	accuracy=0.462246	lr=2.326609
Epoch[79] Batch [399]	Speed: 10134.585957 samples/sec	accuracy=0.461780	lr=2.326044
Epoch[79] Batch [449]	Speed: 11274.915445 samples/sec	accuracy=0.461514	lr=2.325479
Epoch[79] Batch [499]	Speed: 11429.344468 samples/sec	accuracy=0.461190	lr=2.324913
Epoch[79] Batch [549]	Speed: 10936.319823 samples/sec	accuracy=0.461288	lr=2.324347
Epoch[79] Batch [599]	Speed: 11623.766152 samples/sec	accuracy=0.461216	lr=2.323780
[Epoch 79] training: accuracy=0.461270
[Epoch 79] speed: 11031 samples/sec	time cost: 120.912793
[Epoch 79] validation: err-top1=0.652305 err-top5=0.404375
Epoch[80] Batch [49]	Speed: 10686.108622 samples/sec	accuracy=0.470615	lr=2.322928
Epoch[80] Batch [99]	Speed: 11339.780581 samples/sec	accuracy=0.464873	lr=2.322360
Epoch[80] Batch [149]	Speed: 11241.570056 samples/sec	accuracy=0.464473	lr=2.321791
Epoch[80] Batch [199]	Speed: 11163.795113 samples/sec	accuracy=0.463696	lr=2.321222
Epoch[80] Batch [249]	Speed: 10830.666488 samples/sec	accuracy=0.463309	lr=2.320652
Epoch[80] Batch [299]	Speed: 11317.709459 samples/sec	accuracy=0.463252	lr=2.320082
Epoch[80] Batch [349]	Speed: 11220.224500 samples/sec	accuracy=0.462648	lr=2.319511
Epoch[80] Batch [399]	Speed: 11225.903906 samples/sec	accuracy=0.462427	lr=2.318940
Epoch[80] Batch [449]	Speed: 10947.842618 samples/sec	accuracy=0.462219	lr=2.318368
Epoch[80] Batch [499]	Speed: 11054.834031 samples/sec	accuracy=0.461704	lr=2.317796
Epoch[80] Batch [549]	Speed: 11296.169423 samples/sec	accuracy=0.461608	lr=2.317223
Epoch[80] Batch [599]	Speed: 10793.247663 samples/sec	accuracy=0.461185	lr=2.316650
[Epoch 80] training: accuracy=0.461179
[Epoch 80] speed: 11056 samples/sec	time cost: 120.579921
[Epoch 80] validation: err-top1=0.518860 err-top5=0.264547
Epoch[81] Batch [49]	Speed: 10297.824205 samples/sec	accuracy=0.470107	lr=2.315777
Epoch[81] Batch [99]	Speed: 10746.552810 samples/sec	accuracy=0.466182	lr=2.315203
Epoch[81] Batch [149]	Speed: 10716.389736 samples/sec	accuracy=0.464720	lr=2.314628
Epoch[81] Batch [199]	Speed: 10142.241824 samples/sec	accuracy=0.464297	lr=2.314052
Epoch[81] Batch [249]	Speed: 11425.142709 samples/sec	accuracy=0.464439	lr=2.313476
Epoch[81] Batch [299]	Speed: 11398.364669 samples/sec	accuracy=0.464111	lr=2.312899
Epoch[81] Batch [349]	Speed: 10696.482001 samples/sec	accuracy=0.463551	lr=2.312322
Epoch[81] Batch [399]	Speed: 11174.716755 samples/sec	accuracy=0.462786	lr=2.311744
Epoch[81] Batch [449]	Speed: 11280.274691 samples/sec	accuracy=0.462017	lr=2.311166
Epoch[81] Batch [499]	Speed: 11108.846120 samples/sec	accuracy=0.461729	lr=2.310587
Epoch[81] Batch [549]	Speed: 10896.874382 samples/sec	accuracy=0.461395	lr=2.310008
Epoch[81] Batch [599]	Speed: 11036.064350 samples/sec	accuracy=0.461302	lr=2.309428
[Epoch 81] training: accuracy=0.461230
[Epoch 81] speed: 10839 samples/sec	time cost: 122.876086
[Epoch 81] validation: err-top1=0.585063 err-top5=0.330383
Epoch[82] Batch [49]	Speed: 10981.451065 samples/sec	accuracy=0.468115	lr=2.308558
Epoch[82] Batch [99]	Speed: 11164.825919 samples/sec	accuracy=0.467173	lr=2.307977
Epoch[82] Batch [149]	Speed: 11333.267020 samples/sec	accuracy=0.466699	lr=2.307396
Epoch[82] Batch [199]	Speed: 10870.509145 samples/sec	accuracy=0.465366	lr=2.306814
Epoch[82] Batch [249]	Speed: 10680.310619 samples/sec	accuracy=0.465465	lr=2.306231
Epoch[82] Batch [299]	Speed: 10635.834850 samples/sec	accuracy=0.464626	lr=2.305648
Epoch[82] Batch [349]	Speed: 11234.066490 samples/sec	accuracy=0.463916	lr=2.305065
Epoch[82] Batch [399]	Speed: 10548.287446 samples/sec	accuracy=0.463561	lr=2.304481
Epoch[82] Batch [449]	Speed: 11122.938741 samples/sec	accuracy=0.462925	lr=2.303896
Epoch[82] Batch [499]	Speed: 10461.105911 samples/sec	accuracy=0.462531	lr=2.303311
Epoch[82] Batch [549]	Speed: 11295.610902 samples/sec	accuracy=0.462380	lr=2.302726
Epoch[82] Batch [599]	Speed: 11003.551794 samples/sec	accuracy=0.461848	lr=2.302140
[Epoch 82] training: accuracy=0.461884
[Epoch 82] speed: 10909 samples/sec	time cost: 122.413621
[Epoch 82] validation: err-top1=0.529883 err-top5=0.271055
Epoch[83] Batch [49]	Speed: 10810.575567 samples/sec	accuracy=0.468398	lr=2.301248
Epoch[83] Batch [99]	Speed: 11343.056646 samples/sec	accuracy=0.466387	lr=2.300661
Epoch[83] Batch [149]	Speed: 11364.582078 samples/sec	accuracy=0.466423	lr=2.300073
Epoch[83] Batch [199]	Speed: 10758.879245 samples/sec	accuracy=0.464561	lr=2.299485
Epoch[83] Batch [249]	Speed: 10999.703136 samples/sec	accuracy=0.464715	lr=2.298896
Epoch[83] Batch [299]	Speed: 11020.169620 samples/sec	accuracy=0.463936	lr=2.298307
Epoch[83] Batch [349]	Speed: 10581.420067 samples/sec	accuracy=0.463610	lr=2.297717
Epoch[83] Batch [399]	Speed: 10379.057662 samples/sec	accuracy=0.463495	lr=2.297127
Epoch[83] Batch [449]	Speed: 9977.808201 samples/sec	accuracy=0.463262	lr=2.296536
Epoch[83] Batch [499]	Speed: 10609.217779 samples/sec	accuracy=0.462945	lr=2.295945
Epoch[83] Batch [549]	Speed: 10116.273961 samples/sec	accuracy=0.462763	lr=2.295353
Epoch[83] Batch [599]	Speed: 10371.530856 samples/sec	accuracy=0.462621	lr=2.294760
[Epoch 83] training: accuracy=0.462565
[Epoch 83] speed: 10649 samples/sec	time cost: 124.857161
[Epoch 83] validation: err-top1=0.551432 err-top5=0.293355
Epoch[84] Batch [49]	Speed: 11139.431119 samples/sec	accuracy=0.470127	lr=2.293871
Epoch[84] Batch [99]	Speed: 10942.199312 samples/sec	accuracy=0.468247	lr=2.293278
Epoch[84] Batch [149]	Speed: 11564.692754 samples/sec	accuracy=0.466510	lr=2.292684
Epoch[84] Batch [199]	Speed: 11098.953613 samples/sec	accuracy=0.466125	lr=2.292089
Epoch[84] Batch [249]	Speed: 10976.960210 samples/sec	accuracy=0.465314	lr=2.291494
Epoch[84] Batch [299]	Speed: 10429.639829 samples/sec	accuracy=0.465002	lr=2.290899
Epoch[84] Batch [349]	Speed: 11563.850187 samples/sec	accuracy=0.464736	lr=2.290303
Epoch[84] Batch [399]	Speed: 11397.093403 samples/sec	accuracy=0.464550	lr=2.289706
Epoch[84] Batch [449]	Speed: 11261.953582 samples/sec	accuracy=0.464149	lr=2.289109
Epoch[84] Batch [499]	Speed: 10871.436966 samples/sec	accuracy=0.463719	lr=2.288512
Epoch[84] Batch [549]	Speed: 11551.735294 samples/sec	accuracy=0.463485	lr=2.287914
Epoch[84] Batch [599]	Speed: 11137.134161 samples/sec	accuracy=0.463373	lr=2.287315
[Epoch 84] training: accuracy=0.463195
[Epoch 84] speed: 11120 samples/sec	time cost: 119.883650
[Epoch 84] validation: err-top1=0.583184 err-top5=0.321797
Epoch[85] Batch [49]	Speed: 11419.751535 samples/sec	accuracy=0.470391	lr=2.286405
Epoch[85] Batch [99]	Speed: 11108.177261 samples/sec	accuracy=0.468262	lr=2.285805
Epoch[85] Batch [149]	Speed: 11140.172807 samples/sec	accuracy=0.466738	lr=2.285205
Epoch[85] Batch [199]	Speed: 11199.615450 samples/sec	accuracy=0.466863	lr=2.284604
Epoch[85] Batch [249]	Speed: 11118.851842 samples/sec	accuracy=0.465398	lr=2.284003
Epoch[85] Batch [299]	Speed: 11189.595967 samples/sec	accuracy=0.465052	lr=2.283401
Epoch[85] Batch [349]	Speed: 11439.831319 samples/sec	accuracy=0.464399	lr=2.282799
Epoch[85] Batch [399]	Speed: 11221.281878 samples/sec	accuracy=0.464362	lr=2.282196
Epoch[85] Batch [449]	Speed: 11616.058031 samples/sec	accuracy=0.463981	lr=2.281593
Epoch[85] Batch [499]	Speed: 11051.988502 samples/sec	accuracy=0.463812	lr=2.280989
Epoch[85] Batch [549]	Speed: 11235.458299 samples/sec	accuracy=0.463605	lr=2.280385
Epoch[85] Batch [599]	Speed: 11057.381543 samples/sec	accuracy=0.463215	lr=2.279780
[Epoch 85] training: accuracy=0.462980
[Epoch 85] speed: 11197 samples/sec	time cost: 119.004559
[Epoch 85] validation: err-top1=0.554382 err-top5=0.296916
Epoch[86] Batch [49]	Speed: 10599.268435 samples/sec	accuracy=0.468877	lr=2.278860
Epoch[86] Batch [99]	Speed: 10441.648115 samples/sec	accuracy=0.467192	lr=2.278254
Epoch[86] Batch [149]	Speed: 10792.829436 samples/sec	accuracy=0.466833	lr=2.277648
Epoch[86] Batch [199]	Speed: 10826.080865 samples/sec	accuracy=0.466597	lr=2.277041
Epoch[86] Batch [249]	Speed: 11193.728045 samples/sec	accuracy=0.466682	lr=2.276434
Epoch[86] Batch [299]	Speed: 11023.531520 samples/sec	accuracy=0.466328	lr=2.275826
Epoch[86] Batch [349]	Speed: 11101.569422 samples/sec	accuracy=0.465997	lr=2.275218
Epoch[86] Batch [399]	Speed: 11092.252177 samples/sec	accuracy=0.465316	lr=2.274609
Epoch[86] Batch [449]	Speed: 10418.543606 samples/sec	accuracy=0.464833	lr=2.274000
Epoch[86] Batch [499]	Speed: 10042.311999 samples/sec	accuracy=0.464443	lr=2.273390
Epoch[86] Batch [549]	Speed: 10735.884082 samples/sec	accuracy=0.464076	lr=2.272779
Epoch[86] Batch [599]	Speed: 11323.679674 samples/sec	accuracy=0.463494	lr=2.272169
[Epoch 86] training: accuracy=0.463518
[Epoch 86] speed: 10763 samples/sec	time cost: 123.563813
[Epoch 86] validation: err-top1=0.561738 err-top5=0.302188
Epoch[87] Batch [49]	Speed: 11093.899626 samples/sec	accuracy=0.469805	lr=2.271252
Epoch[87] Batch [99]	Speed: 10760.442628 samples/sec	accuracy=0.468599	lr=2.270640
Epoch[87] Batch [149]	Speed: 11265.043297 samples/sec	accuracy=0.467497	lr=2.270027
Epoch[87] Batch [199]	Speed: 10695.137953 samples/sec	accuracy=0.466340	lr=2.269414
Epoch[87] Batch [249]	Speed: 10800.130548 samples/sec	accuracy=0.466127	lr=2.268801
Epoch[87] Batch [299]	Speed: 11502.844548 samples/sec	accuracy=0.465819	lr=2.268187
Epoch[87] Batch [349]	Speed: 11632.724482 samples/sec	accuracy=0.464941	lr=2.267572
Epoch[87] Batch [399]	Speed: 10901.173765 samples/sec	accuracy=0.464061	lr=2.266957
Epoch[87] Batch [449]	Speed: 10940.028386 samples/sec	accuracy=0.463968	lr=2.266342
Epoch[87] Batch [499]	Speed: 10560.760998 samples/sec	accuracy=0.463794	lr=2.265726
Epoch[87] Batch [549]	Speed: 11216.199962 samples/sec	accuracy=0.463663	lr=2.265110
Epoch[87] Batch [599]	Speed: 10886.271124 samples/sec	accuracy=0.463386	lr=2.264493
[Epoch 87] training: accuracy=0.463060
[Epoch 87] speed: 10959 samples/sec	time cost: 121.434902
[Epoch 87] validation: err-top1=0.609436 err-top5=0.356181
Epoch[88] Batch [49]	Speed: 10925.398183 samples/sec	accuracy=0.472969	lr=2.263554
Epoch[88] Batch [99]	Speed: 10666.548889 samples/sec	accuracy=0.471987	lr=2.262936
Epoch[88] Batch [149]	Speed: 10550.778658 samples/sec	accuracy=0.469577	lr=2.262318
Epoch[88] Batch [199]	Speed: 10908.980227 samples/sec	accuracy=0.468889	lr=2.261699
Epoch[88] Batch [249]	Speed: 10996.316618 samples/sec	accuracy=0.467744	lr=2.261079
Epoch[88] Batch [299]	Speed: 10418.976042 samples/sec	accuracy=0.466585	lr=2.260459
Epoch[88] Batch [349]	Speed: 10950.508287 samples/sec	accuracy=0.466166	lr=2.259839
Epoch[88] Batch [399]	Speed: 11061.747288 samples/sec	accuracy=0.465708	lr=2.259218
Epoch[88] Batch [449]	Speed: 11159.192764 samples/sec	accuracy=0.465603	lr=2.258596
Epoch[88] Batch [499]	Speed: 10575.329894 samples/sec	accuracy=0.464991	lr=2.257974
Epoch[88] Batch [549]	Speed: 10790.347593 samples/sec	accuracy=0.464849	lr=2.257352
Epoch[88] Batch [599]	Speed: 11118.705906 samples/sec	accuracy=0.464814	lr=2.256729
[Epoch 88] training: accuracy=0.464727
[Epoch 88] speed: 10825 samples/sec	time cost: 122.647979
[Epoch 88] validation: err-top1=0.549072 err-top5=0.297994
Epoch[89] Batch [49]	Speed: 10493.916099 samples/sec	accuracy=0.476758	lr=2.255794
Epoch[89] Batch [99]	Speed: 10942.873982 samples/sec	accuracy=0.472471	lr=2.255170
Epoch[89] Batch [149]	Speed: 11416.613407 samples/sec	accuracy=0.469840	lr=2.254545
Epoch[89] Batch [199]	Speed: 11424.827854 samples/sec	accuracy=0.468042	lr=2.253920
Epoch[89] Batch [249]	Speed: 10943.383664 samples/sec	accuracy=0.467471	lr=2.253295
Epoch[89] Batch [299]	Speed: 11527.921230 samples/sec	accuracy=0.466903	lr=2.252669
Epoch[89] Batch [349]	Speed: 11210.336662 samples/sec	accuracy=0.466274	lr=2.252042
Epoch[89] Batch [399]	Speed: 10356.080183 samples/sec	accuracy=0.465781	lr=2.251415
Epoch[89] Batch [449]	Speed: 11560.201414 samples/sec	accuracy=0.465549	lr=2.250788
Epoch[89] Batch [499]	Speed: 11382.090151 samples/sec	accuracy=0.465185	lr=2.250160
Epoch[89] Batch [549]	Speed: 10514.882065 samples/sec	accuracy=0.464850	lr=2.249532
Epoch[89] Batch [599]	Speed: 11422.563290 samples/sec	accuracy=0.464668	lr=2.248903
[Epoch 89] training: accuracy=0.464543
[Epoch 89] speed: 11062 samples/sec	time cost: 120.471966
[Epoch 89] validation: err-top1=0.530430 err-top5=0.275918
Epoch[90] Batch [49]	Speed: 10123.746110 samples/sec	accuracy=0.472305	lr=2.247946
Epoch[90] Batch [99]	Speed: 10867.366404 samples/sec	accuracy=0.470400	lr=2.247316
Epoch[90] Batch [149]	Speed: 10134.675396 samples/sec	accuracy=0.469463	lr=2.246685
Epoch[90] Batch [199]	Speed: 11413.948355 samples/sec	accuracy=0.468032	lr=2.246054
Epoch[90] Batch [249]	Speed: 11387.163226 samples/sec	accuracy=0.467316	lr=2.245423
Epoch[90] Batch [299]	Speed: 11237.971246 samples/sec	accuracy=0.467048	lr=2.244791
Epoch[90] Batch [349]	Speed: 11243.957396 samples/sec	accuracy=0.466982	lr=2.244159
Epoch[90] Batch [399]	Speed: 11488.101636 samples/sec	accuracy=0.466487	lr=2.243526
Epoch[90] Batch [449]	Speed: 11162.446821 samples/sec	accuracy=0.466025	lr=2.242892
Epoch[90] Batch [499]	Speed: 10703.982211 samples/sec	accuracy=0.465760	lr=2.242258
Epoch[90] Batch [549]	Speed: 11043.232969 samples/sec	accuracy=0.465379	lr=2.241624
Epoch[90] Batch [599]	Speed: 11298.518783 samples/sec	accuracy=0.465084	lr=2.240989
[Epoch 90] training: accuracy=0.465202
[Epoch 90] speed: 10954 samples/sec	time cost: 121.280153
[Epoch 90] validation: err-top1=0.547791 err-top5=0.290446
Epoch[91] Batch [49]	Speed: 10728.509822 samples/sec	accuracy=0.474961	lr=2.240036
Epoch[91] Batch [99]	Speed: 10629.473545 samples/sec	accuracy=0.473164	lr=2.239400
Epoch[91] Batch [149]	Speed: 11180.706707 samples/sec	accuracy=0.471517	lr=2.238764
Epoch[91] Batch [199]	Speed: 10527.165222 samples/sec	accuracy=0.469475	lr=2.238127
Epoch[91] Batch [249]	Speed: 10626.133664 samples/sec	accuracy=0.468451	lr=2.237490
Epoch[91] Batch [299]	Speed: 11406.434420 samples/sec	accuracy=0.468258	lr=2.236852
Epoch[91] Batch [349]	Speed: 10874.677880 samples/sec	accuracy=0.467708	lr=2.236213
Epoch[91] Batch [399]	Speed: 11683.713446 samples/sec	accuracy=0.466799	lr=2.235575
Epoch[91] Batch [449]	Speed: 11208.313098 samples/sec	accuracy=0.466328	lr=2.234935
Epoch[91] Batch [499]	Speed: 11457.472249 samples/sec	accuracy=0.465834	lr=2.234296
Epoch[91] Batch [549]	Speed: 11093.411644 samples/sec	accuracy=0.465488	lr=2.233656
Epoch[91] Batch [599]	Speed: 10971.344306 samples/sec	accuracy=0.465325	lr=2.233015
[Epoch 91] training: accuracy=0.464978
[Epoch 91] speed: 10973 samples/sec	time cost: 121.341106
[Epoch 91] validation: err-top1=0.527832 err-top5=0.268770
Epoch[92] Batch [49]	Speed: 10220.705370 samples/sec	accuracy=0.474678	lr=2.232040
Epoch[92] Batch [99]	Speed: 10589.173778 samples/sec	accuracy=0.471187	lr=2.231398
Epoch[92] Batch [149]	Speed: 11157.241240 samples/sec	accuracy=0.470000	lr=2.230756
Epoch[92] Batch [199]	Speed: 11216.370437 samples/sec	accuracy=0.469631	lr=2.230113
Epoch[92] Batch [249]	Speed: 11577.806026 samples/sec	accuracy=0.469021	lr=2.229470
Epoch[92] Batch [299]	Speed: 11386.475828 samples/sec	accuracy=0.468683	lr=2.228826
Epoch[92] Batch [349]	Speed: 11123.272609 samples/sec	accuracy=0.468205	lr=2.228182
Epoch[92] Batch [399]	Speed: 11473.535924 samples/sec	accuracy=0.467922	lr=2.227538
Epoch[92] Batch [449]	Speed: 11103.012690 samples/sec	accuracy=0.467493	lr=2.226892
Epoch[92] Batch [499]	Speed: 10700.844626 samples/sec	accuracy=0.467445	lr=2.226247
Epoch[92] Batch [549]	Speed: 11464.241094 samples/sec	accuracy=0.467192	lr=2.225601
Epoch[92] Batch [599]	Speed: 10451.315058 samples/sec	accuracy=0.466902	lr=2.224954
[Epoch 92] training: accuracy=0.466902
[Epoch 92] speed: 11015 samples/sec	time cost: 120.609942
[Epoch 92] validation: err-top1=0.517741 err-top5=0.263692
Epoch[93] Batch [49]	Speed: 11233.517032 samples/sec	accuracy=0.475303	lr=2.223984
Epoch[93] Batch [99]	Speed: 11719.685196 samples/sec	accuracy=0.472461	lr=2.223336
Epoch[93] Batch [149]	Speed: 11049.095547 samples/sec	accuracy=0.471309	lr=2.222688
Epoch[93] Batch [199]	Speed: 11371.536026 samples/sec	accuracy=0.470332	lr=2.222039
Epoch[93] Batch [249]	Speed: 11272.274416 samples/sec	accuracy=0.469369	lr=2.221390
Epoch[93] Batch [299]	Speed: 10970.474451 samples/sec	accuracy=0.469046	lr=2.220741
Epoch[93] Batch [349]	Speed: 11363.307213 samples/sec	accuracy=0.468453	lr=2.220091
Epoch[93] Batch [399]	Speed: 10805.206051 samples/sec	accuracy=0.468225	lr=2.219441
Epoch[93] Batch [449]	Speed: 11001.043394 samples/sec	accuracy=0.467947	lr=2.218790
Epoch[93] Batch [499]	Speed: 10702.683483 samples/sec	accuracy=0.467703	lr=2.218138
Epoch[93] Batch [549]	Speed: 11051.427989 samples/sec	accuracy=0.467172	lr=2.217487
Epoch[93] Batch [599]	Speed: 11717.944489 samples/sec	accuracy=0.466799	lr=2.216834
[Epoch 93] training: accuracy=0.466517
[Epoch 93] speed: 11103 samples/sec	time cost: 119.925048
[Epoch 93] validation: err-top1=0.536702 err-top5=0.278422
Epoch[94] Batch [49]	Speed: 10648.819793 samples/sec	accuracy=0.474961	lr=2.215842
Epoch[94] Batch [99]	Speed: 10052.470306 samples/sec	accuracy=0.473301	lr=2.215189
Epoch[94] Batch [149]	Speed: 11028.607105 samples/sec	accuracy=0.471908	lr=2.214535
Epoch[94] Batch [199]	Speed: 11015.050947 samples/sec	accuracy=0.471594	lr=2.213880
Epoch[94] Batch [249]	Speed: 10973.965075 samples/sec	accuracy=0.470793	lr=2.213226
Epoch[94] Batch [299]	Speed: 11144.081371 samples/sec	accuracy=0.469775	lr=2.212570
Epoch[94] Batch [349]	Speed: 11249.117831 samples/sec	accuracy=0.468877	lr=2.211915
Epoch[94] Batch [399]	Speed: 11136.047539 samples/sec	accuracy=0.468375	lr=2.211258
Epoch[94] Batch [449]	Speed: 10804.338693 samples/sec	accuracy=0.467801	lr=2.210602
Epoch[94] Batch [499]	Speed: 10766.109653 samples/sec	accuracy=0.467273	lr=2.209945
Epoch[94] Batch [549]	Speed: 11055.483392 samples/sec	accuracy=0.467002	lr=2.209287
Epoch[94] Batch [599]	Speed: 11078.614358 samples/sec	accuracy=0.466800	lr=2.208629
[Epoch 94] training: accuracy=0.466596
[Epoch 94] speed: 10880 samples/sec	time cost: 122.611761
[Epoch 94] validation: err-top1=0.543496 err-top5=0.289570
Epoch[95] Batch [49]	Speed: 10907.593884 samples/sec	accuracy=0.475029	lr=2.207628
Epoch[95] Batch [99]	Speed: 11124.558148 samples/sec	accuracy=0.473032	lr=2.206969
Epoch[95] Batch [149]	Speed: 10651.508234 samples/sec	accuracy=0.472526	lr=2.206309
Epoch[95] Batch [199]	Speed: 10598.004930 samples/sec	accuracy=0.470579	lr=2.205649
Epoch[95] Batch [249]	Speed: 11528.580632 samples/sec	accuracy=0.470291	lr=2.204989
Epoch[95] Batch [299]	Speed: 11322.371286 samples/sec	accuracy=0.469590	lr=2.204328
Epoch[95] Batch [349]	Speed: 11150.301645 samples/sec	accuracy=0.468934	lr=2.203667
Epoch[95] Batch [399]	Speed: 11361.160133 samples/sec	accuracy=0.468604	lr=2.203005
Epoch[95] Batch [449]	Speed: 11095.442082 samples/sec	accuracy=0.468545	lr=2.202342
Epoch[95] Batch [499]	Speed: 10466.376251 samples/sec	accuracy=0.468125	lr=2.201680
Epoch[95] Batch [549]	Speed: 11620.906676 samples/sec	accuracy=0.467847	lr=2.201016
Epoch[95] Batch [599]	Speed: 11473.295630 samples/sec	accuracy=0.467935	lr=2.200353
[Epoch 95] training: accuracy=0.467577
[Epoch 95] speed: 11068 samples/sec	time cost: 120.310729
[Epoch 95] validation: err-top1=0.740804 err-top5=0.518555
Epoch[96] Batch [49]	Speed: 10586.268295 samples/sec	accuracy=0.462139	lr=2.199356
Epoch[96] Batch [99]	Speed: 11315.269536 samples/sec	accuracy=0.464023	lr=2.198692
Epoch[96] Batch [149]	Speed: 10957.363200 samples/sec	accuracy=0.464860	lr=2.198026
Epoch[96] Batch [199]	Speed: 11072.794928 samples/sec	accuracy=0.465801	lr=2.197361
Epoch[96] Batch [249]	Speed: 10794.346002 samples/sec	accuracy=0.465256	lr=2.196694
Epoch[96] Batch [299]	Speed: 10882.317936 samples/sec	accuracy=0.464938	lr=2.196028
Epoch[96] Batch [349]	Speed: 10969.139390 samples/sec	accuracy=0.465188	lr=2.195361
Epoch[96] Batch [399]	Speed: 10824.209457 samples/sec	accuracy=0.464744	lr=2.194693
Epoch[96] Batch [449]	Speed: 11611.644121 samples/sec	accuracy=0.464738	lr=2.194025
Epoch[96] Batch [499]	Speed: 11161.610794 samples/sec	accuracy=0.464401	lr=2.193357
Epoch[96] Batch [549]	Speed: 11559.199596 samples/sec	accuracy=0.464534	lr=2.192688
Epoch[96] Batch [599]	Speed: 11245.797152 samples/sec	accuracy=0.464268	lr=2.192019
[Epoch 96] training: accuracy=0.464171
[Epoch 96] speed: 11042 samples/sec	time cost: 120.675472
[Epoch 96] validation: err-top1=0.545859 err-top5=0.289355
Epoch[97] Batch [49]	Speed: 10698.871277 samples/sec	accuracy=0.475303	lr=2.191001
Epoch[97] Batch [99]	Speed: 11189.236826 samples/sec	accuracy=0.473276	lr=2.190330
Epoch[97] Batch [149]	Speed: 10860.239128 samples/sec	accuracy=0.472087	lr=2.189659
Epoch[97] Batch [199]	Speed: 11590.045438 samples/sec	accuracy=0.470701	lr=2.188988
Epoch[97] Batch [249]	Speed: 11491.398177 samples/sec	accuracy=0.469936	lr=2.188316
Epoch[97] Batch [299]	Speed: 11176.054926 samples/sec	accuracy=0.469137	lr=2.187644
Epoch[97] Batch [349]	Speed: 10631.221638 samples/sec	accuracy=0.468722	lr=2.186972
Epoch[97] Batch [399]	Speed: 11436.943446 samples/sec	accuracy=0.468348	lr=2.186299
Epoch[97] Batch [449]	Speed: 11297.061387 samples/sec	accuracy=0.467957	lr=2.185625
Epoch[97] Batch [499]	Speed: 11769.761693 samples/sec	accuracy=0.467816	lr=2.184951
Epoch[97] Batch [549]	Speed: 11360.081338 samples/sec	accuracy=0.467643	lr=2.184277
Epoch[97] Batch [599]	Speed: 11025.499662 samples/sec	accuracy=0.467508	lr=2.183602
[Epoch 97] training: accuracy=0.467373
[Epoch 97] speed: 11182 samples/sec	time cost: 118.772985
[Epoch 97] validation: err-top1=0.534749 err-top5=0.281230
Epoch[98] Batch [49]	Speed: 11241.814277 samples/sec	accuracy=0.476904	lr=2.182589
Epoch[98] Batch [99]	Speed: 11003.947888 samples/sec	accuracy=0.473369	lr=2.181913
Epoch[98] Batch [149]	Speed: 11579.338012 samples/sec	accuracy=0.471624	lr=2.181236
Epoch[98] Batch [199]	Speed: 11479.549290 samples/sec	accuracy=0.470442	lr=2.180559
Epoch[98] Batch [249]	Speed: 10198.352936 samples/sec	accuracy=0.469809	lr=2.179882
Epoch[98] Batch [299]	Speed: 10764.409728 samples/sec	accuracy=0.469657	lr=2.179204
Epoch[98] Batch [349]	Speed: 11065.473283 samples/sec	accuracy=0.469305	lr=2.178526
Epoch[98] Batch [399]	Speed: 11334.886632 samples/sec	accuracy=0.469034	lr=2.177848
Epoch[98] Batch [449]	Speed: 11205.294470 samples/sec	accuracy=0.468728	lr=2.177169
Epoch[98] Batch [499]	Speed: 11077.810270 samples/sec	accuracy=0.468640	lr=2.176489
Epoch[98] Batch [549]	Speed: 11557.838707 samples/sec	accuracy=0.468387	lr=2.175809
Epoch[98] Batch [599]	Speed: 11247.804229 samples/sec	accuracy=0.468092	lr=2.175129
[Epoch 98] training: accuracy=0.467811
[Epoch 98] speed: 11114 samples/sec	time cost: 119.480993
[Epoch 98] validation: err-top1=0.530050 err-top5=0.274455
Epoch[99] Batch [49]	Speed: 10615.297469 samples/sec	accuracy=0.475479	lr=2.174094
Epoch[99] Batch [99]	Speed: 11361.642802 samples/sec	accuracy=0.474429	lr=2.173412
Epoch[99] Batch [149]	Speed: 11012.291929 samples/sec	accuracy=0.473317	lr=2.172730
Epoch[99] Batch [199]	Speed: 11187.569102 samples/sec	accuracy=0.472488	lr=2.172048
Epoch[99] Batch [249]	Speed: 10448.417621 samples/sec	accuracy=0.471279	lr=2.171365
Epoch[99] Batch [299]	Speed: 10486.867061 samples/sec	accuracy=0.470910	lr=2.170682
Epoch[99] Batch [349]	Speed: 11159.975942 samples/sec	accuracy=0.470448	lr=2.169998
Epoch[99] Batch [399]	Speed: 11283.210547 samples/sec	accuracy=0.469897	lr=2.169314
Epoch[99] Batch [449]	Speed: 11267.735921 samples/sec	accuracy=0.469413	lr=2.168630
Epoch[99] Batch [499]	Speed: 11746.388364 samples/sec	accuracy=0.469045	lr=2.167945
Epoch[99] Batch [549]	Speed: 11179.747173 samples/sec	accuracy=0.468822	lr=2.167259
Epoch[99] Batch [599]	Speed: 10352.804313 samples/sec	accuracy=0.468572	lr=2.166574
[Epoch 99] training: accuracy=0.468259
[Epoch 99] speed: 10966 samples/sec	time cost: 121.403717
[Epoch 99] validation: err-top1=0.562227 err-top5=0.305684
Epoch[100] Batch [49]	Speed: 10796.922509 samples/sec	accuracy=0.476045	lr=2.165544
Epoch[100] Batch [99]	Speed: 11036.203587 samples/sec	accuracy=0.475073	lr=2.164857
Epoch[100] Batch [149]	Speed: 11339.524601 samples/sec	accuracy=0.474111	lr=2.164170
Epoch[100] Batch [199]	Speed: 10594.848920 samples/sec	accuracy=0.473296	lr=2.163482
Epoch[100] Batch [249]	Speed: 11266.478553 samples/sec	accuracy=0.472779	lr=2.162794
Epoch[100] Batch [299]	Speed: 11746.735972 samples/sec	accuracy=0.472080	lr=2.162105
Epoch[100] Batch [349]	Speed: 10928.264256 samples/sec	accuracy=0.471512	lr=2.161416
Epoch[100] Batch [399]	Speed: 11478.133778 samples/sec	accuracy=0.471038	lr=2.160726
Epoch[100] Batch [449]	Speed: 11419.703257 samples/sec	accuracy=0.471038	lr=2.160036
Epoch[100] Batch [499]	Speed: 10838.289933 samples/sec	accuracy=0.470687	lr=2.159346
Epoch[100] Batch [549]	Speed: 10822.388328 samples/sec	accuracy=0.470213	lr=2.158655
Epoch[100] Batch [599]	Speed: 11378.488241 samples/sec	accuracy=0.469819	lr=2.157964
[Epoch 100] training: accuracy=0.469602
[Epoch 100] speed: 11105 samples/sec	time cost: 119.834478
[Epoch 100] validation: err-top1=0.530701 err-top5=0.278768
Epoch[101] Batch [49]	Speed: 10919.816300 samples/sec	accuracy=0.478975	lr=2.156913
Epoch[101] Batch [99]	Speed: 11495.023329 samples/sec	accuracy=0.475962	lr=2.156220
Epoch[101] Batch [149]	Speed: 10853.492544 samples/sec	accuracy=0.473809	lr=2.155527
Epoch[101] Batch [199]	Speed: 11213.787210 samples/sec	accuracy=0.473340	lr=2.154834
Epoch[101] Batch [249]	Speed: 10350.595026 samples/sec	accuracy=0.472311	lr=2.154141
Epoch[101] Batch [299]	Speed: 11327.030382 samples/sec	accuracy=0.472160	lr=2.153447
Epoch[101] Batch [349]	Speed: 10479.520899 samples/sec	accuracy=0.471540	lr=2.152752
Epoch[101] Batch [399]	Speed: 10749.247787 samples/sec	accuracy=0.471285	lr=2.152057
Epoch[101] Batch [449]	Speed: 10578.273145 samples/sec	accuracy=0.470845	lr=2.151362
Epoch[101] Batch [499]	Speed: 11188.041214 samples/sec	accuracy=0.470489	lr=2.150666
Epoch[101] Batch [549]	Speed: 11656.553196 samples/sec	accuracy=0.470030	lr=2.149970
Epoch[101] Batch [599]	Speed: 10439.025228 samples/sec	accuracy=0.469730	lr=2.149273
[Epoch 101] training: accuracy=0.469698
[Epoch 101] speed: 10920 samples/sec	time cost: 122.095281
[Epoch 101] validation: err-top1=0.508945 err-top5=0.256211
Epoch[102] Batch [49]	Speed: 10275.964740 samples/sec	accuracy=0.480156	lr=2.148214
Epoch[102] Batch [99]	Speed: 11105.642475 samples/sec	accuracy=0.476152	lr=2.147516
Epoch[102] Batch [149]	Speed: 10370.688401 samples/sec	accuracy=0.473428	lr=2.146818
Epoch[102] Batch [199]	Speed: 10190.044462 samples/sec	accuracy=0.472524	lr=2.146119
Epoch[102] Batch [249]	Speed: 11465.924071 samples/sec	accuracy=0.471713	lr=2.145421
Epoch[102] Batch [299]	Speed: 11217.000830 samples/sec	accuracy=0.471375	lr=2.144721
Epoch[102] Batch [349]	Speed: 11338.105694 samples/sec	accuracy=0.470992	lr=2.144021
Epoch[102] Batch [399]	Speed: 11230.485989 samples/sec	accuracy=0.470826	lr=2.143321
Epoch[102] Batch [449]	Speed: 11196.632444 samples/sec	accuracy=0.470858	lr=2.142621
Epoch[102] Batch [499]	Speed: 11255.600480 samples/sec	accuracy=0.470705	lr=2.141919
Epoch[102] Batch [549]	Speed: 11449.101853 samples/sec	accuracy=0.470489	lr=2.141218
Epoch[102] Batch [599]	Speed: 10765.178136 samples/sec	accuracy=0.470233	lr=2.140516
[Epoch 102] training: accuracy=0.470187
[Epoch 102] speed: 10954 samples/sec	time cost: 121.159622
[Epoch 102] validation: err-top1=0.511515 err-top5=0.258667
Epoch[103] Batch [49]	Speed: 11015.929016 samples/sec	accuracy=0.480518	lr=2.139462
Epoch[103] Batch [99]	Speed: 11351.645265 samples/sec	accuracy=0.477666	lr=2.138759
Epoch[103] Batch [149]	Speed: 11113.059948 samples/sec	accuracy=0.475827	lr=2.138056
Epoch[103] Batch [199]	Speed: 11196.647039 samples/sec	accuracy=0.474397	lr=2.137352
Epoch[103] Batch [249]	Speed: 11442.151507 samples/sec	accuracy=0.473492	lr=2.136648
Epoch[103] Batch [299]	Speed: 11413.762722 samples/sec	accuracy=0.472998	lr=2.135943
Epoch[103] Batch [349]	Speed: 11039.056598 samples/sec	accuracy=0.472532	lr=2.135238
Epoch[103] Batch [399]	Speed: 10677.867241 samples/sec	accuracy=0.472139	lr=2.134533
Epoch[103] Batch [449]	Speed: 11063.897537 samples/sec	accuracy=0.471914	lr=2.133827
Epoch[103] Batch [499]	Speed: 10865.131890 samples/sec	accuracy=0.471373	lr=2.133121
Epoch[103] Batch [549]	Speed: 11433.098868 samples/sec	accuracy=0.471103	lr=2.132414
Epoch[103] Batch [599]	Speed: 11652.164100 samples/sec	accuracy=0.470946	lr=2.131707
[Epoch 103] training: accuracy=0.470859
[Epoch 103] speed: 11142 samples/sec	time cost: 119.718533
[Epoch 103] validation: err-top1=0.560371 err-top5=0.300312
Epoch[104] Batch [49]	Speed: 11285.961686 samples/sec	accuracy=0.477031	lr=2.130631
Epoch[104] Batch [99]	Speed: 11243.438758 samples/sec	accuracy=0.476216	lr=2.129923
Epoch[104] Batch [149]	Speed: 11307.807515 samples/sec	accuracy=0.475283	lr=2.129214
Epoch[104] Batch [199]	Speed: 11269.068671 samples/sec	accuracy=0.474578	lr=2.128505
Epoch[104] Batch [249]	Speed: 11751.585343 samples/sec	accuracy=0.473672	lr=2.127796
Epoch[104] Batch [299]	Speed: 11610.569965 samples/sec	accuracy=0.473237	lr=2.127086
Epoch[104] Batch [349]	Speed: 11636.630734 samples/sec	accuracy=0.471438	lr=2.126376
Epoch[104] Batch [399]	Speed: 11305.502797 samples/sec	accuracy=0.470673	lr=2.125665
Epoch[104] Batch [449]	Speed: 11225.937063 samples/sec	accuracy=0.470303	lr=2.124954
Epoch[104] Batch [499]	Speed: 11385.156511 samples/sec	accuracy=0.470016	lr=2.124243
Epoch[104] Batch [549]	Speed: 11536.954263 samples/sec	accuracy=0.469869	lr=2.123531
Epoch[104] Batch [599]	Speed: 11113.586757 samples/sec	accuracy=0.469700	lr=2.122818
[Epoch 104] training: accuracy=0.469765
[Epoch 104] speed: 11357 samples/sec	time cost: 117.168052
[Epoch 104] validation: err-top1=0.510783 err-top5=0.257731
Epoch[105] Batch [49]	Speed: 11049.340856 samples/sec	accuracy=0.480801	lr=2.121749
Epoch[105] Batch [99]	Speed: 10586.828805 samples/sec	accuracy=0.478457	lr=2.121036
Epoch[105] Batch [149]	Speed: 10582.906222 samples/sec	accuracy=0.476699	lr=2.120322
Epoch[105] Batch [199]	Speed: 11541.208908 samples/sec	accuracy=0.475483	lr=2.119608
Epoch[105] Batch [249]	Speed: 11227.804092 samples/sec	accuracy=0.475033	lr=2.118893
Epoch[105] Batch [299]	Speed: 11235.751633 samples/sec	accuracy=0.474359	lr=2.118178
Epoch[105] Batch [349]	Speed: 11689.272166 samples/sec	accuracy=0.473199	lr=2.117463
Epoch[105] Batch [399]	Speed: 11134.066578 samples/sec	accuracy=0.472545	lr=2.116747
Epoch[105] Batch [449]	Speed: 11004.211496 samples/sec	accuracy=0.472014	lr=2.116031
Epoch[105] Batch [499]	Speed: 11029.311449 samples/sec	accuracy=0.471936	lr=2.115314
Epoch[105] Batch [549]	Speed: 11524.555771 samples/sec	accuracy=0.471522	lr=2.114597
Epoch[105] Batch [599]	Speed: 11244.643591 samples/sec	accuracy=0.471252	lr=2.113879
[Epoch 105] training: accuracy=0.471083
[Epoch 105] speed: 11138 samples/sec	time cost: 119.672848
[Epoch 105] validation: err-top1=0.544861 err-top5=0.288676
Epoch[106] Batch [49]	Speed: 10715.349175 samples/sec	accuracy=0.479131	lr=2.112788
Epoch[106] Batch [99]	Speed: 10144.025465 samples/sec	accuracy=0.476040	lr=2.112070
Epoch[106] Batch [149]	Speed: 10443.155952 samples/sec	accuracy=0.475078	lr=2.111351
Epoch[106] Batch [199]	Speed: 11398.406414 samples/sec	accuracy=0.474138	lr=2.110631
Epoch[106] Batch [249]	Speed: 11609.785034 samples/sec	accuracy=0.473779	lr=2.109912
Epoch[106] Batch [299]	Speed: 11221.322630 samples/sec	accuracy=0.473576	lr=2.109191
Epoch[106] Batch [349]	Speed: 11725.391202 samples/sec	accuracy=0.472981	lr=2.108471
Epoch[106] Batch [399]	Speed: 11697.873297 samples/sec	accuracy=0.472688	lr=2.107750
Epoch[106] Batch [449]	Speed: 11249.795226 samples/sec	accuracy=0.472215	lr=2.107029
Epoch[106] Batch [499]	Speed: 11661.491439 samples/sec	accuracy=0.471807	lr=2.106307
Epoch[106] Batch [549]	Speed: 11022.902599 samples/sec	accuracy=0.471713	lr=2.105585
Epoch[106] Batch [599]	Speed: 11317.238269 samples/sec	accuracy=0.471462	lr=2.104862
[Epoch 106] training: accuracy=0.471386
[Epoch 106] speed: 11168 samples/sec	time cost: 119.299089
[Epoch 106] validation: err-top1=0.530859 err-top5=0.276113
Epoch[107] Batch [49]	Speed: 10373.157301 samples/sec	accuracy=0.480488	lr=2.103778
Epoch[107] Batch [99]	Speed: 11411.436445 samples/sec	accuracy=0.479028	lr=2.103054
Epoch[107] Batch [149]	Speed: 11420.876314 samples/sec	accuracy=0.477627	lr=2.102330
Epoch[107] Batch [199]	Speed: 11335.924144 samples/sec	accuracy=0.476602	lr=2.101606
Epoch[107] Batch [249]	Speed: 11332.749081 samples/sec	accuracy=0.475400	lr=2.100881
Epoch[107] Batch [299]	Speed: 10451.344814 samples/sec	accuracy=0.474465	lr=2.100156
Epoch[107] Batch [349]	Speed: 11667.639465 samples/sec	accuracy=0.474181	lr=2.099430
Epoch[107] Batch [399]	Speed: 11080.154853 samples/sec	accuracy=0.473779	lr=2.098704
Epoch[107] Batch [449]	Speed: 10614.301891 samples/sec	accuracy=0.473490	lr=2.097978
Epoch[107] Batch [499]	Speed: 10306.186278 samples/sec	accuracy=0.473339	lr=2.097251
Epoch[107] Batch [549]	Speed: 11342.226294 samples/sec	accuracy=0.472876	lr=2.096524
Epoch[107] Batch [599]	Speed: 11701.192201 samples/sec	accuracy=0.472532	lr=2.095796
[Epoch 107] training: accuracy=0.472384
[Epoch 107] speed: 11025 samples/sec	time cost: 120.720632
[Epoch 107] validation: err-top1=0.520630 err-top5=0.269694
Epoch[108] Batch [49]	Speed: 10758.736946 samples/sec	accuracy=0.478799	lr=2.094690
Epoch[108] Batch [99]	Speed: 9792.516308 samples/sec	accuracy=0.477935	lr=2.093961
Epoch[108] Batch [149]	Speed: 10856.932164 samples/sec	accuracy=0.477012	lr=2.093232
Epoch[108] Batch [199]	Speed: 11653.298136 samples/sec	accuracy=0.475981	lr=2.092503
Epoch[108] Batch [249]	Speed: 11318.346523 samples/sec	accuracy=0.475346	lr=2.091773
Epoch[108] Batch [299]	Speed: 11074.228439 samples/sec	accuracy=0.474909	lr=2.091043
Epoch[108] Batch [349]	Speed: 10478.435584 samples/sec	accuracy=0.474381	lr=2.090312
Epoch[108] Batch [399]	Speed: 10428.694722 samples/sec	accuracy=0.473629	lr=2.089581
Epoch[108] Batch [449]	Speed: 10962.775913 samples/sec	accuracy=0.473271	lr=2.088850
Epoch[108] Batch [499]	Speed: 10507.893533 samples/sec	accuracy=0.472726	lr=2.088118
Epoch[108] Batch [549]	Speed: 10923.956541 samples/sec	accuracy=0.472758	lr=2.087386
Epoch[108] Batch [599]	Speed: 10681.480661 samples/sec	accuracy=0.472515	lr=2.086654
[Epoch 108] training: accuracy=0.472382
[Epoch 108] speed: 10766 samples/sec	time cost: 123.603944
[Epoch 108] validation: err-top1=0.514766 err-top5=0.258340
Epoch[109] Batch [49]	Speed: 10166.727303 samples/sec	accuracy=0.479336	lr=2.085540
Epoch[109] Batch [99]	Speed: 11397.046526 samples/sec	accuracy=0.478335	lr=2.084806
Epoch[109] Batch [149]	Speed: 11333.005055 samples/sec	accuracy=0.476800	lr=2.084072
Epoch[109] Batch [199]	Speed: 11164.487810 samples/sec	accuracy=0.475879	lr=2.083338
Epoch[109] Batch [249]	Speed: 10497.590036 samples/sec	accuracy=0.475236	lr=2.082603
Epoch[109] Batch [299]	Speed: 11173.950111 samples/sec	accuracy=0.474744	lr=2.081868
Epoch[109] Batch [349]	Speed: 11333.435091 samples/sec	accuracy=0.474272	lr=2.081132
Epoch[109] Batch [399]	Speed: 11022.833006 samples/sec	accuracy=0.474073	lr=2.080397
Epoch[109] Batch [449]	Speed: 11406.048199 samples/sec	accuracy=0.473479	lr=2.079660
Epoch[109] Batch [499]	Speed: 11203.084239 samples/sec	accuracy=0.472874	lr=2.078924
Epoch[109] Batch [549]	Speed: 11110.806044 samples/sec	accuracy=0.472649	lr=2.078187
Epoch[109] Batch [599]	Speed: 10816.475045 samples/sec	accuracy=0.472360	lr=2.077449
[Epoch 109] training: accuracy=0.472439
[Epoch 109] speed: 10987 samples/sec	time cost: 120.820531
[Epoch 109] validation: err-top1=0.545146 err-top5=0.286275
Epoch[110] Batch [49]	Speed: 10623.536321 samples/sec	accuracy=0.483486	lr=2.076342
Epoch[110] Batch [99]	Speed: 10605.765165 samples/sec	accuracy=0.480923	lr=2.075604
Epoch[110] Batch [149]	Speed: 10812.911283 samples/sec	accuracy=0.479948	lr=2.074865
Epoch[110] Batch [199]	Speed: 10872.854594 samples/sec	accuracy=0.478281	lr=2.074126
Epoch[110] Batch [249]	Speed: 10937.842166 samples/sec	accuracy=0.477490	lr=2.073386
Epoch[110] Batch [299]	Speed: 11372.744680 samples/sec	accuracy=0.476569	lr=2.072646
Epoch[110] Batch [349]	Speed: 11692.064172 samples/sec	accuracy=0.475705	lr=2.071906
Epoch[110] Batch [399]	Speed: 11302.947068 samples/sec	accuracy=0.475341	lr=2.071165
Epoch[110] Batch [449]	Speed: 11275.810569 samples/sec	accuracy=0.474932	lr=2.070424
Epoch[110] Batch [499]	Speed: 10631.595590 samples/sec	accuracy=0.474575	lr=2.069683
Epoch[110] Batch [549]	Speed: 11521.946729 samples/sec	accuracy=0.474267	lr=2.068941
Epoch[110] Batch [599]	Speed: 10894.654248 samples/sec	accuracy=0.473888	lr=2.068198
[Epoch 110] training: accuracy=0.473790
[Epoch 110] speed: 11004 samples/sec	time cost: 120.862940
[Epoch 110] validation: err-top1=0.575236 err-top5=0.317851
Epoch[111] Batch [49]	Speed: 11385.167979 samples/sec	accuracy=0.482451	lr=2.067069
Epoch[111] Batch [99]	Speed: 11208.086126 samples/sec	accuracy=0.479756	lr=2.066326
Epoch[111] Batch [149]	Speed: 11570.229162 samples/sec	accuracy=0.478008	lr=2.065582
Epoch[111] Batch [199]	Speed: 11301.638708 samples/sec	accuracy=0.476904	lr=2.064838
Epoch[111] Batch [249]	Speed: 10849.713863 samples/sec	accuracy=0.476150	lr=2.064094
Epoch[111] Batch [299]	Speed: 10783.069984 samples/sec	accuracy=0.475496	lr=2.063349
Epoch[111] Batch [349]	Speed: 11006.577489 samples/sec	accuracy=0.475001	lr=2.062604
Epoch[111] Batch [399]	Speed: 11645.186117 samples/sec	accuracy=0.474823	lr=2.061859
Epoch[111] Batch [449]	Speed: 11251.388414 samples/sec	accuracy=0.474514	lr=2.061113
Epoch[111] Batch [499]	Speed: 10537.089983 samples/sec	accuracy=0.473893	lr=2.060366
Epoch[111] Batch [549]	Speed: 11263.834089 samples/sec	accuracy=0.473670	lr=2.059620
Epoch[111] Batch [599]	Speed: 10888.400062 samples/sec	accuracy=0.473197	lr=2.058872
[Epoch 111] training: accuracy=0.473189
[Epoch 111] speed: 11109 samples/sec	time cost: 119.903407
[Epoch 111] validation: err-top1=0.524961 err-top5=0.271680
Epoch[112] Batch [49]	Speed: 10290.812910 samples/sec	accuracy=0.481553	lr=2.057751
Epoch[112] Batch [99]	Speed: 11677.739929 samples/sec	accuracy=0.479839	lr=2.057003
Epoch[112] Batch [149]	Speed: 11539.141337 samples/sec	accuracy=0.478444	lr=2.056255
Epoch[112] Batch [199]	Speed: 11443.051432 samples/sec	accuracy=0.477888	lr=2.055506
Epoch[112] Batch [249]	Speed: 11052.805627 samples/sec	accuracy=0.476951	lr=2.054757
Epoch[112] Batch [299]	Speed: 11375.352562 samples/sec	accuracy=0.477010	lr=2.054007
Epoch[112] Batch [349]	Speed: 11498.274592 samples/sec	accuracy=0.476332	lr=2.053257
Epoch[112] Batch [399]	Speed: 11400.114294 samples/sec	accuracy=0.475521	lr=2.052507
Epoch[112] Batch [449]	Speed: 11060.232129 samples/sec	accuracy=0.474991	lr=2.051756
Epoch[112] Batch [499]	Speed: 11514.970973 samples/sec	accuracy=0.474369	lr=2.051005
Epoch[112] Batch [549]	Speed: 10266.479429 samples/sec	accuracy=0.474018	lr=2.050254
Epoch[112] Batch [599]	Speed: 11098.892234 samples/sec	accuracy=0.473737	lr=2.049502
[Epoch 112] training: accuracy=0.473684
[Epoch 112] speed: 11130 samples/sec	time cost: 119.556217
[Epoch 112] validation: err-top1=0.551310 err-top5=0.299296
Epoch[113] Batch [49]	Speed: 11453.409525 samples/sec	accuracy=0.484287	lr=2.048359
Epoch[113] Batch [99]	Speed: 11100.160381 samples/sec	accuracy=0.483013	lr=2.047606
Epoch[113] Batch [149]	Speed: 11057.898532 samples/sec	accuracy=0.481429	lr=2.046853
Epoch[113] Batch [199]	Speed: 10876.324677 samples/sec	accuracy=0.480188	lr=2.046099
Epoch[113] Batch [249]	Speed: 11184.995627 samples/sec	accuracy=0.478857	lr=2.045345
Epoch[113] Batch [299]	Speed: 11046.356384 samples/sec	accuracy=0.477964	lr=2.044591
Epoch[113] Batch [349]	Speed: 11219.049218 samples/sec	accuracy=0.477427	lr=2.043836
Epoch[113] Batch [399]	Speed: 10520.380940 samples/sec	accuracy=0.476466	lr=2.043081
Epoch[113] Batch [449]	Speed: 10454.455593 samples/sec	accuracy=0.475909	lr=2.042326
Epoch[113] Batch [499]	Speed: 11065.833077 samples/sec	accuracy=0.475564	lr=2.041570
Epoch[113] Batch [549]	Speed: 11542.841970 samples/sec	accuracy=0.475299	lr=2.040814
Epoch[113] Batch [599]	Speed: 11386.224981 samples/sec	accuracy=0.475018	lr=2.040058
[Epoch 113] training: accuracy=0.474868
[Epoch 113] speed: 11065 samples/sec	time cost: 120.340217
[Epoch 113] validation: err-top1=0.491719 err-top5=0.244082
Epoch[114] Batch [49]	Speed: 11155.646495 samples/sec	accuracy=0.480664	lr=2.038922
Epoch[114] Batch [99]	Speed: 11153.252770 samples/sec	accuracy=0.480127	lr=2.038165
Epoch[114] Batch [149]	Speed: 11297.758239 samples/sec	accuracy=0.478258	lr=2.037407
Epoch[114] Batch [199]	Speed: 10639.042735 samples/sec	accuracy=0.477644	lr=2.036649
Epoch[114] Batch [249]	Speed: 11369.736770 samples/sec	accuracy=0.476848	lr=2.035890
Epoch[114] Batch [299]	Speed: 11719.932083 samples/sec	accuracy=0.476276	lr=2.035132
Epoch[114] Batch [349]	Speed: 10794.792834 samples/sec	accuracy=0.475650	lr=2.034372
Epoch[114] Batch [399]	Speed: 11169.602555 samples/sec	accuracy=0.475349	lr=2.033613
Epoch[114] Batch [449]	Speed: 11060.238395 samples/sec	accuracy=0.475259	lr=2.032853
Epoch[114] Batch [499]	Speed: 11473.950024 samples/sec	accuracy=0.474999	lr=2.032092
Epoch[114] Batch [549]	Speed: 11324.194395 samples/sec	accuracy=0.474963	lr=2.031332
Epoch[114] Batch [599]	Speed: 11278.181379 samples/sec	accuracy=0.474541	lr=2.030570
[Epoch 114] training: accuracy=0.474595
[Epoch 114] speed: 11156 samples/sec	time cost: 119.418142
[Epoch 114] validation: err-top1=0.547567 err-top5=0.289978
Epoch[115] Batch [49]	Speed: 10582.935949 samples/sec	accuracy=0.484297	lr=2.029413
Epoch[115] Batch [99]	Speed: 11137.513071 samples/sec	accuracy=0.482847	lr=2.028651
Epoch[115] Batch [149]	Speed: 10965.346677 samples/sec	accuracy=0.480238	lr=2.027888
Epoch[115] Batch [199]	Speed: 11444.868784 samples/sec	accuracy=0.479041	lr=2.027126
Epoch[115] Batch [249]	Speed: 11361.520778 samples/sec	accuracy=0.479180	lr=2.026363
Epoch[115] Batch [299]	Speed: 11064.087355 samples/sec	accuracy=0.478304	lr=2.025599
Epoch[115] Batch [349]	Speed: 10126.117680 samples/sec	accuracy=0.477891	lr=2.024835
Epoch[115] Batch [399]	Speed: 11206.850807 samples/sec	accuracy=0.477656	lr=2.024071
Epoch[115] Batch [449]	Speed: 11049.291111 samples/sec	accuracy=0.476674	lr=2.023307
Epoch[115] Batch [499]	Speed: 11247.845173 samples/sec	accuracy=0.476404	lr=2.022542
Epoch[115] Batch [549]	Speed: 11373.337358 samples/sec	accuracy=0.475882	lr=2.021776
Epoch[115] Batch [599]	Speed: 11302.643968 samples/sec	accuracy=0.475522	lr=2.021011
[Epoch 115] training: accuracy=0.475347
[Epoch 115] speed: 11045 samples/sec	time cost: 121.011302
[Epoch 115] validation: err-top1=0.518477 err-top5=0.265508
Epoch[116] Batch [49]	Speed: 11139.396161 samples/sec	accuracy=0.481016	lr=2.019846
Epoch[116] Batch [99]	Speed: 10076.674093 samples/sec	accuracy=0.479478	lr=2.019080
Epoch[116] Batch [149]	Speed: 11288.290484 samples/sec	accuracy=0.478262	lr=2.018313
Epoch[116] Batch [199]	Speed: 11376.274854 samples/sec	accuracy=0.477358	lr=2.017545
Epoch[116] Batch [249]	Speed: 10890.025614 samples/sec	accuracy=0.476992	lr=2.016778
Epoch[116] Batch [299]	Speed: 10559.380226 samples/sec	accuracy=0.476331	lr=2.016010
Epoch[116] Batch [349]	Speed: 11106.901537 samples/sec	accuracy=0.476331	lr=2.015241
Epoch[116] Batch [399]	Speed: 11268.718897 samples/sec	accuracy=0.476472	lr=2.014473
Epoch[116] Batch [449]	Speed: 11414.458878 samples/sec	accuracy=0.476305	lr=2.013704
Epoch[116] Batch [499]	Speed: 11334.793600 samples/sec	accuracy=0.476060	lr=2.012934
Epoch[116] Batch [549]	Speed: 10991.141037 samples/sec	accuracy=0.475790	lr=2.012164
Epoch[116] Batch [599]	Speed: 10614.443543 samples/sec	accuracy=0.475470	lr=2.011394
[Epoch 116] training: accuracy=0.475392
[Epoch 116] speed: 10985 samples/sec	time cost: 120.973968
[Epoch 116] validation: err-top1=0.523336 err-top5=0.265198
Epoch[117] Batch [49]	Speed: 10816.416751 samples/sec	accuracy=0.483281	lr=2.010238
Epoch[117] Batch [99]	Speed: 11403.337515 samples/sec	accuracy=0.481509	lr=2.009467
Epoch[117] Batch [149]	Speed: 11197.832521 samples/sec	accuracy=0.480612	lr=2.008696
Epoch[117] Batch [199]	Speed: 11093.056071 samples/sec	accuracy=0.480244	lr=2.007924
Epoch[117] Batch [249]	Speed: 11584.989386 samples/sec	accuracy=0.479723	lr=2.007152
Epoch[117] Batch [299]	Speed: 11245.409366 samples/sec	accuracy=0.478882	lr=2.006380
Epoch[117] Batch [349]	Speed: 10358.420468 samples/sec	accuracy=0.478281	lr=2.005607
Epoch[117] Batch [399]	Speed: 11105.700482 samples/sec	accuracy=0.477727	lr=2.004834
Epoch[117] Batch [449]	Speed: 11434.319123 samples/sec	accuracy=0.477483	lr=2.004060
Epoch[117] Batch [499]	Speed: 11603.069285 samples/sec	accuracy=0.476929	lr=2.003286
Epoch[117] Batch [549]	Speed: 11726.768783 samples/sec	accuracy=0.476961	lr=2.002512
Epoch[117] Batch [599]	Speed: 11121.413109 samples/sec	accuracy=0.476605	lr=2.001737
[Epoch 117] training: accuracy=0.476458
[Epoch 117] speed: 11193 samples/sec	time cost: 118.943066
[Epoch 117] validation: err-top1=0.554728 err-top5=0.301290
Epoch[118] Batch [49]	Speed: 11269.207641 samples/sec	accuracy=0.485615	lr=2.000559
Epoch[118] Batch [99]	Speed: 11279.516898 samples/sec	accuracy=0.482720	lr=1.999784
Epoch[118] Batch [149]	Speed: 11667.999861 samples/sec	accuracy=0.481074	lr=1.999008
Epoch[118] Batch [199]	Speed: 11083.467366 samples/sec	accuracy=0.480596	lr=1.998232
Epoch[118] Batch [249]	Speed: 11103.252363 samples/sec	accuracy=0.479715	lr=1.997456
Epoch[118] Batch [299]	Speed: 11143.305337 samples/sec	accuracy=0.479131	lr=1.996679
Epoch[118] Batch [349]	Speed: 11372.056312 samples/sec	accuracy=0.479424	lr=1.995901
Epoch[118] Batch [399]	Speed: 11098.807912 samples/sec	accuracy=0.478727	lr=1.995124
Epoch[118] Batch [449]	Speed: 11263.652125 samples/sec	accuracy=0.478432	lr=1.994346
Epoch[118] Batch [499]	Speed: 11350.433293 samples/sec	accuracy=0.478092	lr=1.993568
Epoch[118] Batch [549]	Speed: 10416.259694 samples/sec	accuracy=0.477595	lr=1.992789
Epoch[118] Batch [599]	Speed: 11460.597102 samples/sec	accuracy=0.477274	lr=1.992010
[Epoch 118] training: accuracy=0.477252
[Epoch 118] speed: 11181 samples/sec	time cost: 118.999208
[Epoch 118] validation: err-top1=0.609199 err-top5=0.354941
Epoch[119] Batch [49]	Speed: 9248.631765 samples/sec	accuracy=0.481455	lr=1.990841
Epoch[119] Batch [99]	Speed: 10864.148533 samples/sec	accuracy=0.481655	lr=1.990061
Epoch[119] Batch [149]	Speed: 11217.875647 samples/sec	accuracy=0.481051	lr=1.989281
Epoch[119] Batch [199]	Speed: 11170.808756 samples/sec	accuracy=0.480378	lr=1.988501
Epoch[119] Batch [249]	Speed: 10993.753826 samples/sec	accuracy=0.479619	lr=1.987720
Epoch[119] Batch [299]	Speed: 11565.123737 samples/sec	accuracy=0.478953	lr=1.986939
Epoch[119] Batch [349]	Speed: 11279.660865 samples/sec	accuracy=0.478811	lr=1.986157
Epoch[119] Batch [399]	Speed: 11062.065242 samples/sec	accuracy=0.478392	lr=1.985375
Epoch[119] Batch [449]	Speed: 11108.210013 samples/sec	accuracy=0.477806	lr=1.984593
Epoch[119] Batch [499]	Speed: 11375.013632 samples/sec	accuracy=0.477549	lr=1.983810
Epoch[119] Batch [549]	Speed: 11147.520758 samples/sec	accuracy=0.477083	lr=1.983027
Epoch[119] Batch [599]	Speed: 11152.615042 samples/sec	accuracy=0.476633	lr=1.982244
[Epoch 119] training: accuracy=0.476726
[Epoch 119] speed: 10940 samples/sec	time cost: 121.549051
[Epoch 119] validation: err-top1=0.554667 err-top5=0.300496
Epoch[120] Batch [49]	Speed: 10852.085993 samples/sec	accuracy=0.484180	lr=1.981053
Epoch[120] Batch [99]	Speed: 10402.359036 samples/sec	accuracy=0.480103	lr=1.980269
Epoch[120] Batch [149]	Speed: 11146.672790 samples/sec	accuracy=0.481019	lr=1.979484
Epoch[120] Batch [199]	Speed: 11073.611422 samples/sec	accuracy=0.480488	lr=1.978700
Epoch[120] Batch [249]	Speed: 11574.770097 samples/sec	accuracy=0.480244	lr=1.977914
Epoch[120] Batch [299]	Speed: 11495.602667 samples/sec	accuracy=0.480207	lr=1.977129
Epoch[120] Batch [349]	Speed: 11177.967946 samples/sec	accuracy=0.479807	lr=1.976343
Epoch[120] Batch [399]	Speed: 10776.590067 samples/sec	accuracy=0.479376	lr=1.975557
Epoch[120] Batch [449]	Speed: 10559.851434 samples/sec	accuracy=0.478768	lr=1.974770
Epoch[120] Batch [499]	Speed: 10793.172803 samples/sec	accuracy=0.478655	lr=1.973984
Epoch[120] Batch [549]	Speed: 11225.049839 samples/sec	accuracy=0.478401	lr=1.973196
Epoch[120] Batch [599]	Speed: 10571.251382 samples/sec	accuracy=0.478276	lr=1.972409
[Epoch 120] training: accuracy=0.478142
[Epoch 120] speed: 10918 samples/sec	time cost: 121.855707
[Epoch 120] validation: err-top1=0.517266 err-top5=0.263809
Epoch[121] Batch [49]	Speed: 10930.704522 samples/sec	accuracy=0.485186	lr=1.971227
Epoch[121] Batch [99]	Speed: 11153.669563 samples/sec	accuracy=0.483809	lr=1.970439
Epoch[121] Batch [149]	Speed: 11464.177139 samples/sec	accuracy=0.482311	lr=1.969650
Epoch[121] Batch [199]	Speed: 11392.197916 samples/sec	accuracy=0.482095	lr=1.968861
Epoch[121] Batch [249]	Speed: 11171.514237 samples/sec	accuracy=0.481006	lr=1.968072
Epoch[121] Batch [299]	Speed: 11222.995154 samples/sec	accuracy=0.480153	lr=1.967282
Epoch[121] Batch [349]	Speed: 10878.701845 samples/sec	accuracy=0.479445	lr=1.966492
Epoch[121] Batch [399]	Speed: 11193.310879 samples/sec	accuracy=0.479154	lr=1.965701
Epoch[121] Batch [449]	Speed: 11355.770310 samples/sec	accuracy=0.478636	lr=1.964911
Epoch[121] Batch [499]	Speed: 11389.257025 samples/sec	accuracy=0.478454	lr=1.964120
Epoch[121] Batch [549]	Speed: 9859.160297 samples/sec	accuracy=0.478256	lr=1.963328
Epoch[121] Batch [599]	Speed: 10933.302847 samples/sec	accuracy=0.478105	lr=1.962537
[Epoch 121] training: accuracy=0.477907
[Epoch 121] speed: 11046 samples/sec	time cost: 120.314231
[Epoch 121] validation: err-top1=0.517253 err-top5=0.263184
Epoch[122] Batch [49]	Speed: 10166.861593 samples/sec	accuracy=0.483486	lr=1.961333
Epoch[122] Batch [99]	Speed: 10576.148110 samples/sec	accuracy=0.482891	lr=1.960540
Epoch[122] Batch [149]	Speed: 10738.615596 samples/sec	accuracy=0.482083	lr=1.959747
Epoch[122] Batch [199]	Speed: 11268.653262 samples/sec	accuracy=0.481270	lr=1.958954
Epoch[122] Batch [249]	Speed: 11243.889400 samples/sec	accuracy=0.481051	lr=1.958161
Epoch[122] Batch [299]	Speed: 11407.107870 samples/sec	accuracy=0.480365	lr=1.957367
Epoch[122] Batch [349]	Speed: 11373.300314 samples/sec	accuracy=0.480414	lr=1.956573
Epoch[122] Batch [399]	Speed: 10824.293205 samples/sec	accuracy=0.479963	lr=1.955778
Epoch[122] Batch [449]	Speed: 10876.913844 samples/sec	accuracy=0.479735	lr=1.954983
Epoch[122] Batch [499]	Speed: 10831.143100 samples/sec	accuracy=0.479473	lr=1.954188
Epoch[122] Batch [549]	Speed: 11106.861613 samples/sec	accuracy=0.479048	lr=1.953392
Epoch[122] Batch [599]	Speed: 10746.175836 samples/sec	accuracy=0.478720	lr=1.952597
[Epoch 122] training: accuracy=0.478541
[Epoch 122] speed: 10890 samples/sec	time cost: 122.114622
[Epoch 122] validation: err-top1=0.557048 err-top5=0.299337
Epoch[123] Batch [49]	Speed: 11084.649028 samples/sec	accuracy=0.485742	lr=1.951386
Epoch[123] Batch [99]	Speed: 11380.989586 samples/sec	accuracy=0.484380	lr=1.950590
Epoch[123] Batch [149]	Speed: 11234.898418 samples/sec	accuracy=0.482786	lr=1.949793
Epoch[123] Batch [199]	Speed: 11336.976211 samples/sec	accuracy=0.482781	lr=1.948995
Epoch[123] Batch [249]	Speed: 10959.872974 samples/sec	accuracy=0.481518	lr=1.948198
Epoch[123] Batch [299]	Speed: 11192.385930 samples/sec	accuracy=0.481393	lr=1.947400
Epoch[123] Batch [349]	Speed: 11194.813407 samples/sec	accuracy=0.480932	lr=1.946602
Epoch[123] Batch [399]	Speed: 10554.968745 samples/sec	accuracy=0.480297	lr=1.945803
Epoch[123] Batch [449]	Speed: 10938.658658 samples/sec	accuracy=0.479873	lr=1.945004
Epoch[123] Batch [499]	Speed: 10769.064752 samples/sec	accuracy=0.479073	lr=1.944205
Epoch[123] Batch [549]	Speed: 11076.143602 samples/sec	accuracy=0.478705	lr=1.943405
Epoch[123] Batch [599]	Speed: 10493.643042 samples/sec	accuracy=0.478609	lr=1.942605
[Epoch 123] training: accuracy=0.478287
[Epoch 123] speed: 10992 samples/sec	time cost: 121.127881
[Epoch 123] validation: err-top1=0.547129 err-top5=0.286758
Epoch[124] Batch [49]	Speed: 11294.496107 samples/sec	accuracy=0.490361	lr=1.941405
Epoch[124] Batch [99]	Speed: 11399.290696 samples/sec	accuracy=0.486260	lr=1.940604
Epoch[124] Batch [149]	Speed: 11248.011898 samples/sec	accuracy=0.484593	lr=1.939803
Epoch[124] Batch [199]	Speed: 10774.337045 samples/sec	accuracy=0.483274	lr=1.939002
Epoch[124] Batch [249]	Speed: 11657.486214 samples/sec	accuracy=0.482305	lr=1.938200
Epoch[124] Batch [299]	Speed: 11232.735836 samples/sec	accuracy=0.481992	lr=1.937398
Epoch[124] Batch [349]	Speed: 10653.717838 samples/sec	accuracy=0.481733	lr=1.936596
Epoch[124] Batch [399]	Speed: 10993.077651 samples/sec	accuracy=0.481362	lr=1.935793
Epoch[124] Batch [449]	Speed: 11478.359856 samples/sec	accuracy=0.480437	lr=1.934990
Epoch[124] Batch [499]	Speed: 11295.668237 samples/sec	accuracy=0.479658	lr=1.934187
Epoch[124] Batch [549]	Speed: 10840.466632 samples/sec	accuracy=0.479400	lr=1.933384
Epoch[124] Batch [599]	Speed: 11109.419657 samples/sec	accuracy=0.479127	lr=1.932580
[Epoch 124] training: accuracy=0.479022
[Epoch 124] speed: 11144 samples/sec	time cost: 119.456517
[Epoch 124] validation: err-top1=0.530416 err-top5=0.270894
Epoch[125] Batch [49]	Speed: 11286.796277 samples/sec	accuracy=0.485820	lr=1.931357
Epoch[125] Batch [99]	Speed: 11341.051669 samples/sec	accuracy=0.484917	lr=1.930552
Epoch[125] Batch [149]	Speed: 11236.083197 samples/sec	accuracy=0.483717	lr=1.929747
Epoch[125] Batch [199]	Speed: 10466.011536 samples/sec	accuracy=0.482849	lr=1.928942
Epoch[125] Batch [249]	Speed: 10623.208918 samples/sec	accuracy=0.481998	lr=1.928137
Epoch[125] Batch [299]	Speed: 11106.138713 samples/sec	accuracy=0.481252	lr=1.927331
Epoch[125] Batch [349]	Speed: 11571.524694 samples/sec	accuracy=0.480880	lr=1.926524
Epoch[125] Batch [399]	Speed: 11016.467002 samples/sec	accuracy=0.480504	lr=1.925718
Epoch[125] Batch [449]	Speed: 11274.637524 samples/sec	accuracy=0.480358	lr=1.924911
Epoch[125] Batch [499]	Speed: 10787.314965 samples/sec	accuracy=0.480018	lr=1.924104
Epoch[125] Batch [549]	Speed: 11165.280149 samples/sec	accuracy=0.480003	lr=1.923296
Epoch[125] Batch [599]	Speed: 11219.613675 samples/sec	accuracy=0.479649	lr=1.922488
[Epoch 125] training: accuracy=0.479674
[Epoch 125] speed: 11054 samples/sec	time cost: 120.355739
[Epoch 125] validation: err-top1=0.537402 err-top5=0.282188
Epoch[126] Batch [49]	Speed: 10357.732259 samples/sec	accuracy=0.486563	lr=1.921276
Epoch[126] Batch [99]	Speed: 11266.467618 samples/sec	accuracy=0.486182	lr=1.920467
Epoch[126] Batch [149]	Speed: 10740.342030 samples/sec	accuracy=0.485169	lr=1.919658
Epoch[126] Batch [199]	Speed: 11354.046575 samples/sec	accuracy=0.484204	lr=1.918849
Epoch[126] Batch [249]	Speed: 11394.299613 samples/sec	accuracy=0.483451	lr=1.918040
Epoch[126] Batch [299]	Speed: 11673.308216 samples/sec	accuracy=0.482985	lr=1.917230
Epoch[126] Batch [349]	Speed: 11209.050530 samples/sec	accuracy=0.482746	lr=1.916420
Epoch[126] Batch [399]	Speed: 11379.673047 samples/sec	accuracy=0.482007	lr=1.915609
Epoch[126] Batch [449]	Speed: 11243.317789 samples/sec	accuracy=0.481328	lr=1.914798
Epoch[126] Batch [499]	Speed: 11210.313254 samples/sec	accuracy=0.481141	lr=1.913987
Epoch[126] Batch [549]	Speed: 11103.175150 samples/sec	accuracy=0.480869	lr=1.913176
Epoch[126] Batch [599]	Speed: 11113.317882 samples/sec	accuracy=0.480647	lr=1.912364
[Epoch 126] training: accuracy=0.480468
[Epoch 126] speed: 11096 samples/sec	time cost: 119.916647
[Epoch 126] validation: err-top1=0.579671 err-top5=0.322754
Epoch[127] Batch [49]	Speed: 10394.583062 samples/sec	accuracy=0.487549	lr=1.911130
Epoch[127] Batch [99]	Speed: 11311.351403 samples/sec	accuracy=0.486782	lr=1.910317
Epoch[127] Batch [149]	Speed: 11246.625812 samples/sec	accuracy=0.486143	lr=1.909505
Epoch[127] Batch [199]	Speed: 10528.169812 samples/sec	accuracy=0.485149	lr=1.908692
Epoch[127] Batch [249]	Speed: 11369.611864 samples/sec	accuracy=0.484002	lr=1.907878
Epoch[127] Batch [299]	Speed: 11513.167709 samples/sec	accuracy=0.483779	lr=1.907064
Epoch[127] Batch [349]	Speed: 11424.111592 samples/sec	accuracy=0.483309	lr=1.906250
Epoch[127] Batch [399]	Speed: 11386.190871 samples/sec	accuracy=0.482673	lr=1.905436
Epoch[127] Batch [449]	Speed: 11314.725818 samples/sec	accuracy=0.482089	lr=1.904622
Epoch[127] Batch [499]	Speed: 10476.323636 samples/sec	accuracy=0.482030	lr=1.903807
Epoch[127] Batch [549]	Speed: 10391.835164 samples/sec	accuracy=0.481482	lr=1.902991
Epoch[127] Batch [599]	Speed: 11377.172281 samples/sec	accuracy=0.481144	lr=1.902176
[Epoch 127] training: accuracy=0.480778
[Epoch 127] speed: 11032 samples/sec	time cost: 120.249674
[Epoch 127] validation: err-top1=0.522909 err-top5=0.265645
Epoch[128] Batch [49]	Speed: 11240.449428 samples/sec	accuracy=0.488662	lr=1.900952
Epoch[128] Batch [99]	Speed: 10710.495802 samples/sec	accuracy=0.489141	lr=1.900136
Epoch[128] Batch [149]	Speed: 11442.064327 samples/sec	accuracy=0.486273	lr=1.899319
Epoch[128] Batch [199]	Speed: 11271.834809 samples/sec	accuracy=0.485745	lr=1.898502
Epoch[128] Batch [249]	Speed: 10981.299168 samples/sec	accuracy=0.484602	lr=1.897685
Epoch[128] Batch [299]	Speed: 10951.229776 samples/sec	accuracy=0.484012	lr=1.896868
Epoch[128] Batch [349]	Speed: 10918.095525 samples/sec	accuracy=0.483408	lr=1.896050
Epoch[128] Batch [399]	Speed: 11392.406721 samples/sec	accuracy=0.483044	lr=1.895232
Epoch[128] Batch [449]	Speed: 11427.557583 samples/sec	accuracy=0.482571	lr=1.894413
Epoch[128] Batch [499]	Speed: 10978.076055 samples/sec	accuracy=0.482327	lr=1.893595
Epoch[128] Batch [549]	Speed: 10980.411451 samples/sec	accuracy=0.481877	lr=1.892776
Epoch[128] Batch [599]	Speed: 11117.511793 samples/sec	accuracy=0.481536	lr=1.891957
[Epoch 128] training: accuracy=0.481406
[Epoch 128] speed: 11048 samples/sec	time cost: 120.829265
[Epoch 128] validation: err-top1=0.505762 err-top5=0.250586
Epoch[129] Batch [49]	Speed: 10832.220206 samples/sec	accuracy=0.487217	lr=1.890711
Epoch[129] Batch [99]	Speed: 10665.335503 samples/sec	accuracy=0.485068	lr=1.889891
Epoch[129] Batch [149]	Speed: 11012.763200 samples/sec	accuracy=0.482868	lr=1.889070
Epoch[129] Batch [199]	Speed: 11291.286617 samples/sec	accuracy=0.483237	lr=1.888250
Epoch[129] Batch [249]	Speed: 10841.672850 samples/sec	accuracy=0.483494	lr=1.887429
Epoch[129] Batch [299]	Speed: 11168.923745 samples/sec	accuracy=0.483008	lr=1.886608
Epoch[129] Batch [349]	Speed: 10983.546614 samples/sec	accuracy=0.482540	lr=1.885786
Epoch[129] Batch [399]	Speed: 11105.624384 samples/sec	accuracy=0.482208	lr=1.884965
Epoch[129] Batch [449]	Speed: 10593.589865 samples/sec	accuracy=0.481898	lr=1.884142
Epoch[129] Batch [499]	Speed: 11183.571445 samples/sec	accuracy=0.481974	lr=1.883320
Epoch[129] Batch [549]	Speed: 11284.766366 samples/sec	accuracy=0.481760	lr=1.882497
Epoch[129] Batch [599]	Speed: 10702.051438 samples/sec	accuracy=0.481549	lr=1.881675
[Epoch 129] training: accuracy=0.481475
[Epoch 129] speed: 10919 samples/sec	time cost: 121.738183
[Epoch 129] validation: err-top1=0.511495 err-top5=0.258036
Epoch[130] Batch [49]	Speed: 10987.526481 samples/sec	accuracy=0.489092	lr=1.880423
Epoch[130] Batch [99]	Speed: 11133.799309 samples/sec	accuracy=0.488350	lr=1.879599
Epoch[130] Batch [149]	Speed: 11223.025947 samples/sec	accuracy=0.487507	lr=1.878775
Epoch[130] Batch [199]	Speed: 10904.099947 samples/sec	accuracy=0.486575	lr=1.877951
Epoch[130] Batch [249]	Speed: 10962.204268 samples/sec	accuracy=0.485494	lr=1.877127
Epoch[130] Batch [299]	Speed: 11289.374381 samples/sec	accuracy=0.484855	lr=1.876302
Epoch[130] Batch [349]	Speed: 11352.308359 samples/sec	accuracy=0.484311	lr=1.875477
Epoch[130] Batch [399]	Speed: 11710.872833 samples/sec	accuracy=0.484011	lr=1.874651
Epoch[130] Batch [449]	Speed: 11666.950748 samples/sec	accuracy=0.483689	lr=1.873826
Epoch[130] Batch [499]	Speed: 11502.762910 samples/sec	accuracy=0.483244	lr=1.873000
Epoch[130] Batch [549]	Speed: 10933.349327 samples/sec	accuracy=0.482828	lr=1.872173
Epoch[130] Batch [599]	Speed: 10602.909989 samples/sec	accuracy=0.482565	lr=1.871347
[Epoch 130] training: accuracy=0.482495
[Epoch 130] speed: 11138 samples/sec	time cost: 119.448565
[Epoch 130] validation: err-top1=0.508398 err-top5=0.254531
Epoch[131] Batch [49]	Speed: 10983.901381 samples/sec	accuracy=0.491182	lr=1.870106
Epoch[131] Batch [99]	Speed: 11065.181360 samples/sec	accuracy=0.489556	lr=1.869279
Epoch[131] Batch [149]	Speed: 11254.995824 samples/sec	accuracy=0.488083	lr=1.868452
Epoch[131] Batch [199]	Speed: 11222.796032 samples/sec	accuracy=0.486550	lr=1.867624
Epoch[131] Batch [249]	Speed: 11448.897068 samples/sec	accuracy=0.485318	lr=1.866796
Epoch[131] Batch [299]	Speed: 11383.298636 samples/sec	accuracy=0.484295	lr=1.865967
Epoch[131] Batch [349]	Speed: 11088.552502 samples/sec	accuracy=0.483714	lr=1.865138
Epoch[131] Batch [399]	Speed: 11218.196487 samples/sec	accuracy=0.483275	lr=1.864309
Epoch[131] Batch [449]	Speed: 10577.123520 samples/sec	accuracy=0.482936	lr=1.863480
Epoch[131] Batch [499]	Speed: 11270.667619 samples/sec	accuracy=0.482544	lr=1.862651
Epoch[131] Batch [549]	Speed: 11278.172791 samples/sec	accuracy=0.482697	lr=1.861821
Epoch[131] Batch [599]	Speed: 11127.190942 samples/sec	accuracy=0.482642	lr=1.860991
[Epoch 131] training: accuracy=0.482413
[Epoch 131] speed: 11151 samples/sec	time cost: 119.272235
[Epoch 131] validation: err-top1=0.485616 err-top5=0.235250
Epoch[132] Batch [49]	Speed: 11189.077668 samples/sec	accuracy=0.489473	lr=1.859728
Epoch[132] Batch [99]	Speed: 10576.660405 samples/sec	accuracy=0.490762	lr=1.858898
Epoch[132] Batch [149]	Speed: 10148.397390 samples/sec	accuracy=0.490244	lr=1.858066
Epoch[132] Batch [199]	Speed: 11188.836608 samples/sec	accuracy=0.488674	lr=1.857235
Epoch[132] Batch [249]	Speed: 10370.077932 samples/sec	accuracy=0.487516	lr=1.856403
Epoch[132] Batch [299]	Speed: 11166.662223 samples/sec	accuracy=0.486413	lr=1.855571
Epoch[132] Batch [349]	Speed: 11304.606822 samples/sec	accuracy=0.485508	lr=1.854739
Epoch[132] Batch [399]	Speed: 11494.376681 samples/sec	accuracy=0.485189	lr=1.853907
Epoch[132] Batch [449]	Speed: 11284.404351 samples/sec	accuracy=0.484714	lr=1.853074
Epoch[132] Batch [499]	Speed: 11623.877201 samples/sec	accuracy=0.484109	lr=1.852241
Epoch[132] Batch [549]	Speed: 11038.788764 samples/sec	accuracy=0.483732	lr=1.851408
Epoch[132] Batch [599]	Speed: 11382.515475 samples/sec	accuracy=0.483636	lr=1.850574
[Epoch 132] training: accuracy=0.483682
[Epoch 132] speed: 11002 samples/sec	time cost: 120.911490
[Epoch 132] validation: err-top1=0.570000 err-top5=0.311816
Epoch[133] Batch [49]	Speed: 11212.489456 samples/sec	accuracy=0.491816	lr=1.849323
Epoch[133] Batch [99]	Speed: 10909.619492 samples/sec	accuracy=0.490562	lr=1.848489
Epoch[133] Batch [149]	Speed: 11464.272001 samples/sec	accuracy=0.488548	lr=1.847654
Epoch[133] Batch [199]	Speed: 11067.620133 samples/sec	accuracy=0.487776	lr=1.846819
Epoch[133] Batch [249]	Speed: 11081.422152 samples/sec	accuracy=0.486914	lr=1.845984
Epoch[133] Batch [299]	Speed: 11144.524083 samples/sec	accuracy=0.486130	lr=1.845149
Epoch[133] Batch [349]	Speed: 11302.248683 samples/sec	accuracy=0.485865	lr=1.844313
Epoch[133] Batch [399]	Speed: 11336.510297 samples/sec	accuracy=0.485223	lr=1.843477
Epoch[133] Batch [449]	Speed: 11025.835632 samples/sec	accuracy=0.485042	lr=1.842641
Epoch[133] Batch [499]	Speed: 11502.957919 samples/sec	accuracy=0.484454	lr=1.841805
Epoch[133] Batch [549]	Speed: 11269.071333 samples/sec	accuracy=0.484180	lr=1.840968
Epoch[133] Batch [599]	Speed: 10718.197832 samples/sec	accuracy=0.483918	lr=1.840131
[Epoch 133] training: accuracy=0.483896
[Epoch 133] speed: 11086 samples/sec	time cost: 120.004037
[Epoch 133] validation: err-top1=0.540161 err-top5=0.286397
Epoch[134] Batch [49]	Speed: 11080.319217 samples/sec	accuracy=0.492480	lr=1.838858
Epoch[134] Batch [99]	Speed: 11206.550499 samples/sec	accuracy=0.490000	lr=1.838020
Epoch[134] Batch [149]	Speed: 11352.724858 samples/sec	accuracy=0.488529	lr=1.837182
Epoch[134] Batch [199]	Speed: 11327.263094 samples/sec	accuracy=0.487080	lr=1.836344
Epoch[134] Batch [249]	Speed: 10909.485648 samples/sec	accuracy=0.486459	lr=1.835506
Epoch[134] Batch [299]	Speed: 11295.961159 samples/sec	accuracy=0.486014	lr=1.834667
Epoch[134] Batch [349]	Speed: 11485.608586 samples/sec	accuracy=0.485720	lr=1.833828
Epoch[134] Batch [399]	Speed: 11028.799679 samples/sec	accuracy=0.485353	lr=1.832988
Epoch[134] Batch [449]	Speed: 11470.143020 samples/sec	accuracy=0.485003	lr=1.832149
Epoch[134] Batch [499]	Speed: 11056.920678 samples/sec	accuracy=0.484659	lr=1.831309
Epoch[134] Batch [549]	Speed: 11306.661438 samples/sec	accuracy=0.484094	lr=1.830469
Epoch[134] Batch [599]	Speed: 11244.326241 samples/sec	accuracy=0.483590	lr=1.829629
[Epoch 134] training: accuracy=0.483450
[Epoch 134] speed: 11208 samples/sec	time cost: 118.649016
[Epoch 134] validation: err-top1=0.557699 err-top5=0.307129
Epoch[135] Batch [49]	Speed: 10946.455866 samples/sec	accuracy=0.490127	lr=1.828367
Epoch[135] Batch [99]	Speed: 11112.875346 samples/sec	accuracy=0.489287	lr=1.827526
Epoch[135] Batch [149]	Speed: 11116.081728 samples/sec	accuracy=0.488496	lr=1.826685
Epoch[135] Batch [199]	Speed: 11577.731747 samples/sec	accuracy=0.488127	lr=1.825844
Epoch[135] Batch [249]	Speed: 11114.141801 samples/sec	accuracy=0.487613	lr=1.825002
Epoch[135] Batch [299]	Speed: 11164.806474 samples/sec	accuracy=0.486951	lr=1.824160
Epoch[135] Batch [349]	Speed: 11210.466578 samples/sec	accuracy=0.486420	lr=1.823317
Epoch[135] Batch [399]	Speed: 11390.504792 samples/sec	accuracy=0.485968	lr=1.822475
Epoch[135] Batch [449]	Speed: 11415.395108 samples/sec	accuracy=0.485400	lr=1.821632
Epoch[135] Batch [499]	Speed: 11102.871762 samples/sec	accuracy=0.485073	lr=1.820789
Epoch[135] Batch [549]	Speed: 11280.093677 samples/sec	accuracy=0.484907	lr=1.819945
Epoch[135] Batch [599]	Speed: 11238.929625 samples/sec	accuracy=0.484971	lr=1.819102
[Epoch 135] training: accuracy=0.484789
[Epoch 135] speed: 11155 samples/sec	time cost: 119.543791
[Epoch 135] validation: err-top1=0.519355 err-top5=0.265586
Epoch[136] Batch [49]	Speed: 11335.557942 samples/sec	accuracy=0.495283	lr=1.817819
Epoch[136] Batch [99]	Speed: 11549.467042 samples/sec	accuracy=0.491782	lr=1.816974
Epoch[136] Batch [149]	Speed: 10853.481025 samples/sec	accuracy=0.490260	lr=1.816130
Epoch[136] Batch [199]	Speed: 10587.021396 samples/sec	accuracy=0.489551	lr=1.815285
Epoch[136] Batch [249]	Speed: 10689.218682 samples/sec	accuracy=0.488063	lr=1.814440
Epoch[136] Batch [299]	Speed: 11442.222838 samples/sec	accuracy=0.487770	lr=1.813594
Epoch[136] Batch [349]	Speed: 11278.911743 samples/sec	accuracy=0.487104	lr=1.812749
Epoch[136] Batch [399]	Speed: 11795.812238 samples/sec	accuracy=0.486438	lr=1.811903
Epoch[136] Batch [449]	Speed: 10640.128628 samples/sec	accuracy=0.486045	lr=1.811057
Epoch[136] Batch [499]	Speed: 11330.464077 samples/sec	accuracy=0.485968	lr=1.810210
Epoch[136] Batch [549]	Speed: 11165.884202 samples/sec	accuracy=0.485741	lr=1.809364
Epoch[136] Batch [599]	Speed: 10969.258173 samples/sec	accuracy=0.485443	lr=1.808517
[Epoch 136] training: accuracy=0.485383
[Epoch 136] speed: 11087 samples/sec	time cost: 120.045043
[Epoch 136] validation: err-top1=0.516093 err-top5=0.261007
Epoch[137] Batch [49]	Speed: 10160.247114 samples/sec	accuracy=0.492236	lr=1.807229
Epoch[137] Batch [99]	Speed: 11274.391580 samples/sec	accuracy=0.491816	lr=1.806382
Epoch[137] Batch [149]	Speed: 11183.037689 samples/sec	accuracy=0.490768	lr=1.805534
Epoch[137] Batch [199]	Speed: 11042.117182 samples/sec	accuracy=0.490337	lr=1.804686
Epoch[137] Batch [249]	Speed: 10289.384978 samples/sec	accuracy=0.489459	lr=1.803837
Epoch[137] Batch [299]	Speed: 9939.306122 samples/sec	accuracy=0.488890	lr=1.802989
Epoch[137] Batch [349]	Speed: 11463.010468 samples/sec	accuracy=0.488073	lr=1.802140
Epoch[137] Batch [399]	Speed: 11791.833691 samples/sec	accuracy=0.487478	lr=1.801291
Epoch[137] Batch [449]	Speed: 11201.668887 samples/sec	accuracy=0.486861	lr=1.800442
Epoch[137] Batch [499]	Speed: 11294.945504 samples/sec	accuracy=0.486631	lr=1.799592
Epoch[137] Batch [549]	Speed: 11567.485686 samples/sec	accuracy=0.486194	lr=1.798742
Epoch[137] Batch [599]	Speed: 11434.715176 samples/sec	accuracy=0.485872	lr=1.797892
[Epoch 137] training: accuracy=0.485657
[Epoch 137] speed: 11018 samples/sec	time cost: 120.750301
[Epoch 137] validation: err-top1=0.533281 err-top5=0.277246
Epoch[138] Batch [49]	Speed: 11430.344589 samples/sec	accuracy=0.493779	lr=1.796617
Epoch[138] Batch [99]	Speed: 11384.571654 samples/sec	accuracy=0.493257	lr=1.795766
Epoch[138] Batch [149]	Speed: 11327.525093 samples/sec	accuracy=0.492035	lr=1.794915
Epoch[138] Batch [199]	Speed: 11571.714559 samples/sec	accuracy=0.490867	lr=1.794064
Epoch[138] Batch [249]	Speed: 11507.632421 samples/sec	accuracy=0.490568	lr=1.793213
Epoch[138] Batch [299]	Speed: 11235.075047 samples/sec	accuracy=0.489725	lr=1.792361
Epoch[138] Batch [349]	Speed: 11461.161354 samples/sec	accuracy=0.488933	lr=1.791509
Epoch[138] Batch [399]	Speed: 11112.667749 samples/sec	accuracy=0.488248	lr=1.790657
Epoch[138] Batch [449]	Speed: 11223.621599 samples/sec	accuracy=0.487457	lr=1.789804
Epoch[138] Batch [499]	Speed: 11239.346375 samples/sec	accuracy=0.487134	lr=1.788952
Epoch[138] Batch [549]	Speed: 11232.431496 samples/sec	accuracy=0.486719	lr=1.788099
Epoch[138] Batch [599]	Speed: 10403.464431 samples/sec	accuracy=0.486524	lr=1.787246
[Epoch 138] training: accuracy=0.486327
[Epoch 138] speed: 11216 samples/sec	time cost: 118.743797
[Epoch 138] validation: err-top1=0.507385 err-top5=0.255208
Epoch[139] Batch [49]	Speed: 10338.017713 samples/sec	accuracy=0.492998	lr=1.785948
Epoch[139] Batch [99]	Speed: 11298.548506 samples/sec	accuracy=0.493247	lr=1.785095
Epoch[139] Batch [149]	Speed: 11562.116242 samples/sec	accuracy=0.491113	lr=1.784241
Epoch[139] Batch [199]	Speed: 10694.822899 samples/sec	accuracy=0.489990	lr=1.783386
Epoch[139] Batch [249]	Speed: 10715.668647 samples/sec	accuracy=0.489672	lr=1.782532
Epoch[139] Batch [299]	Speed: 10936.618910 samples/sec	accuracy=0.489424	lr=1.781677
Epoch[139] Batch [349]	Speed: 11193.598224 samples/sec	accuracy=0.489102	lr=1.780822
Epoch[139] Batch [399]	Speed: 11728.493857 samples/sec	accuracy=0.488695	lr=1.779967
Epoch[139] Batch [449]	Speed: 11174.741468 samples/sec	accuracy=0.488311	lr=1.779112
Epoch[139] Batch [499]	Speed: 11255.431170 samples/sec	accuracy=0.487754	lr=1.778256
Epoch[139] Batch [549]	Speed: 11281.021920 samples/sec	accuracy=0.487259	lr=1.777400
Epoch[139] Batch [599]	Speed: 11627.975525 samples/sec	accuracy=0.487294	lr=1.776544
[Epoch 139] training: accuracy=0.487110
[Epoch 139] speed: 11092 samples/sec	time cost: 119.800867
[Epoch 139] validation: err-top1=0.500549 err-top5=0.251160
Epoch[140] Batch [49]	Speed: 10352.103876 samples/sec	accuracy=0.495605	lr=1.775259
Epoch[140] Batch [99]	Speed: 11414.364839 samples/sec	accuracy=0.492837	lr=1.774402
Epoch[140] Batch [149]	Speed: 11551.611328 samples/sec	accuracy=0.491729	lr=1.773545
Epoch[140] Batch [199]	Speed: 10683.583662 samples/sec	accuracy=0.491660	lr=1.772688
Epoch[140] Batch [249]	Speed: 10774.478946 samples/sec	accuracy=0.490641	lr=1.771831
Epoch[140] Batch [299]	Speed: 11403.021741 samples/sec	accuracy=0.489631	lr=1.770973
Epoch[140] Batch [349]	Speed: 10884.388505 samples/sec	accuracy=0.488813	lr=1.770115
Epoch[140] Batch [399]	Speed: 11269.087299 samples/sec	accuracy=0.488723	lr=1.769257
Epoch[140] Batch [449]	Speed: 11354.751676 samples/sec	accuracy=0.487699	lr=1.768398
Epoch[140] Batch [499]	Speed: 11652.386970 samples/sec	accuracy=0.487091	lr=1.767540
Epoch[140] Batch [549]	Speed: 11859.467351 samples/sec	accuracy=0.486923	lr=1.766681
Epoch[140] Batch [599]	Speed: 11657.920028 samples/sec	accuracy=0.486687	lr=1.765822
[Epoch 140] training: accuracy=0.486627
[Epoch 140] speed: 11227 samples/sec	time cost: 118.925854
[Epoch 140] validation: err-top1=0.488379 err-top5=0.236523
Epoch[141] Batch [49]	Speed: 10316.448096 samples/sec	accuracy=0.493408	lr=1.764515
Epoch[141] Batch [99]	Speed: 11286.338631 samples/sec	accuracy=0.492505	lr=1.763656
Epoch[141] Batch [149]	Speed: 11596.030677 samples/sec	accuracy=0.491585	lr=1.762796
Epoch[141] Batch [199]	Speed: 10701.759442 samples/sec	accuracy=0.491411	lr=1.761935
Epoch[141] Batch [249]	Speed: 10161.275446 samples/sec	accuracy=0.490924	lr=1.761075
Epoch[141] Batch [299]	Speed: 11642.726372 samples/sec	accuracy=0.490627	lr=1.760214
Epoch[141] Batch [349]	Speed: 11414.410342 samples/sec	accuracy=0.490109	lr=1.759354
Epoch[141] Batch [399]	Speed: 11197.648596 samples/sec	accuracy=0.489722	lr=1.758492
Epoch[141] Batch [449]	Speed: 11178.558534 samples/sec	accuracy=0.489436	lr=1.757631
Epoch[141] Batch [499]	Speed: 10716.177437 samples/sec	accuracy=0.489280	lr=1.756770
Epoch[141] Batch [549]	Speed: 11301.068645 samples/sec	accuracy=0.488934	lr=1.755908
Epoch[141] Batch [599]	Speed: 11627.761773 samples/sec	accuracy=0.488542	lr=1.755046
[Epoch 141] training: accuracy=0.488531
[Epoch 141] speed: 11033 samples/sec	time cost: 120.434578
[Epoch 141] validation: err-top1=0.536682 err-top5=0.278524
Epoch[142] Batch [49]	Speed: 10829.056423 samples/sec	accuracy=0.495029	lr=1.753752
Epoch[142] Batch [99]	Speed: 11427.918199 samples/sec	accuracy=0.494175	lr=1.752890
Epoch[142] Batch [149]	Speed: 10797.306038 samples/sec	accuracy=0.492269	lr=1.752027
Epoch[142] Batch [199]	Speed: 11245.789790 samples/sec	accuracy=0.491494	lr=1.751164
Epoch[142] Batch [249]	Speed: 11409.506335 samples/sec	accuracy=0.491014	lr=1.750301
Epoch[142] Batch [299]	Speed: 11466.026615 samples/sec	accuracy=0.490866	lr=1.749437
Epoch[142] Batch [349]	Speed: 11388.198253 samples/sec	accuracy=0.490633	lr=1.748573
Epoch[142] Batch [399]	Speed: 11341.403253 samples/sec	accuracy=0.490383	lr=1.747709
Epoch[142] Batch [449]	Speed: 10871.905615 samples/sec	accuracy=0.489741	lr=1.746845
Epoch[142] Batch [499]	Speed: 11254.283299 samples/sec	accuracy=0.489185	lr=1.745981
Epoch[142] Batch [549]	Speed: 11508.149818 samples/sec	accuracy=0.488671	lr=1.745116
Epoch[142] Batch [599]	Speed: 11331.784502 samples/sec	accuracy=0.488555	lr=1.744251
[Epoch 142] training: accuracy=0.488721
[Epoch 142] speed: 11211 samples/sec	time cost: 118.979390
[Epoch 142] validation: err-top1=0.540586 err-top5=0.284766
Epoch[143] Batch [49]	Speed: 10564.470983 samples/sec	accuracy=0.496016	lr=1.742936
Epoch[143] Batch [99]	Speed: 11669.110033 samples/sec	accuracy=0.493994	lr=1.742071
Epoch[143] Batch [149]	Speed: 11369.326546 samples/sec	accuracy=0.492682	lr=1.741205
Epoch[143] Batch [199]	Speed: 11198.848598 samples/sec	accuracy=0.491982	lr=1.740340
Epoch[143] Batch [249]	Speed: 11287.430755 samples/sec	accuracy=0.491336	lr=1.739473
Epoch[143] Batch [299]	Speed: 11129.542056 samples/sec	accuracy=0.491134	lr=1.738607
Epoch[143] Batch [349]	Speed: 11135.571720 samples/sec	accuracy=0.491098	lr=1.737741
Epoch[143] Batch [399]	Speed: 11836.978882 samples/sec	accuracy=0.490428	lr=1.736874
Epoch[143] Batch [449]	Speed: 10338.203597 samples/sec	accuracy=0.490017	lr=1.736007
Epoch[143] Batch [499]	Speed: 11465.071656 samples/sec	accuracy=0.489550	lr=1.735140
Epoch[143] Batch [549]	Speed: 10455.847240 samples/sec	accuracy=0.489103	lr=1.734273
Epoch[143] Batch [599]	Speed: 11016.708886 samples/sec	accuracy=0.488632	lr=1.733405
[Epoch 143] training: accuracy=0.488524
[Epoch 143] speed: 11046 samples/sec	time cost: 120.416731
[Epoch 143] validation: err-top1=0.524373 err-top5=0.270487
Epoch[144] Batch [49]	Speed: 11052.153169 samples/sec	accuracy=0.496309	lr=1.732086
Epoch[144] Batch [99]	Speed: 11248.747198 samples/sec	accuracy=0.496777	lr=1.731218
Epoch[144] Batch [149]	Speed: 11299.915015 samples/sec	accuracy=0.495150	lr=1.730349
Epoch[144] Batch [199]	Speed: 11232.956170 samples/sec	accuracy=0.493657	lr=1.729481
Epoch[144] Batch [249]	Speed: 10671.745651 samples/sec	accuracy=0.492840	lr=1.728612
Epoch[144] Batch [299]	Speed: 11370.135887 samples/sec	accuracy=0.491595	lr=1.727743
Epoch[144] Batch [349]	Speed: 11181.644281 samples/sec	accuracy=0.491183	lr=1.726874
Epoch[144] Batch [399]	Speed: 10887.193635 samples/sec	accuracy=0.490955	lr=1.726004
Epoch[144] Batch [449]	Speed: 11475.026028 samples/sec	accuracy=0.490277	lr=1.725135
Epoch[144] Batch [499]	Speed: 11522.376077 samples/sec	accuracy=0.489890	lr=1.724265
Epoch[144] Batch [549]	Speed: 10471.923817 samples/sec	accuracy=0.489598	lr=1.723395
Epoch[144] Batch [599]	Speed: 11087.900682 samples/sec	accuracy=0.489220	lr=1.722524
[Epoch 144] training: accuracy=0.489035
[Epoch 144] speed: 11095 samples/sec	time cost: 120.000833
[Epoch 144] validation: err-top1=0.533223 err-top5=0.279473
Epoch[145] Batch [49]	Speed: 10894.544260 samples/sec	accuracy=0.496836	lr=1.721219
Epoch[145] Batch [99]	Speed: 9748.166062 samples/sec	accuracy=0.496836	lr=1.720348
Epoch[145] Batch [149]	Speed: 11423.179704 samples/sec	accuracy=0.494896	lr=1.719477
Epoch[145] Batch [199]	Speed: 11070.508531 samples/sec	accuracy=0.493035	lr=1.718605
Epoch[145] Batch [249]	Speed: 11167.413639 samples/sec	accuracy=0.492477	lr=1.717734
Epoch[145] Batch [299]	Speed: 11335.250398 samples/sec	accuracy=0.491947	lr=1.716862
Epoch[145] Batch [349]	Speed: 11262.943229 samples/sec	accuracy=0.491482	lr=1.715991
Epoch[145] Batch [399]	Speed: 11277.689488 samples/sec	accuracy=0.490968	lr=1.715118
Epoch[145] Batch [449]	Speed: 11051.580696 samples/sec	accuracy=0.490429	lr=1.714246
Epoch[145] Batch [499]	Speed: 11225.431234 samples/sec	accuracy=0.490167	lr=1.713374
Epoch[145] Batch [549]	Speed: 11623.144886 samples/sec	accuracy=0.489940	lr=1.712501
Epoch[145] Batch [599]	Speed: 10825.231708 samples/sec	accuracy=0.489407	lr=1.711628
[Epoch 145] training: accuracy=0.489319
[Epoch 145] speed: 11040 samples/sec	time cost: 120.769204
[Epoch 145] validation: err-top1=0.495463 err-top5=0.245056
Epoch[146] Batch [49]	Speed: 10805.093512 samples/sec	accuracy=0.497773	lr=1.710301
Epoch[146] Batch [99]	Speed: 10361.921629 samples/sec	accuracy=0.496704	lr=1.709427
Epoch[146] Batch [149]	Speed: 11874.670582 samples/sec	accuracy=0.495238	lr=1.708554
Epoch[146] Batch [199]	Speed: 11039.275641 samples/sec	accuracy=0.494800	lr=1.707680
Epoch[146] Batch [249]	Speed: 11015.316218 samples/sec	accuracy=0.493244	lr=1.706806
Epoch[146] Batch [299]	Speed: 11375.444754 samples/sec	accuracy=0.492980	lr=1.705932
Epoch[146] Batch [349]	Speed: 10610.113588 samples/sec	accuracy=0.492384	lr=1.705057
Epoch[146] Batch [399]	Speed: 11334.255482 samples/sec	accuracy=0.491957	lr=1.704183
Epoch[146] Batch [449]	Speed: 10976.816572 samples/sec	accuracy=0.491587	lr=1.703308
Epoch[146] Batch [499]	Speed: 11320.484909 samples/sec	accuracy=0.491398	lr=1.702433
Epoch[146] Batch [549]	Speed: 11206.590851 samples/sec	accuracy=0.491035	lr=1.701558
Epoch[146] Batch [599]	Speed: 11084.960003 samples/sec	accuracy=0.490956	lr=1.700682
[Epoch 146] training: accuracy=0.490727
[Epoch 146] speed: 11009 samples/sec	time cost: 120.858106
[Epoch 146] validation: err-top1=0.502828 err-top5=0.251628
Epoch[147] Batch [49]	Speed: 10794.284419 samples/sec	accuracy=0.497773	lr=1.699369
Epoch[147] Batch [99]	Speed: 11193.759844 samples/sec	accuracy=0.496284	lr=1.698493
Epoch[147] Batch [149]	Speed: 10868.429000 samples/sec	accuracy=0.495257	lr=1.697616
Epoch[147] Batch [199]	Speed: 11067.031513 samples/sec	accuracy=0.493606	lr=1.696740
Epoch[147] Batch [249]	Speed: 10583.308859 samples/sec	accuracy=0.493346	lr=1.695864
Epoch[147] Batch [299]	Speed: 10904.689913 samples/sec	accuracy=0.493008	lr=1.694987
Epoch[147] Batch [349]	Speed: 10999.613553 samples/sec	accuracy=0.492807	lr=1.694110
Epoch[147] Batch [399]	Speed: 11022.316745 samples/sec	accuracy=0.492593	lr=1.693233
Epoch[147] Batch [449]	Speed: 11411.765724 samples/sec	accuracy=0.491998	lr=1.692355
Epoch[147] Batch [499]	Speed: 11347.081317 samples/sec	accuracy=0.491847	lr=1.691478
Epoch[147] Batch [549]	Speed: 11664.464377 samples/sec	accuracy=0.491459	lr=1.690600
Epoch[147] Batch [599]	Speed: 11207.802715 samples/sec	accuracy=0.491309	lr=1.689722
[Epoch 147] training: accuracy=0.491171
[Epoch 147] speed: 11052 samples/sec	time cost: 120.688131
[Epoch 147] validation: err-top1=0.496465 err-top5=0.245313
Epoch[148] Batch [49]	Speed: 10806.473225 samples/sec	accuracy=0.502227	lr=1.688387
Epoch[148] Batch [99]	Speed: 10734.598798 samples/sec	accuracy=0.498872	lr=1.687509
Epoch[148] Batch [149]	Speed: 11299.425090 samples/sec	accuracy=0.498558	lr=1.686630
Epoch[148] Batch [199]	Speed: 10862.507344 samples/sec	accuracy=0.496597	lr=1.685751
Epoch[148] Batch [249]	Speed: 11502.907703 samples/sec	accuracy=0.495840	lr=1.684872
Epoch[148] Batch [299]	Speed: 10552.600781 samples/sec	accuracy=0.494943	lr=1.683993
Epoch[148] Batch [349]	Speed: 11345.729752 samples/sec	accuracy=0.494042	lr=1.683114
Epoch[148] Batch [399]	Speed: 11020.404315 samples/sec	accuracy=0.493489	lr=1.682234
Epoch[148] Batch [449]	Speed: 11398.589128 samples/sec	accuracy=0.493089	lr=1.681355
Epoch[148] Batch [499]	Speed: 11065.098975 samples/sec	accuracy=0.492750	lr=1.680475
Epoch[148] Batch [549]	Speed: 11432.182251 samples/sec	accuracy=0.492393	lr=1.679594
Epoch[148] Batch [599]	Speed: 11487.098141 samples/sec	accuracy=0.491997	lr=1.678714
[Epoch 148] training: accuracy=0.491683
[Epoch 148] speed: 11110 samples/sec	time cost: 119.552632
[Epoch 148] validation: err-top1=0.494141 err-top5=0.240295
Epoch[149] Batch [49]	Speed: 10616.186696 samples/sec	accuracy=0.499951	lr=1.677393
Epoch[149] Batch [99]	Speed: 11260.817962 samples/sec	accuracy=0.497876	lr=1.676512
Epoch[149] Batch [149]	Speed: 11581.813206 samples/sec	accuracy=0.496318	lr=1.675631
Epoch[149] Batch [199]	Speed: 11475.446675 samples/sec	accuracy=0.495732	lr=1.674750
Epoch[149] Batch [249]	Speed: 10575.651228 samples/sec	accuracy=0.495266	lr=1.673869
Epoch[149] Batch [299]	Speed: 11001.967705 samples/sec	accuracy=0.494453	lr=1.672987
Epoch[149] Batch [349]	Speed: 11336.043524 samples/sec	accuracy=0.494136	lr=1.672105
Epoch[149] Batch [399]	Speed: 11464.903331 samples/sec	accuracy=0.493628	lr=1.671224
Epoch[149] Batch [449]	Speed: 10890.749649 samples/sec	accuracy=0.493340	lr=1.670341
Epoch[149] Batch [499]	Speed: 11435.893755 samples/sec	accuracy=0.492622	lr=1.669459
Epoch[149] Batch [549]	Speed: 11254.596197 samples/sec	accuracy=0.492257	lr=1.668577
Epoch[149] Batch [599]	Speed: 11412.340642 samples/sec	accuracy=0.492075	lr=1.667694
[Epoch 149] training: accuracy=0.491813
[Epoch 149] speed: 11149 samples/sec	time cost: 119.479653
[Epoch 149] validation: err-top1=0.483789 err-top5=0.234141
Epoch[150] Batch [49]	Speed: 10745.664194 samples/sec	accuracy=0.500830	lr=1.666352
Epoch[150] Batch [99]	Speed: 10457.035830 samples/sec	accuracy=0.498477	lr=1.665469
Epoch[150] Batch [149]	Speed: 10765.265290 samples/sec	accuracy=0.497220	lr=1.664585
Epoch[150] Batch [199]	Speed: 10847.666601 samples/sec	accuracy=0.496182	lr=1.663702
Epoch[150] Batch [249]	Speed: 10782.788169 samples/sec	accuracy=0.495338	lr=1.662818
Epoch[150] Batch [299]	Speed: 10899.184758 samples/sec	accuracy=0.494788	lr=1.661934
Epoch[150] Batch [349]	Speed: 11186.662292 samples/sec	accuracy=0.494170	lr=1.661050
Epoch[150] Batch [399]	Speed: 11281.373643 samples/sec	accuracy=0.493650	lr=1.660166
Epoch[150] Batch [449]	Speed: 11514.713506 samples/sec	accuracy=0.493152	lr=1.659282
Epoch[150] Batch [499]	Speed: 11531.936676 samples/sec	accuracy=0.493009	lr=1.658397
Epoch[150] Batch [549]	Speed: 11463.899601 samples/sec	accuracy=0.492993	lr=1.657512
Epoch[150] Batch [599]	Speed: 11743.096431 samples/sec	accuracy=0.492542	lr=1.656627
[Epoch 150] training: accuracy=0.492429
[Epoch 150] speed: 11065 samples/sec	time cost: 120.129201
[Epoch 150] validation: err-top1=0.486694 err-top5=0.234416
Epoch[151] Batch [49]	Speed: 10446.989325 samples/sec	accuracy=0.499102	lr=1.655282
Epoch[151] Batch [99]	Speed: 10728.321964 samples/sec	accuracy=0.499370	lr=1.654396
Epoch[151] Batch [149]	Speed: 10848.304729 samples/sec	accuracy=0.499040	lr=1.653511
Epoch[151] Batch [199]	Speed: 11248.765759 samples/sec	accuracy=0.498203	lr=1.652625
Epoch[151] Batch [249]	Speed: 11247.692002 samples/sec	accuracy=0.497543	lr=1.651739
Epoch[151] Batch [299]	Speed: 11458.809298 samples/sec	accuracy=0.496086	lr=1.650853
Epoch[151] Batch [349]	Speed: 10859.970839 samples/sec	accuracy=0.495541	lr=1.649967
Epoch[151] Batch [399]	Speed: 10662.875940 samples/sec	accuracy=0.495167	lr=1.649080
Epoch[151] Batch [449]	Speed: 11394.175375 samples/sec	accuracy=0.494680	lr=1.648194
Epoch[151] Batch [499]	Speed: 10846.725575 samples/sec	accuracy=0.493911	lr=1.647307
Epoch[151] Batch [549]	Speed: 11144.484177 samples/sec	accuracy=0.493540	lr=1.646420
Epoch[151] Batch [599]	Speed: 10867.981276 samples/sec	accuracy=0.493253	lr=1.645533
[Epoch 151] training: accuracy=0.493091
[Epoch 151] speed: 10954 samples/sec	time cost: 121.331170
[Epoch 151] validation: err-top1=0.506348 err-top5=0.258158
Epoch[152] Batch [49]	Speed: 11220.465449 samples/sec	accuracy=0.498789	lr=1.644201
Epoch[152] Batch [99]	Speed: 11215.747730 samples/sec	accuracy=0.499443	lr=1.643314
Epoch[152] Batch [149]	Speed: 11721.387714 samples/sec	accuracy=0.498079	lr=1.642426
Epoch[152] Batch [199]	Speed: 11234.402949 samples/sec	accuracy=0.497637	lr=1.641538
Epoch[152] Batch [249]	Speed: 11826.632463 samples/sec	accuracy=0.496344	lr=1.640650
Epoch[152] Batch [299]	Speed: 11616.354296 samples/sec	accuracy=0.496064	lr=1.639762
Epoch[152] Batch [349]	Speed: 10890.429869 samples/sec	accuracy=0.495312	lr=1.638873
Epoch[152] Batch [399]	Speed: 11435.598707 samples/sec	accuracy=0.495110	lr=1.637985
Epoch[152] Batch [449]	Speed: 11002.044644 samples/sec	accuracy=0.494455	lr=1.637096
Epoch[152] Batch [499]	Speed: 11040.875318 samples/sec	accuracy=0.494130	lr=1.636207
Epoch[152] Batch [549]	Speed: 11483.390789 samples/sec	accuracy=0.493826	lr=1.635318
Epoch[152] Batch [599]	Speed: 11506.975413 samples/sec	accuracy=0.493594	lr=1.634429
[Epoch 152] training: accuracy=0.493436
[Epoch 152] speed: 11296 samples/sec	time cost: 118.333099
[Epoch 152] validation: err-top1=0.501523 err-top5=0.250391
Epoch[153] Batch [49]	Speed: 10361.040743 samples/sec	accuracy=0.502471	lr=1.633076
Epoch[153] Batch [99]	Speed: 11036.547300 samples/sec	accuracy=0.500469	lr=1.632187
Epoch[153] Batch [149]	Speed: 11417.509318 samples/sec	accuracy=0.498874	lr=1.631297
Epoch[153] Batch [199]	Speed: 11182.930245 samples/sec	accuracy=0.498406	lr=1.630407
Epoch[153] Batch [249]	Speed: 11130.767313 samples/sec	accuracy=0.496703	lr=1.629517
Epoch[153] Batch [299]	Speed: 11412.707273 samples/sec	accuracy=0.496242	lr=1.628626
Epoch[153] Batch [349]	Speed: 11114.168261 samples/sec	accuracy=0.495508	lr=1.627736
Epoch[153] Batch [399]	Speed: 11400.236240 samples/sec	accuracy=0.494888	lr=1.626845
Epoch[153] Batch [449]	Speed: 11495.462058 samples/sec	accuracy=0.494642	lr=1.625954
Epoch[153] Batch [499]	Speed: 10356.030991 samples/sec	accuracy=0.494320	lr=1.625063
Epoch[153] Batch [549]	Speed: 11467.484151 samples/sec	accuracy=0.493903	lr=1.624172
Epoch[153] Batch [599]	Speed: 11253.056351 samples/sec	accuracy=0.493718	lr=1.623280
[Epoch 153] training: accuracy=0.493613
[Epoch 153] speed: 11122 samples/sec	time cost: 119.575924
[Epoch 153] validation: err-top1=0.488790 err-top5=0.240072
Epoch[154] Batch [49]	Speed: 11028.241516 samples/sec	accuracy=0.503174	lr=1.621943
Epoch[154] Batch [99]	Speed: 10887.386822 samples/sec	accuracy=0.501738	lr=1.621051
Epoch[154] Batch [149]	Speed: 11228.140470 samples/sec	accuracy=0.500225	lr=1.620159
Epoch[154] Batch [199]	Speed: 11403.374453 samples/sec	accuracy=0.498943	lr=1.619267
Epoch[154] Batch [249]	Speed: 10305.543321 samples/sec	accuracy=0.498168	lr=1.618375
Epoch[154] Batch [299]	Speed: 11234.602777 samples/sec	accuracy=0.497529	lr=1.617483
Epoch[154] Batch [349]	Speed: 11478.317830 samples/sec	accuracy=0.497012	lr=1.616590
Epoch[154] Batch [399]	Speed: 11145.802679 samples/sec	accuracy=0.496632	lr=1.615697
Epoch[154] Batch [449]	Speed: 11350.200527 samples/sec	accuracy=0.496067	lr=1.614804
Epoch[154] Batch [499]	Speed: 10594.896226 samples/sec	accuracy=0.495604	lr=1.613911
Epoch[154] Batch [549]	Speed: 10842.166306 samples/sec	accuracy=0.495381	lr=1.613018
Epoch[154] Batch [599]	Speed: 11310.391654 samples/sec	accuracy=0.494981	lr=1.612125
[Epoch 154] training: accuracy=0.494822
[Epoch 154] speed: 11001 samples/sec	time cost: 121.305102
[Epoch 154] validation: err-top1=0.497461 err-top5=0.247031
Epoch[155] Batch [49]	Speed: 11226.405963 samples/sec	accuracy=0.501777	lr=1.610767
Epoch[155] Batch [99]	Speed: 11345.608070 samples/sec	accuracy=0.500620	lr=1.609873
Epoch[155] Batch [149]	Speed: 9940.778423 samples/sec	accuracy=0.499762	lr=1.608979
Epoch[155] Batch [199]	Speed: 10757.813441 samples/sec	accuracy=0.498721	lr=1.608085
Epoch[155] Batch [249]	Speed: 11510.488550 samples/sec	accuracy=0.498340	lr=1.607191
Epoch[155] Batch [299]	Speed: 11240.065541 samples/sec	accuracy=0.497497	lr=1.606296
Epoch[155] Batch [349]	Speed: 10225.128563 samples/sec	accuracy=0.496921	lr=1.605402
Epoch[155] Batch [399]	Speed: 11330.538506 samples/sec	accuracy=0.496646	lr=1.604507
Epoch[155] Batch [449]	Speed: 11165.864463 samples/sec	accuracy=0.496514	lr=1.603612
Epoch[155] Batch [499]	Speed: 10754.418287 samples/sec	accuracy=0.496026	lr=1.602717
Epoch[155] Batch [549]	Speed: 10345.663401 samples/sec	accuracy=0.495733	lr=1.601822
Epoch[155] Batch [599]	Speed: 11620.704817 samples/sec	accuracy=0.495430	lr=1.600927
[Epoch 155] training: accuracy=0.495247
[Epoch 155] speed: 10910 samples/sec	time cost: 121.554592
[Epoch 155] validation: err-top1=0.536499 err-top5=0.277710
Epoch[156] Batch [49]	Speed: 11344.401281 samples/sec	accuracy=0.502773	lr=1.599584
Epoch[156] Batch [99]	Speed: 11348.598127 samples/sec	accuracy=0.502124	lr=1.598688
Epoch[156] Batch [149]	Speed: 11081.124813 samples/sec	accuracy=0.500247	lr=1.597792
Epoch[156] Batch [199]	Speed: 11329.950280 samples/sec	accuracy=0.499094	lr=1.596896
Epoch[156] Batch [249]	Speed: 11171.776346 samples/sec	accuracy=0.498406	lr=1.596000
Epoch[156] Batch [299]	Speed: 11232.682664 samples/sec	accuracy=0.497445	lr=1.595104
Epoch[156] Batch [349]	Speed: 11672.833283 samples/sec	accuracy=0.496217	lr=1.594208
Epoch[156] Batch [399]	Speed: 11211.432564 samples/sec	accuracy=0.495889	lr=1.593311
Epoch[156] Batch [449]	Speed: 11459.545511 samples/sec	accuracy=0.495782	lr=1.592414
Epoch[156] Batch [499]	Speed: 10953.286710 samples/sec	accuracy=0.495194	lr=1.591517
Epoch[156] Batch [549]	Speed: 11099.859453 samples/sec	accuracy=0.494716	lr=1.590620
Epoch[156] Batch [599]	Speed: 11562.361827 samples/sec	accuracy=0.494472	lr=1.589723
[Epoch 156] training: accuracy=0.494183
[Epoch 156] speed: 11265 samples/sec	time cost: 118.316986
[Epoch 156] validation: err-top1=0.515762 err-top5=0.263496
Epoch[157] Batch [49]	Speed: 10293.333962 samples/sec	accuracy=0.503467	lr=1.588359
Epoch[157] Batch [99]	Speed: 10793.079500 samples/sec	accuracy=0.501504	lr=1.587462
Epoch[157] Batch [149]	Speed: 11263.504431 samples/sec	accuracy=0.500225	lr=1.586564
Epoch[157] Batch [199]	Speed: 10943.844593 samples/sec	accuracy=0.499700	lr=1.585666
Epoch[157] Batch [249]	Speed: 11481.315029 samples/sec	accuracy=0.499188	lr=1.584768
Epoch[157] Batch [299]	Speed: 11542.378524 samples/sec	accuracy=0.498600	lr=1.583870
Epoch[157] Batch [349]	Speed: 10735.952514 samples/sec	accuracy=0.497928	lr=1.582972
Epoch[157] Batch [399]	Speed: 10809.331915 samples/sec	accuracy=0.497102	lr=1.582074
Epoch[157] Batch [449]	Speed: 10896.052506 samples/sec	accuracy=0.496979	lr=1.581175
Epoch[157] Batch [499]	Speed: 11226.513070 samples/sec	accuracy=0.497027	lr=1.580277
Epoch[157] Batch [549]	Speed: 11112.313241 samples/sec	accuracy=0.496940	lr=1.579378
Epoch[157] Batch [599]	Speed: 10879.662208 samples/sec	accuracy=0.496549	lr=1.578479
[Epoch 157] training: accuracy=0.496459
[Epoch 157] speed: 10993 samples/sec	time cost: 121.013839
[Epoch 157] validation: err-top1=0.501343 err-top5=0.249593
Epoch[158] Batch [49]	Speed: 11102.674297 samples/sec	accuracy=0.503486	lr=1.577112
Epoch[158] Batch [99]	Speed: 11604.722410 samples/sec	accuracy=0.502925	lr=1.576213
Epoch[158] Batch [149]	Speed: 10931.417561 samples/sec	accuracy=0.500788	lr=1.575314
Epoch[158] Batch [199]	Speed: 11492.845565 samples/sec	accuracy=0.500107	lr=1.574414
Epoch[158] Batch [249]	Speed: 10832.290145 samples/sec	accuracy=0.499764	lr=1.573514
Epoch[158] Batch [299]	Speed: 11110.312836 samples/sec	accuracy=0.499352	lr=1.572615
Epoch[158] Batch [349]	Speed: 11374.239141 samples/sec	accuracy=0.498887	lr=1.571715
Epoch[158] Batch [399]	Speed: 10343.398872 samples/sec	accuracy=0.498162	lr=1.570815
Epoch[158] Batch [449]	Speed: 11277.211261 samples/sec	accuracy=0.497975	lr=1.569914
Epoch[158] Batch [499]	Speed: 10520.605653 samples/sec	accuracy=0.497578	lr=1.569014
Epoch[158] Batch [549]	Speed: 11457.570974 samples/sec	accuracy=0.497149	lr=1.568113
Epoch[158] Batch [599]	Speed: 10651.543631 samples/sec	accuracy=0.497028	lr=1.567213
[Epoch 158] training: accuracy=0.497013
[Epoch 158] speed: 11040 samples/sec	time cost: 120.525653
[Epoch 158] validation: err-top1=0.496684 err-top5=0.244670
Epoch[159] Batch [49]	Speed: 10988.569973 samples/sec	accuracy=0.504434	lr=1.565862
Epoch[159] Batch [99]	Speed: 10754.569628 samples/sec	accuracy=0.504072	lr=1.564961
Epoch[159] Batch [149]	Speed: 11841.849520 samples/sec	accuracy=0.502744	lr=1.564059
Epoch[159] Batch [199]	Speed: 11018.537775 samples/sec	accuracy=0.502271	lr=1.563158
Epoch[159] Batch [249]	Speed: 11091.522872 samples/sec	accuracy=0.501428	lr=1.562257
Epoch[159] Batch [299]	Speed: 11226.696477 samples/sec	accuracy=0.500632	lr=1.561355
Epoch[159] Batch [349]	Speed: 11507.921023 samples/sec	accuracy=0.499941	lr=1.560454
Epoch[159] Batch [399]	Speed: 11164.168874 samples/sec	accuracy=0.499501	lr=1.559552
Epoch[159] Batch [449]	Speed: 10543.376360 samples/sec	accuracy=0.498854	lr=1.558650
Epoch[159] Batch [499]	Speed: 10801.009996 samples/sec	accuracy=0.498439	lr=1.557748
Epoch[159] Batch [549]	Speed: 10546.188424 samples/sec	accuracy=0.498184	lr=1.556846
Epoch[159] Batch [599]	Speed: 11218.450827 samples/sec	accuracy=0.497911	lr=1.555944
[Epoch 159] training: accuracy=0.497825
[Epoch 159] speed: 10999 samples/sec	time cost: 121.256944
[Epoch 159] validation: err-top1=0.481934 err-top5=0.229648
Epoch[160] Batch [49]	Speed: 9972.368541 samples/sec	accuracy=0.503154	lr=1.554572
Epoch[160] Batch [99]	Speed: 10683.710692 samples/sec	accuracy=0.501768	lr=1.553669
Epoch[160] Batch [149]	Speed: 11107.717036 samples/sec	accuracy=0.501826	lr=1.552767
Epoch[160] Batch [199]	Speed: 11328.874715 samples/sec	accuracy=0.501011	lr=1.551864
Epoch[160] Batch [249]	Speed: 11212.749978 samples/sec	accuracy=0.500713	lr=1.550961
Epoch[160] Batch [299]	Speed: 11511.193779 samples/sec	accuracy=0.499987	lr=1.550058
Epoch[160] Batch [349]	Speed: 11144.851730 samples/sec	accuracy=0.499284	lr=1.549154
Epoch[160] Batch [399]	Speed: 11414.863264 samples/sec	accuracy=0.499526	lr=1.548251
Epoch[160] Batch [449]	Speed: 11374.625620 samples/sec	accuracy=0.498979	lr=1.547348
Epoch[160] Batch [499]	Speed: 11183.213564 samples/sec	accuracy=0.498828	lr=1.546444
Epoch[160] Batch [549]	Speed: 11487.971657 samples/sec	accuracy=0.498418	lr=1.545540
Epoch[160] Batch [599]	Speed: 11521.399657 samples/sec	accuracy=0.498211	lr=1.544636
[Epoch 160] training: accuracy=0.497925
[Epoch 160] speed: 11130 samples/sec	time cost: 119.327166
[Epoch 160] validation: err-top1=0.534831 err-top5=0.281311
Epoch[161] Batch [49]	Speed: 11577.165946 samples/sec	accuracy=0.505928	lr=1.543280
Epoch[161] Batch [99]	Speed: 11188.005658 samples/sec	accuracy=0.502588	lr=1.542376
Epoch[161] Batch [149]	Speed: 11162.350796 samples/sec	accuracy=0.501748	lr=1.541472
Epoch[161] Batch [199]	Speed: 11558.643381 samples/sec	accuracy=0.501438	lr=1.540568
Epoch[161] Batch [249]	Speed: 11413.268032 samples/sec	accuracy=0.500961	lr=1.539663
Epoch[161] Batch [299]	Speed: 11496.393775 samples/sec	accuracy=0.500822	lr=1.538758
Epoch[161] Batch [349]	Speed: 10104.034225 samples/sec	accuracy=0.500074	lr=1.537854
Epoch[161] Batch [399]	Speed: 10440.940174 samples/sec	accuracy=0.499884	lr=1.536949
Epoch[161] Batch [449]	Speed: 11107.278680 samples/sec	accuracy=0.499300	lr=1.536044
Epoch[161] Batch [499]	Speed: 11123.379774 samples/sec	accuracy=0.499089	lr=1.535139
Epoch[161] Batch [549]	Speed: 11370.134683 samples/sec	accuracy=0.498885	lr=1.534233
Epoch[161] Batch [599]	Speed: 11408.431067 samples/sec	accuracy=0.498527	lr=1.533328
[Epoch 161] training: accuracy=0.498417
[Epoch 161] speed: 11076 samples/sec	time cost: 120.373055
[Epoch 161] validation: err-top1=0.512910 err-top5=0.258125
Epoch[162] Batch [49]	Speed: 11220.172032 samples/sec	accuracy=0.508633	lr=1.531952
Epoch[162] Batch [99]	Speed: 11257.608992 samples/sec	accuracy=0.505908	lr=1.531046
Epoch[162] Batch [149]	Speed: 11445.009683 samples/sec	accuracy=0.504798	lr=1.530140
Epoch[162] Batch [199]	Speed: 11023.525862 samples/sec	accuracy=0.504023	lr=1.529234
Epoch[162] Batch [249]	Speed: 10611.283505 samples/sec	accuracy=0.502873	lr=1.528328
Epoch[162] Batch [299]	Speed: 10584.898323 samples/sec	accuracy=0.501829	lr=1.527422
Epoch[162] Batch [349]	Speed: 10853.849107 samples/sec	accuracy=0.501602	lr=1.526516
Epoch[162] Batch [399]	Speed: 11345.403074 samples/sec	accuracy=0.501057	lr=1.525610
Epoch[162] Batch [449]	Speed: 11249.004105 samples/sec	accuracy=0.500487	lr=1.524703
Epoch[162] Batch [499]	Speed: 10484.227808 samples/sec	accuracy=0.500043	lr=1.523797
Epoch[162] Batch [549]	Speed: 11073.765027 samples/sec	accuracy=0.499773	lr=1.522890
Epoch[162] Batch [599]	Speed: 11103.397893 samples/sec	accuracy=0.499530	lr=1.521983
[Epoch 162] training: accuracy=0.499402
[Epoch 162] speed: 11018 samples/sec	time cost: 120.649999
[Epoch 162] validation: err-top1=0.487956 err-top5=0.236430
Epoch[163] Batch [49]	Speed: 10995.972028 samples/sec	accuracy=0.508457	lr=1.520623
Epoch[163] Batch [99]	Speed: 11159.213349 samples/sec	accuracy=0.507583	lr=1.519716
Epoch[163] Batch [149]	Speed: 11461.131993 samples/sec	accuracy=0.505339	lr=1.518809
Epoch[163] Batch [199]	Speed: 10834.815102 samples/sec	accuracy=0.503972	lr=1.517901
Epoch[163] Batch [249]	Speed: 11122.265879 samples/sec	accuracy=0.503422	lr=1.516994
Epoch[163] Batch [299]	Speed: 11572.152926 samples/sec	accuracy=0.502482	lr=1.516086
Epoch[163] Batch [349]	Speed: 10516.385120 samples/sec	accuracy=0.501917	lr=1.515179
Epoch[163] Batch [399]	Speed: 10772.732064 samples/sec	accuracy=0.501471	lr=1.514271
Epoch[163] Batch [449]	Speed: 11270.265695 samples/sec	accuracy=0.501197	lr=1.513363
Epoch[163] Batch [499]	Speed: 11448.800935 samples/sec	accuracy=0.500930	lr=1.512455
Epoch[163] Batch [549]	Speed: 11169.353910 samples/sec	accuracy=0.500471	lr=1.511547
Epoch[163] Batch [599]	Speed: 11488.423983 samples/sec	accuracy=0.500088	lr=1.510639
[Epoch 163] training: accuracy=0.499813
[Epoch 163] speed: 11137 samples/sec	time cost: 119.608862
[Epoch 163] validation: err-top1=0.484294 err-top5=0.229980
Epoch[164] Batch [49]	Speed: 10409.491415 samples/sec	accuracy=0.510059	lr=1.509259
Epoch[164] Batch [99]	Speed: 10830.039991 samples/sec	accuracy=0.507646	lr=1.508350
Epoch[164] Batch [149]	Speed: 10121.558358 samples/sec	accuracy=0.506735	lr=1.507442
Epoch[164] Batch [199]	Speed: 11515.727389 samples/sec	accuracy=0.504941	lr=1.506533
Epoch[164] Batch [249]	Speed: 11700.208189 samples/sec	accuracy=0.503812	lr=1.505624
Epoch[164] Batch [299]	Speed: 11302.733796 samples/sec	accuracy=0.503122	lr=1.504716
Epoch[164] Batch [349]	Speed: 11422.799944 samples/sec	accuracy=0.502673	lr=1.503807
Epoch[164] Batch [399]	Speed: 11357.859787 samples/sec	accuracy=0.501748	lr=1.502898
Epoch[164] Batch [449]	Speed: 11404.931790 samples/sec	accuracy=0.501287	lr=1.501988
Epoch[164] Batch [499]	Speed: 11173.829469 samples/sec	accuracy=0.500854	lr=1.501079
Epoch[164] Batch [549]	Speed: 11324.706476 samples/sec	accuracy=0.500514	lr=1.500170
Epoch[164] Batch [599]	Speed: 10198.792715 samples/sec	accuracy=0.500365	lr=1.499260
[Epoch 164] training: accuracy=0.500310
[Epoch 164] speed: 10986 samples/sec	time cost: 121.404083
[Epoch 164] validation: err-top1=0.491875 err-top5=0.240820
Epoch[165] Batch [49]	Speed: 10770.322113 samples/sec	accuracy=0.507686	lr=1.497878
Epoch[165] Batch [99]	Speed: 11368.313001 samples/sec	accuracy=0.506064	lr=1.496968
Epoch[165] Batch [149]	Speed: 11304.735362 samples/sec	accuracy=0.506403	lr=1.496058
Epoch[165] Batch [199]	Speed: 11193.989738 samples/sec	accuracy=0.504985	lr=1.495149
Epoch[165] Batch [249]	Speed: 10961.503712 samples/sec	accuracy=0.504355	lr=1.494239
Epoch[165] Batch [299]	Speed: 11093.929714 samples/sec	accuracy=0.503643	lr=1.493328
Epoch[165] Batch [349]	Speed: 11424.909301 samples/sec	accuracy=0.502935	lr=1.492418
Epoch[165] Batch [399]	Speed: 11428.904385 samples/sec	accuracy=0.502523	lr=1.491508
Epoch[165] Batch [449]	Speed: 11233.701843 samples/sec	accuracy=0.502369	lr=1.490598
Epoch[165] Batch [499]	Speed: 11525.242004 samples/sec	accuracy=0.502114	lr=1.489687
Epoch[165] Batch [549]	Speed: 11110.943724 samples/sec	accuracy=0.501722	lr=1.488777
Epoch[165] Batch [599]	Speed: 11172.411617 samples/sec	accuracy=0.501200	lr=1.487866
[Epoch 165] training: accuracy=0.500971
[Epoch 165] speed: 11168 samples/sec	time cost: 118.958079
[Epoch 165] validation: err-top1=0.491089 err-top5=0.244446
Epoch[166] Batch [49]	Speed: 10559.209147 samples/sec	accuracy=0.508809	lr=1.486500
Epoch[166] Batch [99]	Speed: 11492.451933 samples/sec	accuracy=0.507388	lr=1.485589
Epoch[166] Batch [149]	Speed: 10958.702386 samples/sec	accuracy=0.506279	lr=1.484678
Epoch[166] Batch [199]	Speed: 11546.866267 samples/sec	accuracy=0.505425	lr=1.483767
Epoch[166] Batch [249]	Speed: 11334.337138 samples/sec	accuracy=0.504313	lr=1.482856
Epoch[166] Batch [299]	Speed: 11143.787888 samples/sec	accuracy=0.503237	lr=1.481945
Epoch[166] Batch [349]	Speed: 11574.443198 samples/sec	accuracy=0.502886	lr=1.481033
Epoch[166] Batch [399]	Speed: 11549.653079 samples/sec	accuracy=0.502410	lr=1.480122
Epoch[166] Batch [449]	Speed: 11124.185881 samples/sec	accuracy=0.502228	lr=1.479210
Epoch[166] Batch [499]	Speed: 11033.708908 samples/sec	accuracy=0.501904	lr=1.478299
Epoch[166] Batch [549]	Speed: 11201.863754 samples/sec	accuracy=0.501744	lr=1.477387
Epoch[166] Batch [599]	Speed: 11261.251396 samples/sec	accuracy=0.501252	lr=1.476475
[Epoch 166] training: accuracy=0.501038
[Epoch 166] speed: 11192 samples/sec	time cost: 119.182720
[Epoch 166] validation: err-top1=0.499180 err-top5=0.249844
Epoch[167] Batch [49]	Speed: 9638.861745 samples/sec	accuracy=0.510596	lr=1.475089
Epoch[167] Batch [99]	Speed: 11492.365522 samples/sec	accuracy=0.508599	lr=1.474177
Epoch[167] Batch [149]	Speed: 11320.122986 samples/sec	accuracy=0.506204	lr=1.473265
Epoch[167] Batch [199]	Speed: 11470.246864 samples/sec	accuracy=0.505183	lr=1.472353
Epoch[167] Batch [249]	Speed: 11303.110671 samples/sec	accuracy=0.504480	lr=1.471440
Epoch[167] Batch [299]	Speed: 11160.419918 samples/sec	accuracy=0.503369	lr=1.470528
Epoch[167] Batch [349]	Speed: 11657.741246 samples/sec	accuracy=0.503223	lr=1.469616
Epoch[167] Batch [399]	Speed: 10503.492850 samples/sec	accuracy=0.502798	lr=1.468703
Epoch[167] Batch [449]	Speed: 10756.773710 samples/sec	accuracy=0.502402	lr=1.467790
Epoch[167] Batch [499]	Speed: 11142.608039 samples/sec	accuracy=0.502052	lr=1.466878
Epoch[167] Batch [549]	Speed: 11126.994052 samples/sec	accuracy=0.501878	lr=1.465965
Epoch[167] Batch [599]	Speed: 11058.339263 samples/sec	accuracy=0.501635	lr=1.465052
[Epoch 167] training: accuracy=0.501410
[Epoch 167] speed: 11009 samples/sec	time cost: 120.628610
[Epoch 167] validation: err-top1=0.521932 err-top5=0.268087
Epoch[168] Batch [49]	Speed: 11129.660590 samples/sec	accuracy=0.509922	lr=1.463683
Epoch[168] Batch [99]	Speed: 11024.318410 samples/sec	accuracy=0.507881	lr=1.462770
Epoch[168] Batch [149]	Speed: 10714.247074 samples/sec	accuracy=0.507253	lr=1.461856
Epoch[168] Batch [199]	Speed: 11475.258729 samples/sec	accuracy=0.506387	lr=1.460943
Epoch[168] Batch [249]	Speed: 10827.744093 samples/sec	accuracy=0.505457	lr=1.460030
Epoch[168] Batch [299]	Speed: 10723.826862 samples/sec	accuracy=0.505379	lr=1.459116
Epoch[168] Batch [349]	Speed: 11352.094421 samples/sec	accuracy=0.504397	lr=1.458203
Epoch[168] Batch [399]	Speed: 11212.878780 samples/sec	accuracy=0.503920	lr=1.457289
Epoch[168] Batch [449]	Speed: 11190.360386 samples/sec	accuracy=0.503581	lr=1.456376
Epoch[168] Batch [499]	Speed: 10045.910972 samples/sec	accuracy=0.502914	lr=1.455462
Epoch[168] Batch [549]	Speed: 11499.581463 samples/sec	accuracy=0.502524	lr=1.454548
Epoch[168] Batch [599]	Speed: 11263.359695 samples/sec	accuracy=0.502248	lr=1.453634
[Epoch 168] training: accuracy=0.502141
[Epoch 168] speed: 10969 samples/sec	time cost: 121.300251
[Epoch 168] validation: err-top1=0.503540 err-top5=0.248596
Epoch[169] Batch [49]	Speed: 11441.209055 samples/sec	accuracy=0.512061	lr=1.452245
Epoch[169] Batch [99]	Speed: 10564.426028 samples/sec	accuracy=0.510884	lr=1.451331
Epoch[169] Batch [149]	Speed: 11262.200166 samples/sec	accuracy=0.508649	lr=1.450417
Epoch[169] Batch [199]	Speed: 11302.092837 samples/sec	accuracy=0.507852	lr=1.449503
Epoch[169] Batch [249]	Speed: 11087.594408 samples/sec	accuracy=0.506533	lr=1.448588
Epoch[169] Batch [299]	Speed: 11127.174222 samples/sec	accuracy=0.505361	lr=1.447674
Epoch[169] Batch [349]	Speed: 10895.338269 samples/sec	accuracy=0.505226	lr=1.446760
Epoch[169] Batch [399]	Speed: 11618.080977 samples/sec	accuracy=0.504592	lr=1.445845
Epoch[169] Batch [449]	Speed: 11125.451458 samples/sec	accuracy=0.504506	lr=1.444931
Epoch[169] Batch [499]	Speed: 10693.776404 samples/sec	accuracy=0.504117	lr=1.444016
Epoch[169] Batch [549]	Speed: 11303.893059 samples/sec	accuracy=0.503712	lr=1.443101
Epoch[169] Batch [599]	Speed: 11085.977446 samples/sec	accuracy=0.503366	lr=1.442186
[Epoch 169] training: accuracy=0.503304
[Epoch 169] speed: 11065 samples/sec	time cost: 120.420743
[Epoch 169] validation: err-top1=0.496426 err-top5=0.247520
Epoch[170] Batch [49]	Speed: 10507.167841 samples/sec	accuracy=0.510039	lr=1.440814
Epoch[170] Batch [99]	Speed: 10937.257799 samples/sec	accuracy=0.508530	lr=1.439899
Epoch[170] Batch [149]	Speed: 11412.535327 samples/sec	accuracy=0.508060	lr=1.438984
Epoch[170] Batch [199]	Speed: 10920.976650 samples/sec	accuracy=0.506880	lr=1.438069
Epoch[170] Batch [249]	Speed: 10519.227632 samples/sec	accuracy=0.506324	lr=1.437154
Epoch[170] Batch [299]	Speed: 11370.802649 samples/sec	accuracy=0.505535	lr=1.436238
Epoch[170] Batch [349]	Speed: 11231.721825 samples/sec	accuracy=0.505116	lr=1.435323
Epoch[170] Batch [399]	Speed: 10142.938102 samples/sec	accuracy=0.504828	lr=1.434408
Epoch[170] Batch [449]	Speed: 9857.395103 samples/sec	accuracy=0.504112	lr=1.433492
Epoch[170] Batch [499]	Speed: 11168.766617 samples/sec	accuracy=0.503788	lr=1.432577
Epoch[170] Batch [549]	Speed: 11594.717444 samples/sec	accuracy=0.503493	lr=1.431661
Epoch[170] Batch [599]	Speed: 11458.451621 samples/sec	accuracy=0.503232	lr=1.430746
[Epoch 170] training: accuracy=0.503199
[Epoch 170] speed: 10883 samples/sec	time cost: 122.207696
[Epoch 170] validation: err-top1=0.466248 err-top5=0.219543
Epoch[171] Batch [49]	Speed: 10775.291774 samples/sec	accuracy=0.511504	lr=1.429354
Epoch[171] Batch [99]	Speed: 11528.218586 samples/sec	accuracy=0.511206	lr=1.428438
Epoch[171] Batch [149]	Speed: 10694.817307 samples/sec	accuracy=0.510225	lr=1.427522
Epoch[171] Batch [199]	Speed: 11291.129590 samples/sec	accuracy=0.509380	lr=1.426606
Epoch[171] Batch [249]	Speed: 11708.823831 samples/sec	accuracy=0.508094	lr=1.425690
Epoch[171] Batch [299]	Speed: 10992.640700 samples/sec	accuracy=0.507487	lr=1.424774
Epoch[171] Batch [349]	Speed: 10969.151436 samples/sec	accuracy=0.507080	lr=1.423858
Epoch[171] Batch [399]	Speed: 11481.434729 samples/sec	accuracy=0.506541	lr=1.422942
Epoch[171] Batch [449]	Speed: 11034.659979 samples/sec	accuracy=0.506404	lr=1.422025
Epoch[171] Batch [499]	Speed: 11232.396833 samples/sec	accuracy=0.505831	lr=1.421109
Epoch[171] Batch [549]	Speed: 11299.507138 samples/sec	accuracy=0.505569	lr=1.420193
Epoch[171] Batch [599]	Speed: 11094.195359 samples/sec	accuracy=0.505196	lr=1.419276
[Epoch 171] training: accuracy=0.505002
[Epoch 171] speed: 11157 samples/sec	time cost: 119.518728
[Epoch 171] validation: err-top1=0.497715 err-top5=0.244316
Epoch[172] Batch [49]	Speed: 11187.654778 samples/sec	accuracy=0.515439	lr=1.417883
Epoch[172] Batch [99]	Speed: 10319.056350 samples/sec	accuracy=0.512705	lr=1.416966
Epoch[172] Batch [149]	Speed: 11373.064200 samples/sec	accuracy=0.511930	lr=1.416050
Epoch[172] Batch [199]	Speed: 11310.934062 samples/sec	accuracy=0.510854	lr=1.415133
Epoch[172] Batch [249]	Speed: 11450.610040 samples/sec	accuracy=0.509236	lr=1.414216
Epoch[172] Batch [299]	Speed: 11972.289572 samples/sec	accuracy=0.508757	lr=1.413299
Epoch[172] Batch [349]	Speed: 11621.670156 samples/sec	accuracy=0.507581	lr=1.412383
Epoch[172] Batch [399]	Speed: 11477.624291 samples/sec	accuracy=0.507097	lr=1.411466
Epoch[172] Batch [449]	Speed: 11312.110799 samples/sec	accuracy=0.506432	lr=1.410549
Epoch[172] Batch [499]	Speed: 11154.885942 samples/sec	accuracy=0.506277	lr=1.409632
Epoch[172] Batch [549]	Speed: 11209.201188 samples/sec	accuracy=0.506002	lr=1.408714
Epoch[172] Batch [599]	Speed: 11492.512821 samples/sec	accuracy=0.505274	lr=1.407797
[Epoch 172] training: accuracy=0.505187
[Epoch 172] speed: 11286 samples/sec	time cost: 117.839973
[Epoch 172] validation: err-top1=0.494364 err-top5=0.244690
Epoch[173] Batch [49]	Speed: 10821.916031 samples/sec	accuracy=0.511914	lr=1.406421
Epoch[173] Batch [99]	Speed: 10824.157082 samples/sec	accuracy=0.512178	lr=1.405504
Epoch[173] Batch [149]	Speed: 11751.372167 samples/sec	accuracy=0.510967	lr=1.404587
Epoch[173] Batch [199]	Speed: 11304.062640 samples/sec	accuracy=0.509946	lr=1.403669
Epoch[173] Batch [249]	Speed: 11484.442769 samples/sec	accuracy=0.508768	lr=1.402752
Epoch[173] Batch [299]	Speed: 11243.675701 samples/sec	accuracy=0.507969	lr=1.401834
Epoch[173] Batch [349]	Speed: 11451.420614 samples/sec	accuracy=0.507751	lr=1.400917
Epoch[173] Batch [399]	Speed: 11260.550478 samples/sec	accuracy=0.507344	lr=1.399999
Epoch[173] Batch [449]	Speed: 11263.292645 samples/sec	accuracy=0.506980	lr=1.399082
Epoch[173] Batch [499]	Speed: 10923.522011 samples/sec	accuracy=0.506576	lr=1.398164
Epoch[173] Batch [549]	Speed: 11291.051226 samples/sec	accuracy=0.506350	lr=1.397246
Epoch[173] Batch [599]	Speed: 11357.677175 samples/sec	accuracy=0.506121	lr=1.396328
[Epoch 173] training: accuracy=0.506098
[Epoch 173] speed: 11169 samples/sec	time cost: 119.435600
[Epoch 173] validation: err-top1=0.485293 err-top5=0.234668
Epoch[174] Batch [49]	Speed: 10838.085631 samples/sec	accuracy=0.513164	lr=1.394933
Epoch[174] Batch [99]	Speed: 11329.283220 samples/sec	accuracy=0.513379	lr=1.394015
Epoch[174] Batch [149]	Speed: 11169.071293 samples/sec	accuracy=0.511149	lr=1.393097
Epoch[174] Batch [199]	Speed: 10981.157663 samples/sec	accuracy=0.510325	lr=1.392179
Epoch[174] Batch [249]	Speed: 11203.672516 samples/sec	accuracy=0.509383	lr=1.391261
Epoch[174] Batch [299]	Speed: 11764.608567 samples/sec	accuracy=0.508574	lr=1.390343
Epoch[174] Batch [349]	Speed: 11544.899386 samples/sec	accuracy=0.507980	lr=1.389425
Epoch[174] Batch [399]	Speed: 11105.625819 samples/sec	accuracy=0.507579	lr=1.388507
Epoch[174] Batch [449]	Speed: 11102.942656 samples/sec	accuracy=0.507129	lr=1.387588
Epoch[174] Batch [499]	Speed: 11395.654914 samples/sec	accuracy=0.506886	lr=1.386670
Epoch[174] Batch [549]	Speed: 11286.707592 samples/sec	accuracy=0.506912	lr=1.385752
Epoch[174] Batch [599]	Speed: 11295.295719 samples/sec	accuracy=0.506524	lr=1.384834
[Epoch 174] training: accuracy=0.506331
[Epoch 174] speed: 11241 samples/sec	time cost: 118.399691
[Epoch 174] validation: err-top1=0.492411 err-top5=0.243774
Epoch[175] Batch [49]	Speed: 10265.440980 samples/sec	accuracy=0.515215	lr=1.383456
Epoch[175] Batch [99]	Speed: 11203.841734 samples/sec	accuracy=0.513071	lr=1.382537
Epoch[175] Batch [149]	Speed: 11063.620232 samples/sec	accuracy=0.511458	lr=1.381619
Epoch[175] Batch [199]	Speed: 11574.233593 samples/sec	accuracy=0.510251	lr=1.380700
Epoch[175] Batch [249]	Speed: 11453.945271 samples/sec	accuracy=0.509555	lr=1.379782
Epoch[175] Batch [299]	Speed: 11252.528324 samples/sec	accuracy=0.508737	lr=1.378863
Epoch[175] Batch [349]	Speed: 11137.795537 samples/sec	accuracy=0.508403	lr=1.377944
Epoch[175] Batch [399]	Speed: 11519.404984 samples/sec	accuracy=0.508141	lr=1.377026
Epoch[175] Batch [449]	Speed: 11497.936301 samples/sec	accuracy=0.507667	lr=1.376107
Epoch[175] Batch [499]	Speed: 11767.534050 samples/sec	accuracy=0.507321	lr=1.375188
Epoch[175] Batch [549]	Speed: 10035.094584 samples/sec	accuracy=0.506933	lr=1.374269
Epoch[175] Batch [599]	Speed: 10667.847870 samples/sec	accuracy=0.506794	lr=1.373350
[Epoch 175] training: accuracy=0.506593
[Epoch 175] speed: 11088 samples/sec	time cost: 120.148645
[Epoch 175] validation: err-top1=0.473918 err-top5=0.224386
Epoch[176] Batch [49]	Speed: 11281.670269 samples/sec	accuracy=0.513818	lr=1.371954
Epoch[176] Batch [99]	Speed: 11240.465607 samples/sec	accuracy=0.512578	lr=1.371035
Epoch[176] Batch [149]	Speed: 11527.070090 samples/sec	accuracy=0.511976	lr=1.370116
Epoch[176] Batch [199]	Speed: 11419.246003 samples/sec	accuracy=0.510750	lr=1.369197
Epoch[176] Batch [249]	Speed: 11395.299959 samples/sec	accuracy=0.510156	lr=1.368278
Epoch[176] Batch [299]	Speed: 11506.515459 samples/sec	accuracy=0.509585	lr=1.367358
Epoch[176] Batch [349]	Speed: 10907.862038 samples/sec	accuracy=0.509097	lr=1.366439
Epoch[176] Batch [399]	Speed: 11334.841462 samples/sec	accuracy=0.508594	lr=1.365520
Epoch[176] Batch [449]	Speed: 11080.848072 samples/sec	accuracy=0.508360	lr=1.364601
Epoch[176] Batch [499]	Speed: 10665.042595 samples/sec	accuracy=0.507742	lr=1.363682
Epoch[176] Batch [549]	Speed: 11332.533188 samples/sec	accuracy=0.507265	lr=1.362762
Epoch[176] Batch [599]	Speed: 10483.207278 samples/sec	accuracy=0.507250	lr=1.361843
[Epoch 176] training: accuracy=0.507317
[Epoch 176] speed: 11131 samples/sec	time cost: 119.613349
[Epoch 176] validation: err-top1=0.464023 err-top5=0.218730
Epoch[177] Batch [49]	Speed: 11209.417381 samples/sec	accuracy=0.517793	lr=1.360464
Epoch[177] Batch [99]	Speed: 11620.033577 samples/sec	accuracy=0.516157	lr=1.359545
Epoch[177] Batch [149]	Speed: 11183.383329 samples/sec	accuracy=0.515430	lr=1.358625
Epoch[177] Batch [199]	Speed: 11397.683177 samples/sec	accuracy=0.513909	lr=1.357706
Epoch[177] Batch [249]	Speed: 11111.892346 samples/sec	accuracy=0.512842	lr=1.356786
Epoch[177] Batch [299]	Speed: 11577.231792 samples/sec	accuracy=0.512033	lr=1.355867
Epoch[177] Batch [349]	Speed: 11017.512607 samples/sec	accuracy=0.511271	lr=1.354947
Epoch[177] Batch [399]	Speed: 10461.706504 samples/sec	accuracy=0.510461	lr=1.354028
Epoch[177] Batch [449]	Speed: 10540.519750 samples/sec	accuracy=0.509806	lr=1.353108
Epoch[177] Batch [499]	Speed: 10565.295836 samples/sec	accuracy=0.509440	lr=1.352189
Epoch[177] Batch [549]	Speed: 10879.551144 samples/sec	accuracy=0.509138	lr=1.351269
Epoch[177] Batch [599]	Speed: 11339.131522 samples/sec	accuracy=0.508704	lr=1.350349
[Epoch 177] training: accuracy=0.508588
[Epoch 177] speed: 11048 samples/sec	time cost: 120.410371
[Epoch 177] validation: err-top1=0.496704 err-top5=0.245382
Epoch[178] Batch [49]	Speed: 11482.857503 samples/sec	accuracy=0.516582	lr=1.348951
Epoch[178] Batch [99]	Speed: 11395.199584 samples/sec	accuracy=0.516021	lr=1.348032
Epoch[178] Batch [149]	Speed: 11069.360694 samples/sec	accuracy=0.515267	lr=1.347112
Epoch[178] Batch [199]	Speed: 10719.271318 samples/sec	accuracy=0.513669	lr=1.346192
Epoch[178] Batch [249]	Speed: 11547.781497 samples/sec	accuracy=0.512730	lr=1.345272
Epoch[178] Batch [299]	Speed: 11083.256290 samples/sec	accuracy=0.511724	lr=1.344353
Epoch[178] Batch [349]	Speed: 10590.483489 samples/sec	accuracy=0.511385	lr=1.343433
Epoch[178] Batch [399]	Speed: 11053.039439 samples/sec	accuracy=0.510914	lr=1.342513
Epoch[178] Batch [449]	Speed: 11311.632329 samples/sec	accuracy=0.510460	lr=1.341593
Epoch[178] Batch [499]	Speed: 11423.246241 samples/sec	accuracy=0.509716	lr=1.340673
Epoch[178] Batch [549]	Speed: 11783.493776 samples/sec	accuracy=0.509301	lr=1.339753
Epoch[178] Batch [599]	Speed: 11162.007615 samples/sec	accuracy=0.508940	lr=1.338833
[Epoch 178] training: accuracy=0.508784
[Epoch 178] speed: 11205 samples/sec	time cost: 119.051518
[Epoch 178] validation: err-top1=0.461914 err-top5=0.216230
Epoch[179] Batch [49]	Speed: 10047.673347 samples/sec	accuracy=0.516162	lr=1.337453
Epoch[179] Batch [99]	Speed: 11399.451957 samples/sec	accuracy=0.515264	lr=1.336533
Epoch[179] Batch [149]	Speed: 11056.854925 samples/sec	accuracy=0.514867	lr=1.335613
Epoch[179] Batch [199]	Speed: 10877.805842 samples/sec	accuracy=0.513245	lr=1.334693
Epoch[179] Batch [249]	Speed: 11578.716179 samples/sec	accuracy=0.512053	lr=1.333773
Epoch[179] Batch [299]	Speed: 11433.435790 samples/sec	accuracy=0.511138	lr=1.332853
Epoch[179] Batch [349]	Speed: 11515.463095 samples/sec	accuracy=0.510824	lr=1.331933
Epoch[179] Batch [399]	Speed: 10884.426846 samples/sec	accuracy=0.510684	lr=1.331013
Epoch[179] Batch [449]	Speed: 11243.959751 samples/sec	accuracy=0.510289	lr=1.330093
Epoch[179] Batch [499]	Speed: 11788.894497 samples/sec	accuracy=0.509926	lr=1.329173
Epoch[179] Batch [549]	Speed: 11387.099826 samples/sec	accuracy=0.509765	lr=1.328253
Epoch[179] Batch [599]	Speed: 10525.191695 samples/sec	accuracy=0.509631	lr=1.327333
[Epoch 179] training: accuracy=0.509468
[Epoch 179] speed: 11087 samples/sec	time cost: 120.167030
[Epoch 179] validation: err-top1=0.517782 err-top5=0.264486
Epoch[180] Batch [49]	Speed: 11135.363274 samples/sec	accuracy=0.517734	lr=1.325934
Epoch[180] Batch [99]	Speed: 10637.477806 samples/sec	accuracy=0.516777	lr=1.325014
Epoch[180] Batch [149]	Speed: 10825.693925 samples/sec	accuracy=0.514961	lr=1.324094
Epoch[180] Batch [199]	Speed: 11437.614411 samples/sec	accuracy=0.513547	lr=1.323173
Epoch[180] Batch [249]	Speed: 11313.106003 samples/sec	accuracy=0.513021	lr=1.322253
Epoch[180] Batch [299]	Speed: 10824.314756 samples/sec	accuracy=0.511935	lr=1.321333
Epoch[180] Batch [349]	Speed: 10674.710719 samples/sec	accuracy=0.511641	lr=1.320413
Epoch[180] Batch [399]	Speed: 11342.667216 samples/sec	accuracy=0.511340	lr=1.319492
Epoch[180] Batch [449]	Speed: 10965.183467 samples/sec	accuracy=0.511033	lr=1.318572
Epoch[180] Batch [499]	Speed: 11279.841273 samples/sec	accuracy=0.510459	lr=1.317652
Epoch[180] Batch [549]	Speed: 11297.261073 samples/sec	accuracy=0.510263	lr=1.316732
Epoch[180] Batch [599]	Speed: 10713.122485 samples/sec	accuracy=0.510054	lr=1.315811
[Epoch 180] training: accuracy=0.509952
[Epoch 180] speed: 11009 samples/sec	time cost: 121.189248
[Epoch 180] validation: err-top1=0.470500 err-top5=0.221985
Epoch[181] Batch [49]	Speed: 10565.919369 samples/sec	accuracy=0.518779	lr=1.314413
Epoch[181] Batch [99]	Speed: 11217.343592 samples/sec	accuracy=0.518447	lr=1.313492
Epoch[181] Batch [149]	Speed: 10513.400026 samples/sec	accuracy=0.515645	lr=1.312572
Epoch[181] Batch [199]	Speed: 10742.701501 samples/sec	accuracy=0.514836	lr=1.311652
Epoch[181] Batch [249]	Speed: 10787.327428 samples/sec	accuracy=0.513357	lr=1.310731
Epoch[181] Batch [299]	Speed: 11524.742552 samples/sec	accuracy=0.512607	lr=1.309811
Epoch[181] Batch [349]	Speed: 11365.765493 samples/sec	accuracy=0.512553	lr=1.308891
Epoch[181] Batch [399]	Speed: 10898.361151 samples/sec	accuracy=0.512080	lr=1.307970
Epoch[181] Batch [449]	Speed: 10985.310561 samples/sec	accuracy=0.511608	lr=1.307050
Epoch[181] Batch [499]	Speed: 11216.563473 samples/sec	accuracy=0.511436	lr=1.306130
Epoch[181] Batch [549]	Speed: 11320.508780 samples/sec	accuracy=0.510873	lr=1.305209
Epoch[181] Batch [599]	Speed: 10912.976937 samples/sec	accuracy=0.510673	lr=1.304289
[Epoch 181] training: accuracy=0.510593
[Epoch 181] speed: 10978 samples/sec	time cost: 121.223570
[Epoch 181] validation: err-top1=0.465313 err-top5=0.218203
Epoch[182] Batch [49]	Speed: 11182.561341 samples/sec	accuracy=0.520439	lr=1.302908
Epoch[182] Batch [99]	Speed: 10402.024969 samples/sec	accuracy=0.519521	lr=1.301988
Epoch[182] Batch [149]	Speed: 11082.928823 samples/sec	accuracy=0.517601	lr=1.301068
Epoch[182] Batch [199]	Speed: 10790.316960 samples/sec	accuracy=0.516287	lr=1.300147
Epoch[182] Batch [249]	Speed: 11197.284850 samples/sec	accuracy=0.515443	lr=1.299227
Epoch[182] Batch [299]	Speed: 11756.917585 samples/sec	accuracy=0.514147	lr=1.298307
Epoch[182] Batch [349]	Speed: 11448.713958 samples/sec	accuracy=0.513870	lr=1.297386
Epoch[182] Batch [399]	Speed: 11250.174473 samples/sec	accuracy=0.513627	lr=1.296466
Epoch[182] Batch [449]	Speed: 11456.056981 samples/sec	accuracy=0.513411	lr=1.295545
Epoch[182] Batch [499]	Speed: 11119.122424 samples/sec	accuracy=0.513069	lr=1.294625
Epoch[182] Batch [549]	Speed: 11249.176168 samples/sec	accuracy=0.512584	lr=1.293705
Epoch[182] Batch [599]	Speed: 11653.019586 samples/sec	accuracy=0.511965	lr=1.292784
[Epoch 182] training: accuracy=0.511853
[Epoch 182] speed: 11164 samples/sec	time cost: 119.251061
[Epoch 182] validation: err-top1=0.494100 err-top5=0.245890
Epoch[183] Batch [49]	Speed: 10403.213951 samples/sec	accuracy=0.519209	lr=1.291386
Epoch[183] Batch [99]	Speed: 11483.532331 samples/sec	accuracy=0.517954	lr=1.290465
Epoch[183] Batch [149]	Speed: 11326.744509 samples/sec	accuracy=0.517096	lr=1.289545
Epoch[183] Batch [199]	Speed: 11505.831146 samples/sec	accuracy=0.515640	lr=1.288625
Epoch[183] Batch [249]	Speed: 11240.666828 samples/sec	accuracy=0.515320	lr=1.287704
Epoch[183] Batch [299]	Speed: 11297.415597 samples/sec	accuracy=0.514587	lr=1.286784
Epoch[183] Batch [349]	Speed: 11554.124413 samples/sec	accuracy=0.514270	lr=1.285864
Epoch[183] Batch [399]	Speed: 11220.679439 samples/sec	accuracy=0.513501	lr=1.284943
Epoch[183] Batch [449]	Speed: 11402.922441 samples/sec	accuracy=0.513198	lr=1.284023
Epoch[183] Batch [499]	Speed: 10962.706518 samples/sec	accuracy=0.512816	lr=1.283103
Epoch[183] Batch [549]	Speed: 11286.337148 samples/sec	accuracy=0.512515	lr=1.282182
Epoch[183] Batch [599]	Speed: 11456.433761 samples/sec	accuracy=0.512347	lr=1.281262
[Epoch 183] training: accuracy=0.512288
[Epoch 183] speed: 11251 samples/sec	time cost: 118.391829
[Epoch 183] validation: err-top1=0.461602 err-top5=0.214043
Epoch[184] Batch [49]	Speed: 9697.064111 samples/sec	accuracy=0.521289	lr=1.279882
Epoch[184] Batch [99]	Speed: 11445.774321 samples/sec	accuracy=0.518970	lr=1.278962
Epoch[184] Batch [149]	Speed: 11334.477722 samples/sec	accuracy=0.517044	lr=1.278041
Epoch[184] Batch [199]	Speed: 11366.261187 samples/sec	accuracy=0.516008	lr=1.277121
Epoch[184] Batch [249]	Speed: 11516.820507 samples/sec	accuracy=0.515205	lr=1.276201
Epoch[184] Batch [299]	Speed: 11316.006203 samples/sec	accuracy=0.514569	lr=1.275281
Epoch[184] Batch [349]	Speed: 11136.352742 samples/sec	accuracy=0.514258	lr=1.274360
Epoch[184] Batch [399]	Speed: 10625.318474 samples/sec	accuracy=0.513995	lr=1.273440
Epoch[184] Batch [449]	Speed: 11176.114834 samples/sec	accuracy=0.513827	lr=1.272520
Epoch[184] Batch [499]	Speed: 11643.460840 samples/sec	accuracy=0.513485	lr=1.271600
Epoch[184] Batch [549]	Speed: 11451.244140 samples/sec	accuracy=0.513238	lr=1.270680
Epoch[184] Batch [599]	Speed: 10680.335319 samples/sec	accuracy=0.512853	lr=1.269760
[Epoch 184] training: accuracy=0.512809
[Epoch 184] speed: 11064 samples/sec	time cost: 120.210176
[Epoch 184] validation: err-top1=0.468750 err-top5=0.225199
Epoch[185] Batch [49]	Speed: 10908.139901 samples/sec	accuracy=0.522969	lr=1.268361
Epoch[185] Batch [99]	Speed: 11506.665279 samples/sec	accuracy=0.520015	lr=1.267441
Epoch[185] Batch [149]	Speed: 11039.758589 samples/sec	accuracy=0.518444	lr=1.266521
Epoch[185] Batch [199]	Speed: 11359.448581 samples/sec	accuracy=0.517312	lr=1.265601
Epoch[185] Batch [249]	Speed: 11086.316253 samples/sec	accuracy=0.516516	lr=1.264681
Epoch[185] Batch [299]	Speed: 11397.459963 samples/sec	accuracy=0.515254	lr=1.263761
Epoch[185] Batch [349]	Speed: 11107.601555 samples/sec	accuracy=0.514662	lr=1.262841
Epoch[185] Batch [399]	Speed: 11207.888702 samples/sec	accuracy=0.514508	lr=1.261921
Epoch[185] Batch [449]	Speed: 10773.418966 samples/sec	accuracy=0.514125	lr=1.261001
Epoch[185] Batch [499]	Speed: 11253.549929 samples/sec	accuracy=0.513761	lr=1.260081
Epoch[185] Batch [549]	Speed: 11215.917019 samples/sec	accuracy=0.513720	lr=1.259161
Epoch[185] Batch [599]	Speed: 11591.770876 samples/sec	accuracy=0.513446	lr=1.258241
[Epoch 185] training: accuracy=0.513430
[Epoch 185] speed: 11184 samples/sec	time cost: 119.105652
[Epoch 185] validation: err-top1=0.499863 err-top5=0.248262
Epoch[186] Batch [49]	Speed: 10730.352303 samples/sec	accuracy=0.521113	lr=1.256862
Epoch[186] Batch [99]	Speed: 11646.721144 samples/sec	accuracy=0.519678	lr=1.255942
Epoch[186] Batch [149]	Speed: 11379.741490 samples/sec	accuracy=0.518512	lr=1.255022
Epoch[186] Batch [199]	Speed: 11136.360249 samples/sec	accuracy=0.517705	lr=1.254102
Epoch[186] Batch [249]	Speed: 10912.010684 samples/sec	accuracy=0.516121	lr=1.253182
Epoch[186] Batch [299]	Speed: 9784.816233 samples/sec	accuracy=0.515955	lr=1.252263
Epoch[186] Batch [349]	Speed: 11168.078326 samples/sec	accuracy=0.515494	lr=1.251343
Epoch[186] Batch [399]	Speed: 11397.307832 samples/sec	accuracy=0.515426	lr=1.250423
Epoch[186] Batch [449]	Speed: 11149.319537 samples/sec	accuracy=0.515030	lr=1.249503
Epoch[186] Batch [499]	Speed: 11088.355545 samples/sec	accuracy=0.514535	lr=1.248584
Epoch[186] Batch [549]	Speed: 10556.592000 samples/sec	accuracy=0.514221	lr=1.247664
Epoch[186] Batch [599]	Speed: 11334.640744 samples/sec	accuracy=0.514019	lr=1.246745
[Epoch 186] training: accuracy=0.513902
[Epoch 186] speed: 11006 samples/sec	time cost: 120.904490
[Epoch 186] validation: err-top1=0.509725 err-top5=0.256755
Epoch[187] Batch [49]	Speed: 11278.494423 samples/sec	accuracy=0.522803	lr=1.245347
Epoch[187] Batch [99]	Speed: 11733.681101 samples/sec	accuracy=0.521167	lr=1.244427
Epoch[187] Batch [149]	Speed: 11275.987006 samples/sec	accuracy=0.519769	lr=1.243508
Epoch[187] Batch [199]	Speed: 10781.480534 samples/sec	accuracy=0.519016	lr=1.242588
Epoch[187] Batch [249]	Speed: 10920.806150 samples/sec	accuracy=0.517576	lr=1.241669
Epoch[187] Batch [299]	Speed: 11421.856121 samples/sec	accuracy=0.517450	lr=1.240749
Epoch[187] Batch [349]	Speed: 11543.872293 samples/sec	accuracy=0.516837	lr=1.239830
Epoch[187] Batch [399]	Speed: 11631.341187 samples/sec	accuracy=0.515959	lr=1.238911
Epoch[187] Batch [449]	Speed: 11092.839759 samples/sec	accuracy=0.515645	lr=1.237991
Epoch[187] Batch [499]	Speed: 11455.872420 samples/sec	accuracy=0.515632	lr=1.237072
Epoch[187] Batch [549]	Speed: 10183.502894 samples/sec	accuracy=0.515440	lr=1.236153
Epoch[187] Batch [599]	Speed: 10845.885774 samples/sec	accuracy=0.515041	lr=1.235234
[Epoch 187] training: accuracy=0.514868
[Epoch 187] speed: 11149 samples/sec	time cost: 119.536607
[Epoch 187] validation: err-top1=0.462179 err-top5=0.216390
Epoch[188] Batch [49]	Speed: 11528.195069 samples/sec	accuracy=0.523828	lr=1.233836
Epoch[188] Batch [99]	Speed: 11158.427669 samples/sec	accuracy=0.521577	lr=1.232917
Epoch[188] Batch [149]	Speed: 10923.194470 samples/sec	accuracy=0.520485	lr=1.231998
Epoch[188] Batch [199]	Speed: 11486.412449 samples/sec	accuracy=0.519680	lr=1.231079
Epoch[188] Batch [249]	Speed: 11187.940668 samples/sec	accuracy=0.519312	lr=1.230160
Epoch[188] Batch [299]	Speed: 10906.788671 samples/sec	accuracy=0.518791	lr=1.229241
Epoch[188] Batch [349]	Speed: 11135.800385 samples/sec	accuracy=0.518058	lr=1.228322
Epoch[188] Batch [399]	Speed: 11102.719070 samples/sec	accuracy=0.517468	lr=1.227403
Epoch[188] Batch [449]	Speed: 10355.496900 samples/sec	accuracy=0.516796	lr=1.226484
Epoch[188] Batch [499]	Speed: 11380.387667 samples/sec	accuracy=0.516537	lr=1.225565
Epoch[188] Batch [549]	Speed: 10803.398375 samples/sec	accuracy=0.516197	lr=1.224647
Epoch[188] Batch [599]	Speed: 11251.388414 samples/sec	accuracy=0.515758	lr=1.223728
[Epoch 188] training: accuracy=0.515616
[Epoch 188] speed: 11087 samples/sec	time cost: 120.174989
[Epoch 188] validation: err-top1=0.458066 err-top5=0.214082
Epoch[189] Batch [49]	Speed: 10789.910887 samples/sec	accuracy=0.521758	lr=1.222350
Epoch[189] Batch [99]	Speed: 11158.899933 samples/sec	accuracy=0.520674	lr=1.221431
Epoch[189] Batch [149]	Speed: 10433.725108 samples/sec	accuracy=0.519857	lr=1.220512
Epoch[189] Batch [199]	Speed: 10784.161379 samples/sec	accuracy=0.518591	lr=1.219594
Epoch[189] Batch [249]	Speed: 11064.211909 samples/sec	accuracy=0.517871	lr=1.218675
Epoch[189] Batch [299]	Speed: 11431.839013 samples/sec	accuracy=0.517308	lr=1.217757
Epoch[189] Batch [349]	Speed: 11441.495249 samples/sec	accuracy=0.517123	lr=1.216838
Epoch[189] Batch [399]	Speed: 11154.777590 samples/sec	accuracy=0.516688	lr=1.215920
Epoch[189] Batch [449]	Speed: 11402.521322 samples/sec	accuracy=0.516432	lr=1.215001
Epoch[189] Batch [499]	Speed: 11414.272925 samples/sec	accuracy=0.516047	lr=1.214083
Epoch[189] Batch [549]	Speed: 10412.092160 samples/sec	accuracy=0.515629	lr=1.213164
Epoch[189] Batch [599]	Speed: 11148.838822 samples/sec	accuracy=0.515570	lr=1.212246
[Epoch 189] training: accuracy=0.515293
[Epoch 189] speed: 11018 samples/sec	time cost: 120.946879
[Epoch 189] validation: err-top1=0.451721 err-top5=0.209900
Epoch[190] Batch [49]	Speed: 10911.675793 samples/sec	accuracy=0.525410	lr=1.210851
Epoch[190] Batch [99]	Speed: 11077.527696 samples/sec	accuracy=0.523447	lr=1.209932
Epoch[190] Batch [149]	Speed: 10580.266891 samples/sec	accuracy=0.521667	lr=1.209014
Epoch[190] Batch [199]	Speed: 10501.454752 samples/sec	accuracy=0.521152	lr=1.208096
Epoch[190] Batch [249]	Speed: 9754.344881 samples/sec	accuracy=0.519893	lr=1.207178
Epoch[190] Batch [299]	Speed: 10958.990955 samples/sec	accuracy=0.519383	lr=1.206260
Epoch[190] Batch [349]	Speed: 10955.498112 samples/sec	accuracy=0.518740	lr=1.205342
Epoch[190] Batch [399]	Speed: 10673.992841 samples/sec	accuracy=0.518293	lr=1.204424
Epoch[190] Batch [449]	Speed: 11449.392410 samples/sec	accuracy=0.517760	lr=1.203506
Epoch[190] Batch [499]	Speed: 10676.902093 samples/sec	accuracy=0.517521	lr=1.202589
Epoch[190] Batch [549]	Speed: 10434.467307 samples/sec	accuracy=0.517123	lr=1.201671
Epoch[190] Batch [599]	Speed: 10917.648418 samples/sec	accuracy=0.516764	lr=1.200753
[Epoch 190] training: accuracy=0.516497
[Epoch 190] speed: 10720 samples/sec	time cost: 123.885854
[Epoch 190] validation: err-top1=0.490781 err-top5=0.241133
Epoch[191] Batch [49]	Speed: 10099.629420 samples/sec	accuracy=0.525469	lr=1.199377
Epoch[191] Batch [99]	Speed: 11292.739851 samples/sec	accuracy=0.524019	lr=1.198459
Epoch[191] Batch [149]	Speed: 11512.605732 samples/sec	accuracy=0.523180	lr=1.197542
Epoch[191] Batch [199]	Speed: 11322.279653 samples/sec	accuracy=0.521553	lr=1.196624
Epoch[191] Batch [249]	Speed: 11527.009145 samples/sec	accuracy=0.520846	lr=1.195707
Epoch[191] Batch [299]	Speed: 11537.504053 samples/sec	accuracy=0.520275	lr=1.194789
Epoch[191] Batch [349]	Speed: 11220.930374 samples/sec	accuracy=0.519581	lr=1.193872
Epoch[191] Batch [399]	Speed: 11711.910698 samples/sec	accuracy=0.519038	lr=1.192955
Epoch[191] Batch [449]	Speed: 10883.379597 samples/sec	accuracy=0.518652	lr=1.192038
Epoch[191] Batch [499]	Speed: 11509.143115 samples/sec	accuracy=0.518658	lr=1.191120
Epoch[191] Batch [549]	Speed: 11556.650408 samples/sec	accuracy=0.518384	lr=1.190203
Epoch[191] Batch [599]	Speed: 11392.248681 samples/sec	accuracy=0.518070	lr=1.189286
[Epoch 191] training: accuracy=0.517878
[Epoch 191] speed: 11222 samples/sec	time cost: 118.522252
[Epoch 191] validation: err-top1=0.464132 err-top5=0.216736
Epoch[192] Batch [49]	Speed: 9675.312067 samples/sec	accuracy=0.526162	lr=1.187893
Epoch[192] Batch [99]	Speed: 11115.289164 samples/sec	accuracy=0.524668	lr=1.186976
Epoch[192] Batch [149]	Speed: 11352.315861 samples/sec	accuracy=0.523447	lr=1.186059
Epoch[192] Batch [199]	Speed: 11398.108154 samples/sec	accuracy=0.523203	lr=1.185142
Epoch[192] Batch [249]	Speed: 10962.926180 samples/sec	accuracy=0.522258	lr=1.184225
Epoch[192] Batch [299]	Speed: 11536.592930 samples/sec	accuracy=0.521603	lr=1.183309
Epoch[192] Batch [349]	Speed: 11341.697354 samples/sec	accuracy=0.520776	lr=1.182392
Epoch[192] Batch [399]	Speed: 11439.060468 samples/sec	accuracy=0.520229	lr=1.181475
Epoch[192] Batch [449]	Speed: 10635.523543 samples/sec	accuracy=0.520242	lr=1.180559
Epoch[192] Batch [499]	Speed: 11531.908190 samples/sec	accuracy=0.519950	lr=1.179643
Epoch[192] Batch [549]	Speed: 10382.132328 samples/sec	accuracy=0.519484	lr=1.178726
Epoch[192] Batch [599]	Speed: 10715.209896 samples/sec	accuracy=0.519021	lr=1.177810
[Epoch 192] training: accuracy=0.518767
[Epoch 192] speed: 10961 samples/sec	time cost: 121.329154
[Epoch 192] validation: err-top1=0.466471 err-top5=0.220459
Epoch[193] Batch [49]	Speed: 11106.260770 samples/sec	accuracy=0.528145	lr=1.176435
Epoch[193] Batch [99]	Speed: 11454.576992 samples/sec	accuracy=0.525098	lr=1.175519
Epoch[193] Batch [149]	Speed: 11015.216775 samples/sec	accuracy=0.523955	lr=1.174603
Epoch[193] Batch [199]	Speed: 10867.457695 samples/sec	accuracy=0.523108	lr=1.173687
Epoch[193] Batch [249]	Speed: 11056.092129 samples/sec	accuracy=0.522305	lr=1.172771
Epoch[193] Batch [299]	Speed: 11360.794702 samples/sec	accuracy=0.521270	lr=1.171855
Epoch[193] Batch [349]	Speed: 10781.582838 samples/sec	accuracy=0.520819	lr=1.170939
Epoch[193] Batch [399]	Speed: 11006.893125 samples/sec	accuracy=0.520142	lr=1.170024
Epoch[193] Batch [449]	Speed: 10283.849323 samples/sec	accuracy=0.519743	lr=1.169108
Epoch[193] Batch [499]	Speed: 10218.450232 samples/sec	accuracy=0.519273	lr=1.168192
Epoch[193] Batch [549]	Speed: 11232.085168 samples/sec	accuracy=0.518919	lr=1.167277
Epoch[193] Batch [599]	Speed: 11463.309380 samples/sec	accuracy=0.518370	lr=1.166361
[Epoch 193] training: accuracy=0.518161
[Epoch 193] speed: 10960 samples/sec	time cost: 121.713774
[Epoch 193] validation: err-top1=0.464531 err-top5=0.219160
Epoch[194] Batch [49]	Speed: 11233.178275 samples/sec	accuracy=0.527314	lr=1.164970
Epoch[194] Batch [99]	Speed: 10644.168670 samples/sec	accuracy=0.526074	lr=1.164054
Epoch[194] Batch [149]	Speed: 11064.943329 samples/sec	accuracy=0.525420	lr=1.163139
Epoch[194] Batch [199]	Speed: 11568.094785 samples/sec	accuracy=0.523530	lr=1.162224
Epoch[194] Batch [249]	Speed: 11183.741221 samples/sec	accuracy=0.522775	lr=1.161309
Epoch[194] Batch [299]	Speed: 10196.626636 samples/sec	accuracy=0.521738	lr=1.160394
Epoch[194] Batch [349]	Speed: 10636.600551 samples/sec	accuracy=0.521335	lr=1.159479
Epoch[194] Batch [399]	Speed: 11518.951450 samples/sec	accuracy=0.521021	lr=1.158564
Epoch[194] Batch [449]	Speed: 11235.623481 samples/sec	accuracy=0.520307	lr=1.157649
Epoch[194] Batch [499]	Speed: 11304.511906 samples/sec	accuracy=0.520429	lr=1.156734
Epoch[194] Batch [549]	Speed: 11093.208497 samples/sec	accuracy=0.519858	lr=1.155819
Epoch[194] Batch [599]	Speed: 11330.857751 samples/sec	accuracy=0.519598	lr=1.154905
[Epoch 194] training: accuracy=0.519425
[Epoch 194] speed: 11049 samples/sec	time cost: 120.481938
[Epoch 194] validation: err-top1=0.468770 err-top5=0.224670
Epoch[195] Batch [49]	Speed: 11331.853566 samples/sec	accuracy=0.528252	lr=1.153515
Epoch[195] Batch [99]	Speed: 10893.304146 samples/sec	accuracy=0.525854	lr=1.152600
Epoch[195] Batch [149]	Speed: 11582.929528 samples/sec	accuracy=0.524873	lr=1.151686
Epoch[195] Batch [199]	Speed: 10352.665066 samples/sec	accuracy=0.524192	lr=1.150772
Epoch[195] Batch [249]	Speed: 11133.012297 samples/sec	accuracy=0.523664	lr=1.149857
Epoch[195] Batch [299]	Speed: 10798.797797 samples/sec	accuracy=0.522785	lr=1.148943
Epoch[195] Batch [349]	Speed: 11051.759569 samples/sec	accuracy=0.522118	lr=1.148029
Epoch[195] Batch [399]	Speed: 11400.337612 samples/sec	accuracy=0.521670	lr=1.147115
Epoch[195] Batch [449]	Speed: 10238.797993 samples/sec	accuracy=0.520891	lr=1.146201
Epoch[195] Batch [499]	Speed: 11227.757424 samples/sec	accuracy=0.520644	lr=1.145287
Epoch[195] Batch [549]	Speed: 11449.127490 samples/sec	accuracy=0.520401	lr=1.144373
Epoch[195] Batch [599]	Speed: 11401.081463 samples/sec	accuracy=0.519972	lr=1.143460
[Epoch 195] training: accuracy=0.519986
[Epoch 195] speed: 11026 samples/sec	time cost: 120.647241
[Epoch 195] validation: err-top1=0.472012 err-top5=0.226426
Epoch[196] Batch [49]	Speed: 10831.051598 samples/sec	accuracy=0.528193	lr=1.142089
Epoch[196] Batch [99]	Speed: 11244.949771 samples/sec	accuracy=0.527427	lr=1.141176
Epoch[196] Batch [149]	Speed: 10642.153144 samples/sec	accuracy=0.525824	lr=1.140262
Epoch[196] Batch [199]	Speed: 11076.624068 samples/sec	accuracy=0.524778	lr=1.139349
Epoch[196] Batch [249]	Speed: 11244.397776 samples/sec	accuracy=0.523254	lr=1.138436
Epoch[196] Batch [299]	Speed: 11237.709845 samples/sec	accuracy=0.523311	lr=1.137523
Epoch[196] Batch [349]	Speed: 11234.903708 samples/sec	accuracy=0.522973	lr=1.136610
Epoch[196] Batch [399]	Speed: 11159.605071 samples/sec	accuracy=0.522148	lr=1.135697
Epoch[196] Batch [449]	Speed: 10992.090973 samples/sec	accuracy=0.521603	lr=1.134784
Epoch[196] Batch [499]	Speed: 11626.919434 samples/sec	accuracy=0.521518	lr=1.133871
Epoch[196] Batch [549]	Speed: 11562.502833 samples/sec	accuracy=0.521265	lr=1.132958
Epoch[196] Batch [599]	Speed: 11545.613804 samples/sec	accuracy=0.520942	lr=1.132045
[Epoch 196] training: accuracy=0.520839
[Epoch 196] speed: 11112 samples/sec	time cost: 119.735740
[Epoch 196] validation: err-top1=0.457499 err-top5=0.214294
Epoch[197] Batch [49]	Speed: 10885.778060 samples/sec	accuracy=0.529502	lr=1.130658
Epoch[197] Batch [99]	Speed: 10084.538653 samples/sec	accuracy=0.527441	lr=1.129746
Epoch[197] Batch [149]	Speed: 10835.522519 samples/sec	accuracy=0.526120	lr=1.128833
Epoch[197] Batch [199]	Speed: 11291.627700 samples/sec	accuracy=0.525022	lr=1.127921
Epoch[197] Batch [249]	Speed: 11093.644884 samples/sec	accuracy=0.524301	lr=1.127009
Epoch[197] Batch [299]	Speed: 10949.264891 samples/sec	accuracy=0.523358	lr=1.126097
Epoch[197] Batch [349]	Speed: 11087.744681 samples/sec	accuracy=0.522764	lr=1.125185
Epoch[197] Batch [399]	Speed: 11013.976435 samples/sec	accuracy=0.522472	lr=1.124273
Epoch[197] Batch [449]	Speed: 10740.949327 samples/sec	accuracy=0.521995	lr=1.123361
Epoch[197] Batch [499]	Speed: 10629.829747 samples/sec	accuracy=0.521354	lr=1.122449
Epoch[197] Batch [549]	Speed: 11143.607179 samples/sec	accuracy=0.521160	lr=1.121537
Epoch[197] Batch [599]	Speed: 11497.406896 samples/sec	accuracy=0.520929	lr=1.120626
[Epoch 197] training: accuracy=0.520881
[Epoch 197] speed: 10905 samples/sec	time cost: 121.547898
[Epoch 197] validation: err-top1=0.484843 err-top5=0.243164
Epoch[198] Batch [49]	Speed: 10941.715106 samples/sec	accuracy=0.531396	lr=1.119258
Epoch[198] Batch [99]	Speed: 11374.934100 samples/sec	accuracy=0.530269	lr=1.118347
Epoch[198] Batch [149]	Speed: 11366.806259 samples/sec	accuracy=0.527946	lr=1.117436
Epoch[198] Batch [199]	Speed: 10936.266078 samples/sec	accuracy=0.526592	lr=1.116525
Epoch[198] Batch [249]	Speed: 11170.167857 samples/sec	accuracy=0.526453	lr=1.115614
Epoch[198] Batch [299]	Speed: 11254.732156 samples/sec	accuracy=0.525638	lr=1.114703
Epoch[198] Batch [349]	Speed: 11004.586490 samples/sec	accuracy=0.525052	lr=1.113792
Epoch[198] Batch [399]	Speed: 10256.035325 samples/sec	accuracy=0.523829	lr=1.112881
Epoch[198] Batch [449]	Speed: 11339.082427 samples/sec	accuracy=0.523483	lr=1.111970
Epoch[198] Batch [499]	Speed: 10868.334117 samples/sec	accuracy=0.523121	lr=1.111059
Epoch[198] Batch [549]	Speed: 11393.284028 samples/sec	accuracy=0.522573	lr=1.110149
Epoch[198] Batch [599]	Speed: 11537.871952 samples/sec	accuracy=0.522446	lr=1.109238
[Epoch 198] training: accuracy=0.522225
[Epoch 198] speed: 11093 samples/sec	time cost: 120.082619
[Epoch 198] validation: err-top1=0.489355 err-top5=0.238340
Epoch[199] Batch [49]	Speed: 11044.290477 samples/sec	accuracy=0.527500	lr=1.107855
Epoch[199] Batch [99]	Speed: 11082.171862 samples/sec	accuracy=0.527168	lr=1.106945
Epoch[199] Batch [149]	Speed: 11410.238651 samples/sec	accuracy=0.526729	lr=1.106034
Epoch[199] Batch [199]	Speed: 11630.124499 samples/sec	accuracy=0.526411	lr=1.105124
Epoch[199] Batch [249]	Speed: 10760.215370 samples/sec	accuracy=0.525658	lr=1.104215
Epoch[199] Batch [299]	Speed: 11281.881858 samples/sec	accuracy=0.525656	lr=1.103305
Epoch[199] Batch [349]	Speed: 10956.403046 samples/sec	accuracy=0.525059	lr=1.102395
Epoch[199] Batch [399]	Speed: 11233.303727 samples/sec	accuracy=0.524230	lr=1.101485
Epoch[199] Batch [449]	Speed: 11154.436323 samples/sec	accuracy=0.523751	lr=1.100576
Epoch[199] Batch [499]	Speed: 11052.292812 samples/sec	accuracy=0.523577	lr=1.099666
Epoch[199] Batch [549]	Speed: 11226.587019 samples/sec	accuracy=0.523120	lr=1.098757
Epoch[199] Batch [599]	Speed: 10922.516113 samples/sec	accuracy=0.522635	lr=1.097848
[Epoch 199] training: accuracy=0.522489
[Epoch 199] speed: 11129 samples/sec	time cost: 119.406080
[Epoch 199] validation: err-top1=0.474955 err-top5=0.227844
Epoch[200] Batch [49]	Speed: 10335.669973 samples/sec	accuracy=0.532461	lr=1.096484
Epoch[200] Batch [99]	Speed: 11584.309143 samples/sec	accuracy=0.529165	lr=1.095575
Epoch[200] Batch [149]	Speed: 10082.019664 samples/sec	accuracy=0.528464	lr=1.094666
Epoch[200] Batch [199]	Speed: 11377.283490 samples/sec	accuracy=0.527490	lr=1.093758
Epoch[200] Batch [249]	Speed: 11409.062929 samples/sec	accuracy=0.526691	lr=1.092849
Epoch[200] Batch [299]	Speed: 11489.019559 samples/sec	accuracy=0.526322	lr=1.091941
Epoch[200] Batch [349]	Speed: 11191.621235 samples/sec	accuracy=0.525841	lr=1.091032
Epoch[200] Batch [399]	Speed: 10874.702111 samples/sec	accuracy=0.525471	lr=1.090124
Epoch[200] Batch [449]	Speed: 11651.303998 samples/sec	accuracy=0.524911	lr=1.089216
Epoch[200] Batch [499]	Speed: 10621.045831 samples/sec	accuracy=0.524229	lr=1.088307
Epoch[200] Batch [549]	Speed: 11048.002447 samples/sec	accuracy=0.523970	lr=1.087399
Epoch[200] Batch [599]	Speed: 11112.972822 samples/sec	accuracy=0.523629	lr=1.086491
[Epoch 200] training: accuracy=0.523557
[Epoch 200] speed: 10975 samples/sec	time cost: 121.495174
[Epoch 200] validation: err-top1=0.475371 err-top5=0.230176
Epoch[201] Batch [49]	Speed: 10667.667695 samples/sec	accuracy=0.529160	lr=1.085112
Epoch[201] Batch [99]	Speed: 10644.668845 samples/sec	accuracy=0.529829	lr=1.084204
Epoch[201] Batch [149]	Speed: 11053.220920 samples/sec	accuracy=0.529593	lr=1.083296
Epoch[201] Batch [199]	Speed: 10733.524117 samples/sec	accuracy=0.527805	lr=1.082389
Epoch[201] Batch [249]	Speed: 10046.294463 samples/sec	accuracy=0.527240	lr=1.081482
Epoch[201] Batch [299]	Speed: 11278.385137 samples/sec	accuracy=0.526717	lr=1.080575
Epoch[201] Batch [349]	Speed: 10910.263268 samples/sec	accuracy=0.526161	lr=1.079667
Epoch[201] Batch [399]	Speed: 11383.848060 samples/sec	accuracy=0.525541	lr=1.078760
Epoch[201] Batch [449]	Speed: 11282.971935 samples/sec	accuracy=0.525498	lr=1.077854
Epoch[201] Batch [499]	Speed: 11001.773812 samples/sec	accuracy=0.525271	lr=1.076947
Epoch[201] Batch [549]	Speed: 10952.835878 samples/sec	accuracy=0.524950	lr=1.076040
Epoch[201] Batch [599]	Speed: 11421.665978 samples/sec	accuracy=0.524504	lr=1.075134
[Epoch 201] training: accuracy=0.524263
[Epoch 201] speed: 10907 samples/sec	time cost: 121.767891
[Epoch 201] validation: err-top1=0.487813 err-top5=0.240173
Epoch[202] Batch [49]	Speed: 11368.907324 samples/sec	accuracy=0.532803	lr=1.073756
Epoch[202] Batch [99]	Speed: 11303.571166 samples/sec	accuracy=0.531968	lr=1.072850
Epoch[202] Batch [149]	Speed: 11545.605735 samples/sec	accuracy=0.530316	lr=1.071943
Epoch[202] Batch [199]	Speed: 11000.053031 samples/sec	accuracy=0.529429	lr=1.071037
Epoch[202] Batch [249]	Speed: 11169.942426 samples/sec	accuracy=0.528295	lr=1.070132
Epoch[202] Batch [299]	Speed: 11433.998587 samples/sec	accuracy=0.527759	lr=1.069226
Epoch[202] Batch [349]	Speed: 11669.017141 samples/sec	accuracy=0.527232	lr=1.068320
Epoch[202] Batch [399]	Speed: 11378.914502 samples/sec	accuracy=0.526709	lr=1.067414
Epoch[202] Batch [449]	Speed: 11245.446465 samples/sec	accuracy=0.526366	lr=1.066509
Epoch[202] Batch [499]	Speed: 11632.187947 samples/sec	accuracy=0.525824	lr=1.065604
Epoch[202] Batch [549]	Speed: 11091.727102 samples/sec	accuracy=0.525609	lr=1.064698
Epoch[202] Batch [599]	Speed: 11343.527891 samples/sec	accuracy=0.525447	lr=1.063793
[Epoch 202] training: accuracy=0.525166
[Epoch 202] speed: 11317 samples/sec	time cost: 117.584685
[Epoch 202] validation: err-top1=0.444512 err-top5=0.205859
Epoch[203] Batch [49]	Speed: 10243.089555 samples/sec	accuracy=0.536592	lr=1.062436
Epoch[203] Batch [99]	Speed: 11587.423856 samples/sec	accuracy=0.535508	lr=1.061531
Epoch[203] Batch [149]	Speed: 11162.270438 samples/sec	accuracy=0.533675	lr=1.060626
Epoch[203] Batch [199]	Speed: 11303.650894 samples/sec	accuracy=0.532163	lr=1.059722
Epoch[203] Batch [249]	Speed: 10957.246072 samples/sec	accuracy=0.531129	lr=1.058817
Epoch[203] Batch [299]	Speed: 9795.528019 samples/sec	accuracy=0.530389	lr=1.057913
Epoch[203] Batch [349]	Speed: 11159.662194 samples/sec	accuracy=0.529614	lr=1.057009
Epoch[203] Batch [399]	Speed: 11531.725202 samples/sec	accuracy=0.528595	lr=1.056105
Epoch[203] Batch [449]	Speed: 10917.832418 samples/sec	accuracy=0.527835	lr=1.055201
Epoch[203] Batch [499]	Speed: 11048.645889 samples/sec	accuracy=0.527146	lr=1.054297
Epoch[203] Batch [549]	Speed: 10676.783983 samples/sec	accuracy=0.526807	lr=1.053393
Epoch[203] Batch [599]	Speed: 11333.006251 samples/sec	accuracy=0.526305	lr=1.052490
[Epoch 203] training: accuracy=0.526124
[Epoch 203] speed: 10954 samples/sec	time cost: 121.430977
[Epoch 203] validation: err-top1=0.447795 err-top5=0.207845
Epoch[204] Batch [49]	Speed: 10935.543218 samples/sec	accuracy=0.534883	lr=1.051117
Epoch[204] Batch [99]	Speed: 10884.686966 samples/sec	accuracy=0.532407	lr=1.050213
Epoch[204] Batch [149]	Speed: 10591.158315 samples/sec	accuracy=0.531725	lr=1.049310
Epoch[204] Batch [199]	Speed: 10662.168652 samples/sec	accuracy=0.530527	lr=1.048407
Epoch[204] Batch [249]	Speed: 10079.250964 samples/sec	accuracy=0.530117	lr=1.047504
Epoch[204] Batch [299]	Speed: 10673.280629 samples/sec	accuracy=0.529557	lr=1.046601
Epoch[204] Batch [349]	Speed: 11300.387736 samples/sec	accuracy=0.528774	lr=1.045699
Epoch[204] Batch [399]	Speed: 11356.941680 samples/sec	accuracy=0.528586	lr=1.044796
Epoch[204] Batch [449]	Speed: 11123.516326 samples/sec	accuracy=0.528081	lr=1.043894
Epoch[204] Batch [499]	Speed: 11296.093366 samples/sec	accuracy=0.527545	lr=1.042992
Epoch[204] Batch [549]	Speed: 11386.709782 samples/sec	accuracy=0.527238	lr=1.042090
Epoch[204] Batch [599]	Speed: 11580.970953 samples/sec	accuracy=0.527030	lr=1.041188
[Epoch 204] training: accuracy=0.526997
[Epoch 204] speed: 10969 samples/sec	time cost: 121.243201
[Epoch 204] validation: err-top1=0.473999 err-top5=0.228312
Epoch[205] Batch [49]	Speed: 11220.438481 samples/sec	accuracy=0.533740	lr=1.039835
Epoch[205] Batch [99]	Speed: 10911.977138 samples/sec	accuracy=0.532466	lr=1.038933
Epoch[205] Batch [149]	Speed: 11281.133627 samples/sec	accuracy=0.531279	lr=1.038032
Epoch[205] Batch [199]	Speed: 11362.778709 samples/sec	accuracy=0.530249	lr=1.037130
Epoch[205] Batch [249]	Speed: 10675.393934 samples/sec	accuracy=0.529520	lr=1.036229
Epoch[205] Batch [299]	Speed: 11163.894935 samples/sec	accuracy=0.529435	lr=1.035328
Epoch[205] Batch [349]	Speed: 10879.348865 samples/sec	accuracy=0.529069	lr=1.034427
Epoch[205] Batch [399]	Speed: 11247.579483 samples/sec	accuracy=0.528499	lr=1.033526
Epoch[205] Batch [449]	Speed: 11562.383927 samples/sec	accuracy=0.528167	lr=1.032625
Epoch[205] Batch [499]	Speed: 11538.171372 samples/sec	accuracy=0.527692	lr=1.031724
Epoch[205] Batch [549]	Speed: 10689.217618 samples/sec	accuracy=0.527288	lr=1.030824
Epoch[205] Batch [599]	Speed: 11433.588887 samples/sec	accuracy=0.527015	lr=1.029924
[Epoch 205] training: accuracy=0.526773
[Epoch 205] speed: 11127 samples/sec	time cost: 119.992440
[Epoch 205] validation: err-top1=0.449766 err-top5=0.208633
Epoch[206] Batch [49]	Speed: 11127.301931 samples/sec	accuracy=0.535840	lr=1.028555
Epoch[206] Batch [99]	Speed: 10832.400246 samples/sec	accuracy=0.533921	lr=1.027655
Epoch[206] Batch [149]	Speed: 11486.335345 samples/sec	accuracy=0.533441	lr=1.026756
Epoch[206] Batch [199]	Speed: 11219.991183 samples/sec	accuracy=0.532444	lr=1.025856
Epoch[206] Batch [249]	Speed: 11033.113119 samples/sec	accuracy=0.531869	lr=1.024956
Epoch[206] Batch [299]	Speed: 11148.670104 samples/sec	accuracy=0.531058	lr=1.024057
Epoch[206] Batch [349]	Speed: 11387.438269 samples/sec	accuracy=0.530091	lr=1.023157
Epoch[206] Batch [399]	Speed: 11297.078919 samples/sec	accuracy=0.529823	lr=1.022258
Epoch[206] Batch [449]	Speed: 11192.717564 samples/sec	accuracy=0.529231	lr=1.021359
Epoch[206] Batch [499]	Speed: 10139.622834 samples/sec	accuracy=0.529013	lr=1.020460
Epoch[206] Batch [549]	Speed: 11320.449701 samples/sec	accuracy=0.528764	lr=1.019562
Epoch[206] Batch [599]	Speed: 11010.367733 samples/sec	accuracy=0.528455	lr=1.018663
[Epoch 206] training: accuracy=0.528140
[Epoch 206] speed: 11017 samples/sec	time cost: 120.603231
[Epoch 206] validation: err-top1=0.479370 err-top5=0.233988
Epoch[207] Batch [49]	Speed: 10626.326374 samples/sec	accuracy=0.532871	lr=1.017315
Epoch[207] Batch [99]	Speed: 11128.565909 samples/sec	accuracy=0.533838	lr=1.016417
Epoch[207] Batch [149]	Speed: 11215.911162 samples/sec	accuracy=0.532588	lr=1.015519
Epoch[207] Batch [199]	Speed: 10998.478117 samples/sec	accuracy=0.532480	lr=1.014621
Epoch[207] Batch [249]	Speed: 11169.057932 samples/sec	accuracy=0.531988	lr=1.013723
Epoch[207] Batch [299]	Speed: 10997.689280 samples/sec	accuracy=0.531533	lr=1.012825
Epoch[207] Batch [349]	Speed: 11120.772970 samples/sec	accuracy=0.530984	lr=1.011928
Epoch[207] Batch [399]	Speed: 11418.037766 samples/sec	accuracy=0.530864	lr=1.011030
Epoch[207] Batch [449]	Speed: 10767.220021 samples/sec	accuracy=0.530339	lr=1.010133
Epoch[207] Batch [499]	Speed: 11238.586719 samples/sec	accuracy=0.529862	lr=1.009236
Epoch[207] Batch [549]	Speed: 11429.797360 samples/sec	accuracy=0.529268	lr=1.008339
Epoch[207] Batch [599]	Speed: 11557.945700 samples/sec	accuracy=0.528735	lr=1.007442
[Epoch 207] training: accuracy=0.528605
[Epoch 207] speed: 11110 samples/sec	time cost: 120.016711
[Epoch 207] validation: err-top1=0.444883 err-top5=0.202148
Epoch[208] Batch [49]	Speed: 9609.301176 samples/sec	accuracy=0.538066	lr=1.006079
Epoch[208] Batch [99]	Speed: 11083.488818 samples/sec	accuracy=0.536016	lr=1.005183
Epoch[208] Batch [149]	Speed: 11355.211285 samples/sec	accuracy=0.534714	lr=1.004287
Epoch[208] Batch [199]	Speed: 11414.430666 samples/sec	accuracy=0.534041	lr=1.003390
Epoch[208] Batch [249]	Speed: 11547.542120 samples/sec	accuracy=0.532760	lr=1.002494
Epoch[208] Batch [299]	Speed: 11535.862587 samples/sec	accuracy=0.532393	lr=1.001599
Epoch[208] Batch [349]	Speed: 11486.573420 samples/sec	accuracy=0.532167	lr=1.000703
Epoch[208] Batch [399]	Speed: 11640.827030 samples/sec	accuracy=0.531726	lr=0.999807
Epoch[208] Batch [449]	Speed: 11167.805647 samples/sec	accuracy=0.531141	lr=0.998912
Epoch[208] Batch [499]	Speed: 11363.758795 samples/sec	accuracy=0.530594	lr=0.998017
Epoch[208] Batch [549]	Speed: 11061.360127 samples/sec	accuracy=0.530213	lr=0.997122
Epoch[208] Batch [599]	Speed: 10909.138166 samples/sec	accuracy=0.529633	lr=0.996227
[Epoch 208] training: accuracy=0.529429
[Epoch 208] speed: 11116 samples/sec	time cost: 119.708771
[Epoch 208] validation: err-top1=0.472249 err-top5=0.226624
Epoch[209] Batch [49]	Speed: 10340.629916 samples/sec	accuracy=0.540684	lr=0.994867
Epoch[209] Batch [99]	Speed: 11363.309918 samples/sec	accuracy=0.538096	lr=0.993972
Epoch[209] Batch [149]	Speed: 11346.602282 samples/sec	accuracy=0.537116	lr=0.993078
Epoch[209] Batch [199]	Speed: 11463.427174 samples/sec	accuracy=0.535249	lr=0.992183
Epoch[209] Batch [249]	Speed: 11036.326096 samples/sec	accuracy=0.534492	lr=0.991289
Epoch[209] Batch [299]	Speed: 11242.417811 samples/sec	accuracy=0.533840	lr=0.990395
Epoch[209] Batch [349]	Speed: 11908.441702 samples/sec	accuracy=0.533378	lr=0.989501
Epoch[209] Batch [399]	Speed: 11262.727920 samples/sec	accuracy=0.532372	lr=0.988608
Epoch[209] Batch [449]	Speed: 11532.828792 samples/sec	accuracy=0.531650	lr=0.987714
Epoch[209] Batch [499]	Speed: 11466.658444 samples/sec	accuracy=0.531169	lr=0.986821
Epoch[209] Batch [549]	Speed: 10936.591897 samples/sec	accuracy=0.530711	lr=0.985928
Epoch[209] Batch [599]	Speed: 11214.967539 samples/sec	accuracy=0.530330	lr=0.985035
[Epoch 209] training: accuracy=0.530199
[Epoch 209] speed: 11207 samples/sec	time cost: 118.594434
[Epoch 209] validation: err-top1=0.477905 err-top5=0.234924
Epoch[210] Batch [49]	Speed: 11120.806947 samples/sec	accuracy=0.536572	lr=0.983696
Epoch[210] Batch [99]	Speed: 10964.410875 samples/sec	accuracy=0.535239	lr=0.982803
Epoch[210] Batch [149]	Speed: 11185.276720 samples/sec	accuracy=0.534482	lr=0.981910
Epoch[210] Batch [199]	Speed: 11128.747284 samples/sec	accuracy=0.533528	lr=0.981018
Epoch[210] Batch [249]	Speed: 11386.778309 samples/sec	accuracy=0.532576	lr=0.980126
Epoch[210] Batch [299]	Speed: 11328.367038 samples/sec	accuracy=0.532279	lr=0.979234
Epoch[210] Batch [349]	Speed: 11438.758858 samples/sec	accuracy=0.532030	lr=0.978342
Epoch[210] Batch [399]	Speed: 11523.033300 samples/sec	accuracy=0.531467	lr=0.977451
Epoch[210] Batch [449]	Speed: 10764.654161 samples/sec	accuracy=0.531120	lr=0.976559
Epoch[210] Batch [499]	Speed: 11523.620721 samples/sec	accuracy=0.530953	lr=0.975668
Epoch[210] Batch [549]	Speed: 11025.002396 samples/sec	accuracy=0.530949	lr=0.974777
Epoch[210] Batch [599]	Speed: 11103.166826 samples/sec	accuracy=0.530802	lr=0.973886
[Epoch 210] training: accuracy=0.530536
[Epoch 210] speed: 11179 samples/sec	time cost: 119.316242
[Epoch 210] validation: err-top1=0.452422 err-top5=0.212402
Epoch[211] Batch [49]	Speed: 10757.458579 samples/sec	accuracy=0.539189	lr=0.972532
Epoch[211] Batch [99]	Speed: 11010.932557 samples/sec	accuracy=0.538008	lr=0.971641
Epoch[211] Batch [149]	Speed: 11166.348099 samples/sec	accuracy=0.537298	lr=0.970751
Epoch[211] Batch [199]	Speed: 11425.157905 samples/sec	accuracy=0.536226	lr=0.969860
Epoch[211] Batch [249]	Speed: 11221.671813 samples/sec	accuracy=0.535271	lr=0.968970
Epoch[211] Batch [299]	Speed: 11656.944862 samples/sec	accuracy=0.534746	lr=0.968080
Epoch[211] Batch [349]	Speed: 11007.336852 samples/sec	accuracy=0.533650	lr=0.967190
Epoch[211] Batch [399]	Speed: 11357.344704 samples/sec	accuracy=0.533185	lr=0.966301
Epoch[211] Batch [449]	Speed: 10650.619684 samples/sec	accuracy=0.533202	lr=0.965411
Epoch[211] Batch [499]	Speed: 11456.152014 samples/sec	accuracy=0.532729	lr=0.964522
Epoch[211] Batch [549]	Speed: 11361.678267 samples/sec	accuracy=0.532401	lr=0.963633
Epoch[211] Batch [599]	Speed: 10676.407907 samples/sec	accuracy=0.532048	lr=0.962744
[Epoch 211] training: accuracy=0.531905
[Epoch 211] speed: 11118 samples/sec	time cost: 119.410750
[Epoch 211] validation: err-top1=0.447245 err-top5=0.203125
Epoch[212] Batch [49]	Speed: 10914.717179 samples/sec	accuracy=0.537373	lr=0.961411
Epoch[212] Batch [99]	Speed: 11013.431350 samples/sec	accuracy=0.537612	lr=0.960523
Epoch[212] Batch [149]	Speed: 11478.921255 samples/sec	accuracy=0.536735	lr=0.959634
Epoch[212] Batch [199]	Speed: 11546.048333 samples/sec	accuracy=0.536782	lr=0.958746
Epoch[212] Batch [249]	Speed: 11452.806335 samples/sec	accuracy=0.536010	lr=0.957858
Epoch[212] Batch [299]	Speed: 11115.856174 samples/sec	accuracy=0.535423	lr=0.956970
Epoch[212] Batch [349]	Speed: 11499.011883 samples/sec	accuracy=0.534897	lr=0.956083
Epoch[212] Batch [399]	Speed: 11306.073308 samples/sec	accuracy=0.533971	lr=0.955195
Epoch[212] Batch [449]	Speed: 11348.074888 samples/sec	accuracy=0.533274	lr=0.954308
Epoch[212] Batch [499]	Speed: 10781.383915 samples/sec	accuracy=0.532668	lr=0.953421
Epoch[212] Batch [549]	Speed: 11350.053554 samples/sec	accuracy=0.532161	lr=0.952534
Epoch[212] Batch [599]	Speed: 11295.700916 samples/sec	accuracy=0.532065	lr=0.951647
[Epoch 212] training: accuracy=0.531967
[Epoch 212] speed: 11230 samples/sec	time cost: 118.766493
[Epoch 212] validation: err-top1=0.454805 err-top5=0.212676
Epoch[213] Batch [49]	Speed: 10969.805899 samples/sec	accuracy=0.538633	lr=0.950299
Epoch[213] Batch [99]	Speed: 11119.145453 samples/sec	accuracy=0.537983	lr=0.949413
Epoch[213] Batch [149]	Speed: 10612.381044 samples/sec	accuracy=0.537454	lr=0.948527
Epoch[213] Batch [199]	Speed: 11841.613793 samples/sec	accuracy=0.536782	lr=0.947641
Epoch[213] Batch [249]	Speed: 10951.985432 samples/sec	accuracy=0.536182	lr=0.946755
Epoch[213] Batch [299]	Speed: 11211.393347 samples/sec	accuracy=0.535859	lr=0.945869
Epoch[213] Batch [349]	Speed: 11174.581850 samples/sec	accuracy=0.535571	lr=0.944984
Epoch[213] Batch [399]	Speed: 11474.897571 samples/sec	accuracy=0.535236	lr=0.944099
Epoch[213] Batch [449]	Speed: 11703.666191 samples/sec	accuracy=0.534962	lr=0.943213
Epoch[213] Batch [499]	Speed: 11208.318948 samples/sec	accuracy=0.534480	lr=0.942329
Epoch[213] Batch [549]	Speed: 10933.857008 samples/sec	accuracy=0.533881	lr=0.941444
Epoch[213] Batch [599]	Speed: 11476.285301 samples/sec	accuracy=0.533634	lr=0.940559
[Epoch 213] training: accuracy=0.533525
[Epoch 213] speed: 11192 samples/sec	time cost: 118.678497
[Epoch 213] validation: err-top1=0.445190 err-top5=0.206767
Epoch[214] Batch [49]	Speed: 11008.388907 samples/sec	accuracy=0.541543	lr=0.939233
Epoch[214] Batch [99]	Speed: 11647.006025 samples/sec	accuracy=0.540278	lr=0.938349
Epoch[214] Batch [149]	Speed: 10809.194807 samples/sec	accuracy=0.538776	lr=0.937465
Epoch[214] Batch [199]	Speed: 11355.063882 samples/sec	accuracy=0.537578	lr=0.936581
Epoch[214] Batch [249]	Speed: 10737.643462 samples/sec	accuracy=0.537088	lr=0.935697
Epoch[214] Batch [299]	Speed: 10719.646406 samples/sec	accuracy=0.536535	lr=0.934814
Epoch[214] Batch [349]	Speed: 11190.793370 samples/sec	accuracy=0.535921	lr=0.933931
Epoch[214] Batch [399]	Speed: 10842.515283 samples/sec	accuracy=0.535414	lr=0.933048
Epoch[214] Batch [449]	Speed: 10532.856475 samples/sec	accuracy=0.535129	lr=0.932165
Epoch[214] Batch [499]	Speed: 10416.668951 samples/sec	accuracy=0.534943	lr=0.931282
Epoch[214] Batch [549]	Speed: 11023.181828 samples/sec	accuracy=0.534553	lr=0.930400
Epoch[214] Batch [599]	Speed: 11414.021154 samples/sec	accuracy=0.534166	lr=0.929517
[Epoch 214] training: accuracy=0.534071
[Epoch 214] speed: 10962 samples/sec	time cost: 121.552897
[Epoch 214] validation: err-top1=0.460000 err-top5=0.216387
Epoch[215] Batch [49]	Speed: 10957.203582 samples/sec	accuracy=0.542383	lr=0.928177
Epoch[215] Batch [99]	Speed: 10804.480027 samples/sec	accuracy=0.540664	lr=0.927295
Epoch[215] Batch [149]	Speed: 10959.898704 samples/sec	accuracy=0.540290	lr=0.926413
Epoch[215] Batch [199]	Speed: 11641.430309 samples/sec	accuracy=0.539607	lr=0.925532
Epoch[215] Batch [249]	Speed: 11068.395360 samples/sec	accuracy=0.538207	lr=0.924651
Epoch[215] Batch [299]	Speed: 11062.470118 samples/sec	accuracy=0.537972	lr=0.923770
Epoch[215] Batch [349]	Speed: 11431.355534 samples/sec	accuracy=0.537366	lr=0.922889
Epoch[215] Batch [399]	Speed: 11147.598589 samples/sec	accuracy=0.536874	lr=0.922008
Epoch[215] Batch [449]	Speed: 10876.330736 samples/sec	accuracy=0.536025	lr=0.921127
Epoch[215] Batch [499]	Speed: 10985.149004 samples/sec	accuracy=0.535507	lr=0.920247
Epoch[215] Batch [549]	Speed: 11313.847156 samples/sec	accuracy=0.534900	lr=0.919367
Epoch[215] Batch [599]	Speed: 10147.536609 samples/sec	accuracy=0.534600	lr=0.918487
[Epoch 215] training: accuracy=0.534333
[Epoch 215] speed: 11024 samples/sec	time cost: 120.729587
[Epoch 215] validation: err-top1=0.487244 err-top5=0.240824
Epoch[216] Batch [49]	Speed: 10919.240242 samples/sec	accuracy=0.542383	lr=0.917150
Epoch[216] Batch [99]	Speed: 11089.193804 samples/sec	accuracy=0.539653	lr=0.916271
Epoch[216] Batch [149]	Speed: 11233.902528 samples/sec	accuracy=0.539079	lr=0.915391
Epoch[216] Batch [199]	Speed: 10124.990234 samples/sec	accuracy=0.538623	lr=0.914512
Epoch[216] Batch [249]	Speed: 10898.507444 samples/sec	accuracy=0.538041	lr=0.913633
Epoch[216] Batch [299]	Speed: 11178.585592 samples/sec	accuracy=0.537526	lr=0.912755
Epoch[216] Batch [349]	Speed: 11457.371999 samples/sec	accuracy=0.537105	lr=0.911876
Epoch[216] Batch [399]	Speed: 11164.213855 samples/sec	accuracy=0.536687	lr=0.910998
Epoch[216] Batch [449]	Speed: 11068.709988 samples/sec	accuracy=0.536099	lr=0.910120
Epoch[216] Batch [499]	Speed: 11473.630940 samples/sec	accuracy=0.535874	lr=0.909242
Epoch[216] Batch [549]	Speed: 11517.256887 samples/sec	accuracy=0.535500	lr=0.908364
Epoch[216] Batch [599]	Speed: 10906.188231 samples/sec	accuracy=0.535216	lr=0.907487
[Epoch 216] training: accuracy=0.535191
[Epoch 216] speed: 11060 samples/sec	time cost: 120.143697
[Epoch 216] validation: err-top1=0.452047 err-top5=0.208069
Epoch[217] Batch [49]	Speed: 10931.797070 samples/sec	accuracy=0.547324	lr=0.906171
Epoch[217] Batch [99]	Speed: 11624.445689 samples/sec	accuracy=0.542222	lr=0.905294
Epoch[217] Batch [149]	Speed: 10967.040374 samples/sec	accuracy=0.541572	lr=0.904417
Epoch[217] Batch [199]	Speed: 10868.755466 samples/sec	accuracy=0.541030	lr=0.903540
Epoch[217] Batch [249]	Speed: 11501.323803 samples/sec	accuracy=0.540361	lr=0.902664
Epoch[217] Batch [299]	Speed: 11408.987769 samples/sec	accuracy=0.539982	lr=0.901788
Epoch[217] Batch [349]	Speed: 11418.944223 samples/sec	accuracy=0.539283	lr=0.900912
Epoch[217] Batch [399]	Speed: 11247.630146 samples/sec	accuracy=0.538680	lr=0.900036
Epoch[217] Batch [449]	Speed: 11579.184732 samples/sec	accuracy=0.537958	lr=0.899160
Epoch[217] Batch [499]	Speed: 11492.506363 samples/sec	accuracy=0.537428	lr=0.898285
Epoch[217] Batch [549]	Speed: 11207.197920 samples/sec	accuracy=0.537156	lr=0.897410
Epoch[217] Batch [599]	Speed: 10932.120119 samples/sec	accuracy=0.536829	lr=0.896535
[Epoch 217] training: accuracy=0.536634
[Epoch 217] speed: 11238 samples/sec	time cost: 118.703034
[Epoch 217] validation: err-top1=0.450254 err-top5=0.205352
Epoch[218] Batch [49]	Speed: 11253.131829 samples/sec	accuracy=0.545068	lr=0.895205
Epoch[218] Batch [99]	Speed: 11407.360850 samples/sec	accuracy=0.544375	lr=0.894331
Epoch[218] Batch [149]	Speed: 11070.649495 samples/sec	accuracy=0.542523	lr=0.893456
Epoch[218] Batch [199]	Speed: 10443.467018 samples/sec	accuracy=0.541851	lr=0.892582
Epoch[218] Batch [249]	Speed: 11009.380771 samples/sec	accuracy=0.540922	lr=0.891708
Epoch[218] Batch [299]	Speed: 11067.804090 samples/sec	accuracy=0.540174	lr=0.890835
Epoch[218] Batch [349]	Speed: 11456.709410 samples/sec	accuracy=0.539249	lr=0.889961
Epoch[218] Batch [399]	Speed: 11257.162266 samples/sec	accuracy=0.538810	lr=0.889088
Epoch[218] Batch [449]	Speed: 10695.396029 samples/sec	accuracy=0.538360	lr=0.888215
Epoch[218] Batch [499]	Speed: 11200.021988 samples/sec	accuracy=0.537987	lr=0.887342
Epoch[218] Batch [549]	Speed: 10961.905178 samples/sec	accuracy=0.537341	lr=0.886469
Epoch[218] Batch [599]	Speed: 11084.196757 samples/sec	accuracy=0.536868	lr=0.885597
[Epoch 218] training: accuracy=0.536637
[Epoch 218] speed: 11030 samples/sec	time cost: 120.461425
[Epoch 218] validation: err-top1=0.460164 err-top5=0.214905
Epoch[219] Batch [49]	Speed: 10808.595816 samples/sec	accuracy=0.545244	lr=0.884289
Epoch[219] Batch [99]	Speed: 11034.475705 samples/sec	accuracy=0.543843	lr=0.883417
Epoch[219] Batch [149]	Speed: 11483.460178 samples/sec	accuracy=0.543369	lr=0.882545
Epoch[219] Batch [199]	Speed: 11472.065511 samples/sec	accuracy=0.542437	lr=0.881673
Epoch[219] Batch [249]	Speed: 11328.805687 samples/sec	accuracy=0.541473	lr=0.880802
Epoch[219] Batch [299]	Speed: 11251.095441 samples/sec	accuracy=0.540824	lr=0.879931
Epoch[219] Batch [349]	Speed: 11075.936518 samples/sec	accuracy=0.540467	lr=0.879060
Epoch[219] Batch [399]	Speed: 11060.011399 samples/sec	accuracy=0.539830	lr=0.878189
Epoch[219] Batch [449]	Speed: 10348.242568 samples/sec	accuracy=0.539096	lr=0.877319
Epoch[219] Batch [499]	Speed: 10843.360310 samples/sec	accuracy=0.538705	lr=0.876449
Epoch[219] Batch [549]	Speed: 10614.700887 samples/sec	accuracy=0.538260	lr=0.875579
Epoch[219] Batch [599]	Speed: 11408.415916 samples/sec	accuracy=0.538160	lr=0.874709
[Epoch 219] training: accuracy=0.538051
[Epoch 219] speed: 11060 samples/sec	time cost: 120.381855
[Epoch 219] validation: err-top1=0.453008 err-top5=0.209141
Epoch[220] Batch [49]	Speed: 10429.544349 samples/sec	accuracy=0.545947	lr=0.873387
Epoch[220] Batch [99]	Speed: 11386.108767 samples/sec	accuracy=0.545137	lr=0.872518
Epoch[220] Batch [149]	Speed: 11361.511762 samples/sec	accuracy=0.543848	lr=0.871649
Epoch[220] Batch [199]	Speed: 10897.466055 samples/sec	accuracy=0.542617	lr=0.870780
Epoch[220] Batch [249]	Speed: 11451.572972 samples/sec	accuracy=0.541518	lr=0.869911
Epoch[220] Batch [299]	Speed: 11302.983358 samples/sec	accuracy=0.540815	lr=0.869043
Epoch[220] Batch [349]	Speed: 11198.052948 samples/sec	accuracy=0.540352	lr=0.868175
Epoch[220] Batch [399]	Speed: 11470.159561 samples/sec	accuracy=0.539736	lr=0.867307
Epoch[220] Batch [449]	Speed: 10570.117590 samples/sec	accuracy=0.539266	lr=0.866439
Epoch[220] Batch [499]	Speed: 11213.422708 samples/sec	accuracy=0.538808	lr=0.865571
Epoch[220] Batch [549]	Speed: 10821.256193 samples/sec	accuracy=0.538495	lr=0.864704
Epoch[220] Batch [599]	Speed: 11203.634815 samples/sec	accuracy=0.538153	lr=0.863837
[Epoch 220] training: accuracy=0.538025
[Epoch 220] speed: 11081 samples/sec	time cost: 119.897069
[Epoch 220] validation: err-top1=0.508382 err-top5=0.262227
Epoch[221] Batch [49]	Speed: 10088.432410 samples/sec	accuracy=0.547295	lr=0.862537
Epoch[221] Batch [99]	Speed: 10981.594544 samples/sec	accuracy=0.545898	lr=0.861670
Epoch[221] Batch [149]	Speed: 10744.772231 samples/sec	accuracy=0.544499	lr=0.860804
Epoch[221] Batch [199]	Speed: 10883.499013 samples/sec	accuracy=0.543408	lr=0.859938
Epoch[221] Batch [249]	Speed: 11184.221165 samples/sec	accuracy=0.542775	lr=0.859072
Epoch[221] Batch [299]	Speed: 11377.382043 samples/sec	accuracy=0.542339	lr=0.858206
Epoch[221] Batch [349]	Speed: 11393.164951 samples/sec	accuracy=0.541438	lr=0.857340
Epoch[221] Batch [399]	Speed: 11712.409257 samples/sec	accuracy=0.540704	lr=0.856475
Epoch[221] Batch [449]	Speed: 11272.797789 samples/sec	accuracy=0.540475	lr=0.855610
Epoch[221] Batch [499]	Speed: 11115.250330 samples/sec	accuracy=0.539747	lr=0.854745
Epoch[221] Batch [549]	Speed: 11310.865252 samples/sec	accuracy=0.539482	lr=0.853881
Epoch[221] Batch [599]	Speed: 11153.115777 samples/sec	accuracy=0.539162	lr=0.853016
[Epoch 221] training: accuracy=0.539175
[Epoch 221] speed: 11058 samples/sec	time cost: 120.359118
[Epoch 221] validation: err-top1=0.455200 err-top5=0.212158
Epoch[222] Batch [49]	Speed: 11153.713301 samples/sec	accuracy=0.546309	lr=0.851703
Epoch[222] Batch [99]	Speed: 11283.839879 samples/sec	accuracy=0.546885	lr=0.850839
Epoch[222] Batch [149]	Speed: 11582.524391 samples/sec	accuracy=0.545589	lr=0.849976
Epoch[222] Batch [199]	Speed: 11107.884517 samples/sec	accuracy=0.544346	lr=0.849112
Epoch[222] Batch [249]	Speed: 11458.614560 samples/sec	accuracy=0.543344	lr=0.848249
Epoch[222] Batch [299]	Speed: 11441.023753 samples/sec	accuracy=0.542744	lr=0.847386
Epoch[222] Batch [349]	Speed: 10863.088422 samples/sec	accuracy=0.542165	lr=0.846524
Epoch[222] Batch [399]	Speed: 11257.362610 samples/sec	accuracy=0.541669	lr=0.845661
Epoch[222] Batch [449]	Speed: 11142.040609 samples/sec	accuracy=0.541066	lr=0.844799
Epoch[222] Batch [499]	Speed: 11432.775966 samples/sec	accuracy=0.540551	lr=0.843937
Epoch[222] Batch [549]	Speed: 11699.388783 samples/sec	accuracy=0.540226	lr=0.843075
Epoch[222] Batch [599]	Speed: 10740.517685 samples/sec	accuracy=0.539946	lr=0.842214
[Epoch 222] training: accuracy=0.539789
[Epoch 222] speed: 11212 samples/sec	time cost: 119.071664
[Epoch 222] validation: err-top1=0.440566 err-top5=0.202734
Epoch[223] Batch [49]	Speed: 11606.862483 samples/sec	accuracy=0.549482	lr=0.840905
Epoch[223] Batch [99]	Speed: 11103.571272 samples/sec	accuracy=0.547026	lr=0.840044
Epoch[223] Batch [149]	Speed: 11307.245759 samples/sec	accuracy=0.545824	lr=0.839183
Epoch[223] Batch [199]	Speed: 11139.369292 samples/sec	accuracy=0.545122	lr=0.838323
Epoch[223] Batch [249]	Speed: 11191.489130 samples/sec	accuracy=0.544328	lr=0.837462
Epoch[223] Batch [299]	Speed: 10889.610897 samples/sec	accuracy=0.543156	lr=0.836602
Epoch[223] Batch [349]	Speed: 11575.060516 samples/sec	accuracy=0.542306	lr=0.835743
Epoch[223] Batch [399]	Speed: 11061.093773 samples/sec	accuracy=0.542012	lr=0.834883
Epoch[223] Batch [449]	Speed: 11267.832585 samples/sec	accuracy=0.541502	lr=0.834024
Epoch[223] Batch [499]	Speed: 11351.581060 samples/sec	accuracy=0.541012	lr=0.833165
Epoch[223] Batch [549]	Speed: 11313.444531 samples/sec	accuracy=0.541009	lr=0.832306
Epoch[223] Batch [599]	Speed: 11380.091556 samples/sec	accuracy=0.540741	lr=0.831447
[Epoch 223] training: accuracy=0.540777
[Epoch 223] speed: 11250 samples/sec	time cost: 118.251487
[Epoch 223] validation: err-top1=0.430237 err-top5=0.191264
Epoch[224] Batch [49]	Speed: 10397.851950 samples/sec	accuracy=0.548154	lr=0.830160
Epoch[224] Batch [99]	Speed: 11089.198672 samples/sec	accuracy=0.546890	lr=0.829302
Epoch[224] Batch [149]	Speed: 11343.301700 samples/sec	accuracy=0.546380	lr=0.828444
Epoch[224] Batch [199]	Speed: 11077.965135 samples/sec	accuracy=0.545828	lr=0.827586
Epoch[224] Batch [249]	Speed: 11278.441113 samples/sec	accuracy=0.545342	lr=0.826729
Epoch[224] Batch [299]	Speed: 11102.285700 samples/sec	accuracy=0.544790	lr=0.825872
Epoch[224] Batch [349]	Speed: 11245.296893 samples/sec	accuracy=0.544318	lr=0.825015
Epoch[224] Batch [399]	Speed: 10717.712119 samples/sec	accuracy=0.544125	lr=0.824158
Epoch[224] Batch [449]	Speed: 11246.730950 samples/sec	accuracy=0.543351	lr=0.823302
Epoch[224] Batch [499]	Speed: 11385.737806 samples/sec	accuracy=0.542812	lr=0.822446
Epoch[224] Batch [549]	Speed: 11481.840191 samples/sec	accuracy=0.542611	lr=0.821590
Epoch[224] Batch [599]	Speed: 11361.412282 samples/sec	accuracy=0.542157	lr=0.820734
[Epoch 224] training: accuracy=0.542066
[Epoch 224] speed: 11135 samples/sec	time cost: 119.855959
[Epoch 224] validation: err-top1=0.431582 err-top5=0.197090
Epoch[225] Batch [49]	Speed: 10275.497630 samples/sec	accuracy=0.551436	lr=0.819434
Epoch[225] Batch [99]	Speed: 11336.901698 samples/sec	accuracy=0.550229	lr=0.818579
Epoch[225] Batch [149]	Speed: 11310.560238 samples/sec	accuracy=0.549095	lr=0.817724
Epoch[225] Batch [199]	Speed: 11339.440474 samples/sec	accuracy=0.547778	lr=0.816870
Epoch[225] Batch [249]	Speed: 11406.768560 samples/sec	accuracy=0.546961	lr=0.816016
Epoch[225] Batch [299]	Speed: 11173.867260 samples/sec	accuracy=0.546245	lr=0.815161
Epoch[225] Batch [349]	Speed: 11588.126976 samples/sec	accuracy=0.545840	lr=0.814308
Epoch[225] Batch [399]	Speed: 11370.581390 samples/sec	accuracy=0.545020	lr=0.813454
Epoch[225] Batch [449]	Speed: 11070.776765 samples/sec	accuracy=0.544580	lr=0.812601
Epoch[225] Batch [499]	Speed: 10694.937148 samples/sec	accuracy=0.544186	lr=0.811748
Epoch[225] Batch [549]	Speed: 11093.278409 samples/sec	accuracy=0.543509	lr=0.810895
Epoch[225] Batch [599]	Speed: 11157.087918 samples/sec	accuracy=0.542970	lr=0.810042
[Epoch 225] training: accuracy=0.542933
[Epoch 225] speed: 11109 samples/sec	time cost: 119.556662
[Epoch 225] validation: err-top1=0.449890 err-top5=0.209859
Epoch[226] Batch [49]	Speed: 10456.791165 samples/sec	accuracy=0.550352	lr=0.808764
Epoch[226] Batch [99]	Speed: 10515.738326 samples/sec	accuracy=0.548242	lr=0.807912
Epoch[226] Batch [149]	Speed: 11390.291828 samples/sec	accuracy=0.547210	lr=0.807060
Epoch[226] Batch [199]	Speed: 11553.115567 samples/sec	accuracy=0.546343	lr=0.806208
Epoch[226] Batch [249]	Speed: 11090.343181 samples/sec	accuracy=0.545754	lr=0.805357
Epoch[226] Batch [299]	Speed: 10823.438328 samples/sec	accuracy=0.545243	lr=0.804506
Epoch[226] Batch [349]	Speed: 11072.514893 samples/sec	accuracy=0.544344	lr=0.803655
Epoch[226] Batch [399]	Speed: 10905.489002 samples/sec	accuracy=0.543961	lr=0.802805
Epoch[226] Batch [449]	Speed: 10952.917718 samples/sec	accuracy=0.543734	lr=0.801955
Epoch[226] Batch [499]	Speed: 10692.646259 samples/sec	accuracy=0.543724	lr=0.801105
Epoch[226] Batch [549]	Speed: 10958.996827 samples/sec	accuracy=0.543380	lr=0.800255
Epoch[226] Batch [599]	Speed: 11302.913455 samples/sec	accuracy=0.543026	lr=0.799405
[Epoch 226] training: accuracy=0.542889
[Epoch 226] speed: 10960 samples/sec	time cost: 121.319717
[Epoch 226] validation: err-top1=0.449117 err-top5=0.208394
Epoch[227] Batch [49]	Speed: 11084.580941 samples/sec	accuracy=0.554414	lr=0.798115
Epoch[227] Batch [99]	Speed: 11163.587060 samples/sec	accuracy=0.551089	lr=0.797266
Epoch[227] Batch [149]	Speed: 10968.537948 samples/sec	accuracy=0.549665	lr=0.796417
Epoch[227] Batch [199]	Speed: 11128.938469 samples/sec	accuracy=0.548289	lr=0.795569
Epoch[227] Batch [249]	Speed: 11345.590087 samples/sec	accuracy=0.547598	lr=0.794721
Epoch[227] Batch [299]	Speed: 11178.052021 samples/sec	accuracy=0.546852	lr=0.793873
Epoch[227] Batch [349]	Speed: 11032.174216 samples/sec	accuracy=0.546281	lr=0.793025
Epoch[227] Batch [399]	Speed: 11118.919487 samples/sec	accuracy=0.545555	lr=0.792178
Epoch[227] Batch [449]	Speed: 11330.886747 samples/sec	accuracy=0.545311	lr=0.791331
Epoch[227] Batch [499]	Speed: 11555.128461 samples/sec	accuracy=0.544963	lr=0.790484
Epoch[227] Batch [549]	Speed: 11273.444602 samples/sec	accuracy=0.544695	lr=0.789637
Epoch[227] Batch [599]	Speed: 11586.575473 samples/sec	accuracy=0.544360	lr=0.788791
[Epoch 227] training: accuracy=0.544372
[Epoch 227] speed: 11214 samples/sec	time cost: 118.701957
[Epoch 227] validation: err-top1=0.447754 err-top5=0.208262
Epoch[228] Batch [49]	Speed: 11300.009853 samples/sec	accuracy=0.553311	lr=0.787522
Epoch[228] Batch [99]	Speed: 11398.984827 samples/sec	accuracy=0.551606	lr=0.786676
Epoch[228] Batch [149]	Speed: 10674.708332 samples/sec	accuracy=0.551064	lr=0.785831
Epoch[228] Batch [199]	Speed: 11435.557907 samples/sec	accuracy=0.550237	lr=0.784986
Epoch[228] Batch [249]	Speed: 11405.902200 samples/sec	accuracy=0.549248	lr=0.784141
Epoch[228] Batch [299]	Speed: 11232.268757 samples/sec	accuracy=0.548226	lr=0.783296
Epoch[228] Batch [349]	Speed: 10068.965545 samples/sec	accuracy=0.547432	lr=0.782452
Epoch[228] Batch [399]	Speed: 10566.129915 samples/sec	accuracy=0.546783	lr=0.781607
Epoch[228] Batch [449]	Speed: 11424.540973 samples/sec	accuracy=0.546329	lr=0.780763
Epoch[228] Batch [499]	Speed: 11240.396476 samples/sec	accuracy=0.545843	lr=0.779920
Epoch[228] Batch [549]	Speed: 11480.811705 samples/sec	accuracy=0.545249	lr=0.779076
Epoch[228] Batch [599]	Speed: 11474.291810 samples/sec	accuracy=0.544753	lr=0.778233
[Epoch 228] training: accuracy=0.544719
[Epoch 228] speed: 11089 samples/sec	time cost: 120.047459
[Epoch 228] validation: err-top1=0.446411 err-top5=0.209554
Epoch[229] Batch [49]	Speed: 9676.892947 samples/sec	accuracy=0.553379	lr=0.776952
Epoch[229] Batch [99]	Speed: 11210.473894 samples/sec	accuracy=0.550557	lr=0.776110
Epoch[229] Batch [149]	Speed: 10997.352488 samples/sec	accuracy=0.549440	lr=0.775268
Epoch[229] Batch [199]	Speed: 11314.730885 samples/sec	accuracy=0.549111	lr=0.774426
Epoch[229] Batch [249]	Speed: 11526.567696 samples/sec	accuracy=0.548262	lr=0.773584
Epoch[229] Batch [299]	Speed: 11199.727011 samples/sec	accuracy=0.547839	lr=0.772743
Epoch[229] Batch [349]	Speed: 11355.932744 samples/sec	accuracy=0.547390	lr=0.771902
Epoch[229] Batch [399]	Speed: 10707.795926 samples/sec	accuracy=0.547382	lr=0.771061
Epoch[229] Batch [449]	Speed: 11447.473848 samples/sec	accuracy=0.547323	lr=0.770220
Epoch[229] Batch [499]	Speed: 11607.704429 samples/sec	accuracy=0.547022	lr=0.769380
Epoch[229] Batch [549]	Speed: 11804.785716 samples/sec	accuracy=0.546687	lr=0.768540
Epoch[229] Batch [599]	Speed: 10940.935505 samples/sec	accuracy=0.546445	lr=0.767700
[Epoch 229] training: accuracy=0.546129
[Epoch 229] speed: 11080 samples/sec	time cost: 120.387245
[Epoch 229] validation: err-top1=0.443730 err-top5=0.204590
Epoch[230] Batch [49]	Speed: 10832.018591 samples/sec	accuracy=0.553721	lr=0.766424
Epoch[230] Batch [99]	Speed: 11551.155565 samples/sec	accuracy=0.552412	lr=0.765585
Epoch[230] Batch [149]	Speed: 11494.001092 samples/sec	accuracy=0.551455	lr=0.764746
Epoch[230] Batch [199]	Speed: 11548.427647 samples/sec	accuracy=0.550510	lr=0.763907
Epoch[230] Batch [249]	Speed: 11049.179968 samples/sec	accuracy=0.550141	lr=0.763069
Epoch[230] Batch [299]	Speed: 11681.484893 samples/sec	accuracy=0.549357	lr=0.762231
Epoch[230] Batch [349]	Speed: 11115.478449 samples/sec	accuracy=0.548661	lr=0.761393
Epoch[230] Batch [399]	Speed: 11666.489643 samples/sec	accuracy=0.548195	lr=0.760556
Epoch[230] Batch [449]	Speed: 11068.228782 samples/sec	accuracy=0.547697	lr=0.759718
Epoch[230] Batch [499]	Speed: 11288.105059 samples/sec	accuracy=0.547334	lr=0.758881
Epoch[230] Batch [549]	Speed: 10982.957634 samples/sec	accuracy=0.547317	lr=0.758045
Epoch[230] Batch [599]	Speed: 11251.298221 samples/sec	accuracy=0.547001	lr=0.757208
[Epoch 230] training: accuracy=0.546820
[Epoch 230] speed: 11277 samples/sec	time cost: 117.873184
[Epoch 230] validation: err-top1=0.445943 err-top5=0.206136
Epoch[231] Batch [49]	Speed: 10951.480533 samples/sec	accuracy=0.554023	lr=0.755954
Epoch[231] Batch [99]	Speed: 10096.622940 samples/sec	accuracy=0.552983	lr=0.755118
Epoch[231] Batch [149]	Speed: 11427.834276 samples/sec	accuracy=0.552057	lr=0.754283
Epoch[231] Batch [199]	Speed: 11412.479225 samples/sec	accuracy=0.551829	lr=0.753448
Epoch[231] Batch [249]	Speed: 11609.562849 samples/sec	accuracy=0.550887	lr=0.752613
Epoch[231] Batch [299]	Speed: 11306.373021 samples/sec	accuracy=0.550724	lr=0.751778
Epoch[231] Batch [349]	Speed: 10050.369456 samples/sec	accuracy=0.550106	lr=0.750944
Epoch[231] Batch [399]	Speed: 10853.325790 samples/sec	accuracy=0.549545	lr=0.750110
Epoch[231] Batch [449]	Speed: 11100.375544 samples/sec	accuracy=0.548962	lr=0.749276
Epoch[231] Batch [499]	Speed: 11358.273089 samples/sec	accuracy=0.548438	lr=0.748442
Epoch[231] Batch [549]	Speed: 11286.842251 samples/sec	accuracy=0.547969	lr=0.747609
Epoch[231] Batch [599]	Speed: 11734.584828 samples/sec	accuracy=0.547480	lr=0.746776
[Epoch 231] training: accuracy=0.547341
[Epoch 231] speed: 11056 samples/sec	time cost: 120.595624
[Epoch 231] validation: err-top1=0.453945 err-top5=0.214922
Epoch[232] Batch [49]	Speed: 10130.434276 samples/sec	accuracy=0.556592	lr=0.745510
Epoch[232] Batch [99]	Speed: 11388.558805 samples/sec	accuracy=0.554170	lr=0.744678
Epoch[232] Batch [149]	Speed: 10778.151840 samples/sec	accuracy=0.553398	lr=0.743846
Epoch[232] Batch [199]	Speed: 11825.955132 samples/sec	accuracy=0.552429	lr=0.743014
Epoch[232] Batch [249]	Speed: 11531.286178 samples/sec	accuracy=0.552049	lr=0.742183
Epoch[232] Batch [299]	Speed: 11727.542394 samples/sec	accuracy=0.551061	lr=0.741352
Epoch[232] Batch [349]	Speed: 10588.737019 samples/sec	accuracy=0.550611	lr=0.740521
Epoch[232] Batch [399]	Speed: 10988.342536 samples/sec	accuracy=0.550018	lr=0.739690
Epoch[232] Batch [449]	Speed: 11297.713365 samples/sec	accuracy=0.549529	lr=0.738860
Epoch[232] Batch [499]	Speed: 10284.636353 samples/sec	accuracy=0.549085	lr=0.738030
Epoch[232] Batch [549]	Speed: 10954.552254 samples/sec	accuracy=0.548807	lr=0.737200
Epoch[232] Batch [599]	Speed: 10804.386529 samples/sec	accuracy=0.548631	lr=0.736371
[Epoch 232] training: accuracy=0.548464
[Epoch 232] speed: 10972 samples/sec	time cost: 121.109625
[Epoch 232] validation: err-top1=0.448079 err-top5=0.208211
Epoch[233] Batch [49]	Speed: 11297.495238 samples/sec	accuracy=0.556621	lr=0.735127
Epoch[233] Batch [99]	Speed: 11213.340442 samples/sec	accuracy=0.554023	lr=0.734298
Epoch[233] Batch [149]	Speed: 11393.527631 samples/sec	accuracy=0.554583	lr=0.733470
Epoch[233] Batch [199]	Speed: 11555.227632 samples/sec	accuracy=0.553394	lr=0.732641
Epoch[233] Batch [249]	Speed: 11330.582147 samples/sec	accuracy=0.552574	lr=0.731813
Epoch[233] Batch [299]	Speed: 10787.759319 samples/sec	accuracy=0.552207	lr=0.730986
Epoch[233] Batch [349]	Speed: 11375.880428 samples/sec	accuracy=0.551508	lr=0.730158
Epoch[233] Batch [399]	Speed: 11392.357466 samples/sec	accuracy=0.551458	lr=0.729331
Epoch[233] Batch [449]	Speed: 10528.061680 samples/sec	accuracy=0.550893	lr=0.728504
Epoch[233] Batch [499]	Speed: 10736.767592 samples/sec	accuracy=0.550642	lr=0.727678
Epoch[233] Batch [549]	Speed: 11542.738979 samples/sec	accuracy=0.550044	lr=0.726852
Epoch[233] Batch [599]	Speed: 11206.466872 samples/sec	accuracy=0.549551	lr=0.726026
[Epoch 233] training: accuracy=0.549465
[Epoch 233] speed: 11166 samples/sec	time cost: 119.242009
[Epoch 233] validation: err-top1=0.435913 err-top5=0.201640
Epoch[234] Batch [49]	Speed: 11081.869907 samples/sec	accuracy=0.558818	lr=0.724771
Epoch[234] Batch [99]	Speed: 11316.790079 samples/sec	accuracy=0.556426	lr=0.723946
Epoch[234] Batch [149]	Speed: 11640.212772 samples/sec	accuracy=0.555042	lr=0.723121
Epoch[234] Batch [199]	Speed: 11363.019506 samples/sec	accuracy=0.554709	lr=0.722296
Epoch[234] Batch [249]	Speed: 11417.465005 samples/sec	accuracy=0.554621	lr=0.721472
Epoch[234] Batch [299]	Speed: 11544.593411 samples/sec	accuracy=0.553910	lr=0.720648
Epoch[234] Batch [349]	Speed: 10716.219147 samples/sec	accuracy=0.553013	lr=0.719824
Epoch[234] Batch [399]	Speed: 11427.781977 samples/sec	accuracy=0.552468	lr=0.719000
Epoch[234] Batch [449]	Speed: 11391.773077 samples/sec	accuracy=0.552201	lr=0.718177
Epoch[234] Batch [499]	Speed: 11480.193965 samples/sec	accuracy=0.551469	lr=0.717354
Epoch[234] Batch [549]	Speed: 11153.096083 samples/sec	accuracy=0.550889	lr=0.716532
Epoch[234] Batch [599]	Speed: 10610.405060 samples/sec	accuracy=0.550701	lr=0.715710
[Epoch 234] training: accuracy=0.550420
[Epoch 234] speed: 11248 samples/sec	time cost: 118.456961
[Epoch 234] validation: err-top1=0.441758 err-top5=0.202051
Epoch[235] Batch [49]	Speed: 10903.217748 samples/sec	accuracy=0.558311	lr=0.714477
Epoch[235] Batch [99]	Speed: 11296.722650 samples/sec	accuracy=0.556333	lr=0.713655
Epoch[235] Batch [149]	Speed: 11615.784086 samples/sec	accuracy=0.555856	lr=0.712834
Epoch[235] Batch [199]	Speed: 11146.482153 samples/sec	accuracy=0.554844	lr=0.712013
Epoch[235] Batch [249]	Speed: 11392.453862 samples/sec	accuracy=0.554127	lr=0.711192
Epoch[235] Batch [299]	Speed: 11243.107938 samples/sec	accuracy=0.553789	lr=0.710372
Epoch[235] Batch [349]	Speed: 11272.388613 samples/sec	accuracy=0.553393	lr=0.709552
Epoch[235] Batch [399]	Speed: 11100.900291 samples/sec	accuracy=0.552806	lr=0.708732
Epoch[235] Batch [449]	Speed: 10854.476442 samples/sec	accuracy=0.552369	lr=0.707912
Epoch[235] Batch [499]	Speed: 11368.570283 samples/sec	accuracy=0.552097	lr=0.707093
Epoch[235] Batch [549]	Speed: 11047.098799 samples/sec	accuracy=0.551490	lr=0.706274
Epoch[235] Batch [599]	Speed: 11052.677917 samples/sec	accuracy=0.551191	lr=0.705456
[Epoch 235] training: accuracy=0.550926
[Epoch 235] speed: 11171 samples/sec	time cost: 119.183567
[Epoch 235] validation: err-top1=0.441589 err-top5=0.203349
Epoch[236] Batch [49]	Speed: 10762.805276 samples/sec	accuracy=0.558818	lr=0.704212
Epoch[236] Batch [99]	Speed: 11492.903689 samples/sec	accuracy=0.558398	lr=0.703394
Epoch[236] Batch [149]	Speed: 11239.594323 samples/sec	accuracy=0.557497	lr=0.702576
Epoch[236] Batch [199]	Speed: 11056.095829 samples/sec	accuracy=0.556445	lr=0.701759
Epoch[236] Batch [249]	Speed: 11187.950577 samples/sec	accuracy=0.556107	lr=0.700942
Epoch[236] Batch [299]	Speed: 11106.289489 samples/sec	accuracy=0.555062	lr=0.700125
Epoch[236] Batch [349]	Speed: 10900.635361 samples/sec	accuracy=0.554495	lr=0.699309
Epoch[236] Batch [399]	Speed: 11446.356637 samples/sec	accuracy=0.553960	lr=0.698493
Epoch[236] Batch [449]	Speed: 11397.496560 samples/sec	accuracy=0.553448	lr=0.697677
Epoch[236] Batch [499]	Speed: 11523.511580 samples/sec	accuracy=0.552774	lr=0.696862
Epoch[236] Batch [549]	Speed: 11577.120385 samples/sec	accuracy=0.552268	lr=0.696047
Epoch[236] Batch [599]	Speed: 11138.816636 samples/sec	accuracy=0.551908	lr=0.695232
[Epoch 236] training: accuracy=0.551777
[Epoch 236] speed: 11183 samples/sec	time cost: 119.400201
[Epoch 236] validation: err-top1=0.415430 err-top5=0.183828
Epoch[237] Batch [49]	Speed: 10077.067030 samples/sec	accuracy=0.562422	lr=0.693994
Epoch[237] Batch [99]	Speed: 11357.576260 samples/sec	accuracy=0.559590	lr=0.693180
Epoch[237] Batch [149]	Speed: 11361.616654 samples/sec	accuracy=0.558434	lr=0.692366
Epoch[237] Batch [199]	Speed: 10788.394754 samples/sec	accuracy=0.557527	lr=0.691552
Epoch[237] Batch [249]	Speed: 11016.693344 samples/sec	accuracy=0.556254	lr=0.690739
Epoch[237] Batch [299]	Speed: 11207.414036 samples/sec	accuracy=0.555373	lr=0.689926
Epoch[237] Batch [349]	Speed: 11612.019589 samples/sec	accuracy=0.554856	lr=0.689114
Epoch[237] Batch [399]	Speed: 11484.534282 samples/sec	accuracy=0.554462	lr=0.688302
Epoch[237] Batch [449]	Speed: 11410.771579 samples/sec	accuracy=0.554248	lr=0.687490
Epoch[237] Batch [499]	Speed: 11185.454995 samples/sec	accuracy=0.553704	lr=0.686678
Epoch[237] Batch [549]	Speed: 10481.213620 samples/sec	accuracy=0.553179	lr=0.685867
Epoch[237] Batch [599]	Speed: 10750.403383 samples/sec	accuracy=0.553003	lr=0.685056
[Epoch 237] training: accuracy=0.552965
[Epoch 237] speed: 11035 samples/sec	time cost: 120.412868
[Epoch 237] validation: err-top1=0.435242 err-top5=0.199443
Epoch[238] Batch [49]	Speed: 11088.426540 samples/sec	accuracy=0.557822	lr=0.683840
Epoch[238] Batch [99]	Speed: 10785.985375 samples/sec	accuracy=0.557109	lr=0.683029
Epoch[238] Batch [149]	Speed: 11490.083022 samples/sec	accuracy=0.556592	lr=0.682219
Epoch[238] Batch [199]	Speed: 11416.946929 samples/sec	accuracy=0.556743	lr=0.681410
Epoch[238] Batch [249]	Speed: 11174.085291 samples/sec	accuracy=0.555838	lr=0.680601
Epoch[238] Batch [299]	Speed: 11529.379071 samples/sec	accuracy=0.555319	lr=0.679791
Epoch[238] Batch [349]	Speed: 11066.154688 samples/sec	accuracy=0.555187	lr=0.678983
Epoch[238] Batch [399]	Speed: 11529.965591 samples/sec	accuracy=0.554814	lr=0.678174
Epoch[238] Batch [449]	Speed: 11240.946902 samples/sec	accuracy=0.554549	lr=0.677366
Epoch[238] Batch [499]	Speed: 11716.792722 samples/sec	accuracy=0.554075	lr=0.676559
Epoch[238] Batch [549]	Speed: 10800.221528 samples/sec	accuracy=0.553709	lr=0.675751
Epoch[238] Batch [599]	Speed: 11554.086803 samples/sec	accuracy=0.553515	lr=0.674944
[Epoch 238] training: accuracy=0.553374
[Epoch 238] speed: 11249 samples/sec	time cost: 118.431523
[Epoch 238] validation: err-top1=0.460429 err-top5=0.214193
Epoch[239] Batch [49]	Speed: 10611.137218 samples/sec	accuracy=0.561943	lr=0.673718
Epoch[239] Batch [99]	Speed: 11256.391349 samples/sec	accuracy=0.560869	lr=0.672911
Epoch[239] Batch [149]	Speed: 11050.317653 samples/sec	accuracy=0.560068	lr=0.672105
Epoch[239] Batch [199]	Speed: 10436.424960 samples/sec	accuracy=0.558477	lr=0.671300
Epoch[239] Batch [249]	Speed: 11255.529097 samples/sec	accuracy=0.557426	lr=0.670494
Epoch[239] Batch [299]	Speed: 11730.419033 samples/sec	accuracy=0.557266	lr=0.669689
Epoch[239] Batch [349]	Speed: 11333.194350 samples/sec	accuracy=0.556705	lr=0.668884
Epoch[239] Batch [399]	Speed: 11376.536413 samples/sec	accuracy=0.556099	lr=0.668080
Epoch[239] Batch [449]	Speed: 11416.398251 samples/sec	accuracy=0.555587	lr=0.667276
Epoch[239] Batch [499]	Speed: 10740.064860 samples/sec	accuracy=0.555184	lr=0.666472
Epoch[239] Batch [549]	Speed: 11310.026801 samples/sec	accuracy=0.554819	lr=0.665668
Epoch[239] Batch [599]	Speed: 11145.939782 samples/sec	accuracy=0.554383	lr=0.664865
[Epoch 239] training: accuracy=0.554262
[Epoch 239] speed: 11108 samples/sec	time cost: 119.936171
[Epoch 239] validation: err-top1=0.417480 err-top5=0.184648
Epoch[240] Batch [49]	Speed: 10754.508498 samples/sec	accuracy=0.561670	lr=0.663661
Epoch[240] Batch [99]	Speed: 11471.656755 samples/sec	accuracy=0.560640	lr=0.662858
Epoch[240] Batch [149]	Speed: 11459.855862 samples/sec	accuracy=0.559382	lr=0.662056
Epoch[240] Batch [199]	Speed: 11292.651370 samples/sec	accuracy=0.558491	lr=0.661255
Epoch[240] Batch [249]	Speed: 11364.202595 samples/sec	accuracy=0.557932	lr=0.660453
Epoch[240] Batch [299]	Speed: 11134.289408 samples/sec	accuracy=0.557308	lr=0.659652
Epoch[240] Batch [349]	Speed: 11565.757192 samples/sec	accuracy=0.556839	lr=0.658851
Epoch[240] Batch [399]	Speed: 10846.183497 samples/sec	accuracy=0.556731	lr=0.658051
Epoch[240] Batch [449]	Speed: 11199.392626 samples/sec	accuracy=0.556221	lr=0.657251
Epoch[240] Batch [499]	Speed: 11131.431683 samples/sec	accuracy=0.555813	lr=0.656451
Epoch[240] Batch [549]	Speed: 11198.329734 samples/sec	accuracy=0.555395	lr=0.655651
Epoch[240] Batch [599]	Speed: 11070.789606 samples/sec	accuracy=0.555174	lr=0.654852
[Epoch 240] training: accuracy=0.555018
[Epoch 240] speed: 11165 samples/sec	time cost: 119.283027
[Epoch 240] validation: err-top1=0.475566 err-top5=0.232849
Epoch[241] Batch [49]	Speed: 11329.318484 samples/sec	accuracy=0.562490	lr=0.653638
Epoch[241] Batch [99]	Speed: 11641.024225 samples/sec	accuracy=0.561118	lr=0.652840
Epoch[241] Batch [149]	Speed: 11074.997452 samples/sec	accuracy=0.560986	lr=0.652042
Epoch[241] Batch [199]	Speed: 11450.711088 samples/sec	accuracy=0.560500	lr=0.651244
Epoch[241] Batch [249]	Speed: 10962.184402 samples/sec	accuracy=0.559881	lr=0.650446
Epoch[241] Batch [299]	Speed: 11393.697494 samples/sec	accuracy=0.559263	lr=0.649649
Epoch[241] Batch [349]	Speed: 10906.082441 samples/sec	accuracy=0.558698	lr=0.648853
Epoch[241] Batch [399]	Speed: 11292.667404 samples/sec	accuracy=0.558219	lr=0.648056
Epoch[241] Batch [449]	Speed: 11313.545557 samples/sec	accuracy=0.557836	lr=0.647260
Epoch[241] Batch [499]	Speed: 11851.449779 samples/sec	accuracy=0.557482	lr=0.646464
Epoch[241] Batch [549]	Speed: 11258.814798 samples/sec	accuracy=0.556923	lr=0.645669
Epoch[241] Batch [599]	Speed: 11196.658131 samples/sec	accuracy=0.556342	lr=0.644874
[Epoch 241] training: accuracy=0.556191
[Epoch 241] speed: 11271 samples/sec	time cost: 118.267801
[Epoch 241] validation: err-top1=0.425605 err-top5=0.190684
Epoch[242] Batch [49]	Speed: 10923.173357 samples/sec	accuracy=0.566836	lr=0.643682
Epoch[242] Batch [99]	Speed: 11523.987735 samples/sec	accuracy=0.563613	lr=0.642887
Epoch[242] Batch [149]	Speed: 11550.715061 samples/sec	accuracy=0.561455	lr=0.642093
Epoch[242] Batch [199]	Speed: 11383.716205 samples/sec	accuracy=0.559973	lr=0.641300
Epoch[242] Batch [249]	Speed: 11222.885768 samples/sec	accuracy=0.559242	lr=0.640507
Epoch[242] Batch [299]	Speed: 11377.810631 samples/sec	accuracy=0.558540	lr=0.639714
Epoch[242] Batch [349]	Speed: 11300.702014 samples/sec	accuracy=0.558139	lr=0.638921
Epoch[242] Batch [399]	Speed: 11103.525630 samples/sec	accuracy=0.557755	lr=0.638129
Epoch[242] Batch [449]	Speed: 10653.176119 samples/sec	accuracy=0.557517	lr=0.637337
Epoch[242] Batch [499]	Speed: 10402.062002 samples/sec	accuracy=0.557276	lr=0.636545
Epoch[242] Batch [549]	Speed: 11215.972377 samples/sec	accuracy=0.556950	lr=0.635754
Epoch[242] Batch [599]	Speed: 11322.766785 samples/sec	accuracy=0.556635	lr=0.634963
[Epoch 242] training: accuracy=0.556556
[Epoch 242] speed: 11134 samples/sec	time cost: 119.597276
[Epoch 242] validation: err-top1=0.421651 err-top5=0.189962
Epoch[243] Batch [49]	Speed: 11027.618852 samples/sec	accuracy=0.562617	lr=0.633761
Epoch[243] Batch [99]	Speed: 11035.386646 samples/sec	accuracy=0.562412	lr=0.632971
Epoch[243] Batch [149]	Speed: 10914.286712 samples/sec	accuracy=0.561712	lr=0.632181
Epoch[243] Batch [199]	Speed: 10821.027450 samples/sec	accuracy=0.560801	lr=0.631392
Epoch[243] Batch [249]	Speed: 11236.818114 samples/sec	accuracy=0.559977	lr=0.630603
Epoch[243] Batch [299]	Speed: 11272.825305 samples/sec	accuracy=0.558913	lr=0.629814
Epoch[243] Batch [349]	Speed: 11836.137599 samples/sec	accuracy=0.558989	lr=0.629025
Epoch[243] Batch [399]	Speed: 10972.456487 samples/sec	accuracy=0.558790	lr=0.628237
Epoch[243] Batch [449]	Speed: 10582.762020 samples/sec	accuracy=0.558244	lr=0.627449
Epoch[243] Batch [499]	Speed: 11348.382829 samples/sec	accuracy=0.558029	lr=0.626662
Epoch[243] Batch [549]	Speed: 11119.577549 samples/sec	accuracy=0.557832	lr=0.625875
Epoch[243] Batch [599]	Speed: 11495.693127 samples/sec	accuracy=0.557594	lr=0.625088
[Epoch 243] training: accuracy=0.557561
[Epoch 243] speed: 11098 samples/sec	time cost: 120.402831
[Epoch 243] validation: err-top1=0.421855 err-top5=0.188516
Epoch[244] Batch [49]	Speed: 11362.438424 samples/sec	accuracy=0.564023	lr=0.623893
Epoch[244] Batch [99]	Speed: 11479.211794 samples/sec	accuracy=0.563560	lr=0.623107
Epoch[244] Batch [149]	Speed: 11661.947084 samples/sec	accuracy=0.562868	lr=0.622321
Epoch[244] Batch [199]	Speed: 11229.031998 samples/sec	accuracy=0.562427	lr=0.621536
Epoch[244] Batch [249]	Speed: 10807.341197 samples/sec	accuracy=0.561883	lr=0.620751
Epoch[244] Batch [299]	Speed: 11231.789674 samples/sec	accuracy=0.561243	lr=0.619967
Epoch[244] Batch [349]	Speed: 10161.465126 samples/sec	accuracy=0.560441	lr=0.619182
Epoch[244] Batch [399]	Speed: 11312.050913 samples/sec	accuracy=0.559998	lr=0.618398
Epoch[244] Batch [449]	Speed: 10697.573789 samples/sec	accuracy=0.559260	lr=0.617615
Epoch[244] Batch [499]	Speed: 10552.635524 samples/sec	accuracy=0.558523	lr=0.616832
Epoch[244] Batch [549]	Speed: 10278.053487 samples/sec	accuracy=0.558354	lr=0.616049
Epoch[244] Batch [599]	Speed: 11098.404099 samples/sec	accuracy=0.558248	lr=0.615266
[Epoch 244] training: accuracy=0.558001
[Epoch 244] speed: 10941 samples/sec	time cost: 121.348320
[Epoch 244] validation: err-top1=0.411519 err-top5=0.178324
Epoch[245] Batch [49]	Speed: 11053.080115 samples/sec	accuracy=0.565820	lr=0.614093
Epoch[245] Batch [99]	Speed: 10899.702272 samples/sec	accuracy=0.565254	lr=0.613312
Epoch[245] Batch [149]	Speed: 10281.641798 samples/sec	accuracy=0.563626	lr=0.612530
Epoch[245] Batch [199]	Speed: 11458.439087 samples/sec	accuracy=0.562617	lr=0.611749
Epoch[245] Batch [249]	Speed: 11437.233386 samples/sec	accuracy=0.561570	lr=0.610969
Epoch[245] Batch [299]	Speed: 11081.248607 samples/sec	accuracy=0.560789	lr=0.610188
Epoch[245] Batch [349]	Speed: 11852.948077 samples/sec	accuracy=0.560713	lr=0.609408
Epoch[245] Batch [399]	Speed: 11752.568691 samples/sec	accuracy=0.560614	lr=0.608629
Epoch[245] Batch [449]	Speed: 11056.623229 samples/sec	accuracy=0.560443	lr=0.607850
Epoch[245] Batch [499]	Speed: 11329.713274 samples/sec	accuracy=0.560186	lr=0.607071
Epoch[245] Batch [549]	Speed: 11583.803308 samples/sec	accuracy=0.559710	lr=0.606292
Epoch[245] Batch [599]	Speed: 11672.619148 samples/sec	accuracy=0.559384	lr=0.605514
[Epoch 245] training: accuracy=0.559291
[Epoch 245] speed: 11203 samples/sec	time cost: 118.853784
[Epoch 245] validation: err-top1=0.435282 err-top5=0.197815
Epoch[246] Batch [49]	Speed: 10814.271482 samples/sec	accuracy=0.566992	lr=0.604332
Epoch[246] Batch [99]	Speed: 10722.687144 samples/sec	accuracy=0.565913	lr=0.603554
Epoch[246] Batch [149]	Speed: 11028.366113 samples/sec	accuracy=0.564548	lr=0.602778
Epoch[246] Batch [199]	Speed: 11569.256768 samples/sec	accuracy=0.563948	lr=0.602001
Epoch[246] Batch [249]	Speed: 11331.306758 samples/sec	accuracy=0.563166	lr=0.601225
Epoch[246] Batch [299]	Speed: 11203.759901 samples/sec	accuracy=0.562799	lr=0.600449
Epoch[246] Batch [349]	Speed: 11521.293648 samples/sec	accuracy=0.562305	lr=0.599673
Epoch[246] Batch [399]	Speed: 11474.974522 samples/sec	accuracy=0.561700	lr=0.598898
Epoch[246] Batch [449]	Speed: 11312.585733 samples/sec	accuracy=0.561166	lr=0.598123
Epoch[246] Batch [499]	Speed: 10885.791855 samples/sec	accuracy=0.560992	lr=0.597349
Epoch[246] Batch [549]	Speed: 10783.113029 samples/sec	accuracy=0.560696	lr=0.596574
Epoch[246] Batch [599]	Speed: 11118.747931 samples/sec	accuracy=0.560306	lr=0.595801
[Epoch 246] training: accuracy=0.560080
[Epoch 246] speed: 11139 samples/sec	time cost: 119.491308
[Epoch 246] validation: err-top1=0.417578 err-top5=0.186758
Epoch[247] Batch [49]	Speed: 10627.398364 samples/sec	accuracy=0.570479	lr=0.594641
Epoch[247] Batch [99]	Speed: 11781.747638 samples/sec	accuracy=0.567944	lr=0.593868
Epoch[247] Batch [149]	Speed: 11391.104156 samples/sec	accuracy=0.566787	lr=0.593095
Epoch[247] Batch [199]	Speed: 11676.068795 samples/sec	accuracy=0.565449	lr=0.592323
Epoch[247] Batch [249]	Speed: 11314.791992 samples/sec	accuracy=0.564617	lr=0.591551
Epoch[247] Batch [299]	Speed: 11201.820807 samples/sec	accuracy=0.563993	lr=0.590779
Epoch[247] Batch [349]	Speed: 11289.026016 samples/sec	accuracy=0.563001	lr=0.590008
Epoch[247] Batch [399]	Speed: 11605.432648 samples/sec	accuracy=0.562201	lr=0.589237
Epoch[247] Batch [449]	Speed: 9716.097311 samples/sec	accuracy=0.561583	lr=0.588467
Epoch[247] Batch [499]	Speed: 10359.992324 samples/sec	accuracy=0.561413	lr=0.587697
Epoch[247] Batch [549]	Speed: 11320.245614 samples/sec	accuracy=0.561336	lr=0.586927
Epoch[247] Batch [599]	Speed: 11281.945277 samples/sec	accuracy=0.561125	lr=0.586158
[Epoch 247] training: accuracy=0.561113
[Epoch 247] speed: 11072 samples/sec	time cost: 120.118044
[Epoch 247] validation: err-top1=0.431030 err-top5=0.194478
Epoch[248] Batch [49]	Speed: 10796.018760 samples/sec	accuracy=0.568398	lr=0.584989
Epoch[248] Batch [99]	Speed: 10753.874895 samples/sec	accuracy=0.568447	lr=0.584221
Epoch[248] Batch [149]	Speed: 11291.459680 samples/sec	accuracy=0.568115	lr=0.583453
Epoch[248] Batch [199]	Speed: 11451.318331 samples/sec	accuracy=0.566624	lr=0.582685
Epoch[248] Batch [249]	Speed: 11379.287430 samples/sec	accuracy=0.565617	lr=0.581917
Epoch[248] Batch [299]	Speed: 11449.948232 samples/sec	accuracy=0.564730	lr=0.581150
Epoch[248] Batch [349]	Speed: 11607.338650 samples/sec	accuracy=0.564294	lr=0.580384
Epoch[248] Batch [399]	Speed: 11425.351508 samples/sec	accuracy=0.564163	lr=0.579617
Epoch[248] Batch [449]	Speed: 10497.890240 samples/sec	accuracy=0.563511	lr=0.578851
Epoch[248] Batch [499]	Speed: 11086.732351 samples/sec	accuracy=0.562747	lr=0.578086
Epoch[248] Batch [549]	Speed: 11268.617488 samples/sec	accuracy=0.562404	lr=0.577321
Epoch[248] Batch [599]	Speed: 11387.094996 samples/sec	accuracy=0.562187	lr=0.576556
[Epoch 248] training: accuracy=0.562220
[Epoch 248] speed: 11184 samples/sec	time cost: 118.975291
[Epoch 248] validation: err-top1=0.411172 err-top5=0.178613
Epoch[249] Batch [49]	Speed: 11281.469060 samples/sec	accuracy=0.571631	lr=0.575409
Epoch[249] Batch [99]	Speed: 10961.400483 samples/sec	accuracy=0.569302	lr=0.574645
Epoch[249] Batch [149]	Speed: 11217.754055 samples/sec	accuracy=0.568353	lr=0.573882
Epoch[249] Batch [199]	Speed: 11555.964472 samples/sec	accuracy=0.567617	lr=0.573118
Epoch[249] Batch [249]	Speed: 11309.458571 samples/sec	accuracy=0.567004	lr=0.572356
Epoch[249] Batch [299]	Speed: 11579.555294 samples/sec	accuracy=0.565884	lr=0.571593
Epoch[249] Batch [349]	Speed: 10500.199312 samples/sec	accuracy=0.565300	lr=0.570831
Epoch[249] Batch [399]	Speed: 10497.981588 samples/sec	accuracy=0.564718	lr=0.570069
Epoch[249] Batch [449]	Speed: 11101.493668 samples/sec	accuracy=0.564456	lr=0.569308
Epoch[249] Batch [499]	Speed: 10429.366055 samples/sec	accuracy=0.563832	lr=0.568547
Epoch[249] Batch [549]	Speed: 10633.822478 samples/sec	accuracy=0.563414	lr=0.567786
Epoch[249] Batch [599]	Speed: 11019.086758 samples/sec	accuracy=0.563083	lr=0.567026
[Epoch 249] training: accuracy=0.562910
[Epoch 249] speed: 10999 samples/sec	time cost: 120.853008
[Epoch 249] validation: err-top1=0.435506 err-top5=0.201762
Epoch[250] Batch [49]	Speed: 11257.638795 samples/sec	accuracy=0.572080	lr=0.565871
Epoch[250] Batch [99]	Speed: 11351.066246 samples/sec	accuracy=0.569468	lr=0.565111
Epoch[250] Batch [149]	Speed: 11278.491758 samples/sec	accuracy=0.568962	lr=0.564352
Epoch[250] Batch [199]	Speed: 10545.332633 samples/sec	accuracy=0.568320	lr=0.563594
Epoch[250] Batch [249]	Speed: 11643.328585 samples/sec	accuracy=0.567104	lr=0.562836
Epoch[250] Batch [299]	Speed: 10947.430462 samples/sec	accuracy=0.566733	lr=0.562078
Epoch[250] Batch [349]	Speed: 11474.717308 samples/sec	accuracy=0.566020	lr=0.561320
Epoch[250] Batch [399]	Speed: 11384.721333 samples/sec	accuracy=0.565494	lr=0.560563
Epoch[250] Batch [449]	Speed: 11200.099386 samples/sec	accuracy=0.565007	lr=0.559806
Epoch[250] Batch [499]	Speed: 11518.921484 samples/sec	accuracy=0.564877	lr=0.559050
Epoch[250] Batch [549]	Speed: 11035.177114 samples/sec	accuracy=0.564172	lr=0.558294
Epoch[250] Batch [599]	Speed: 10218.966389 samples/sec	accuracy=0.563737	lr=0.557538
[Epoch 250] training: accuracy=0.563714
[Epoch 250] speed: 11077 samples/sec	time cost: 120.136071
[Epoch 250] validation: err-top1=0.410360 err-top5=0.179281
Epoch[251] Batch [49]	Speed: 10777.799421 samples/sec	accuracy=0.573848	lr=0.556390
Epoch[251] Batch [99]	Speed: 11328.229295 samples/sec	accuracy=0.570605	lr=0.555635
Epoch[251] Batch [149]	Speed: 11371.331598 samples/sec	accuracy=0.569681	lr=0.554881
Epoch[251] Batch [199]	Speed: 11137.429605 samples/sec	accuracy=0.568845	lr=0.554127
Epoch[251] Batch [249]	Speed: 11388.854149 samples/sec	accuracy=0.568059	lr=0.553373
Epoch[251] Batch [299]	Speed: 10260.115855 samples/sec	accuracy=0.567201	lr=0.552620
Epoch[251] Batch [349]	Speed: 10772.279761 samples/sec	accuracy=0.566755	lr=0.551867
Epoch[251] Batch [399]	Speed: 10973.261894 samples/sec	accuracy=0.566498	lr=0.551115
Epoch[251] Batch [449]	Speed: 11307.363048 samples/sec	accuracy=0.566033	lr=0.550363
Epoch[251] Batch [499]	Speed: 10833.993269 samples/sec	accuracy=0.565725	lr=0.549611
Epoch[251] Batch [549]	Speed: 10641.735206 samples/sec	accuracy=0.565374	lr=0.548860
Epoch[251] Batch [599]	Speed: 11347.356825 samples/sec	accuracy=0.565279	lr=0.548109
[Epoch 251] training: accuracy=0.565138
[Epoch 251] speed: 10967 samples/sec	time cost: 121.317205
[Epoch 251] validation: err-top1=0.420508 err-top5=0.185352
Epoch[252] Batch [49]	Speed: 10959.307783 samples/sec	accuracy=0.571396	lr=0.546983
Epoch[252] Batch [99]	Speed: 10925.882057 samples/sec	accuracy=0.570713	lr=0.546233
Epoch[252] Batch [149]	Speed: 11111.386106 samples/sec	accuracy=0.570456	lr=0.545483
Epoch[252] Batch [199]	Speed: 11174.692623 samples/sec	accuracy=0.569639	lr=0.544734
Epoch[252] Batch [249]	Speed: 11524.206965 samples/sec	accuracy=0.569184	lr=0.543985
Epoch[252] Batch [299]	Speed: 11245.129070 samples/sec	accuracy=0.568867	lr=0.543236
Epoch[252] Batch [349]	Speed: 11356.066358 samples/sec	accuracy=0.568191	lr=0.542488
Epoch[252] Batch [399]	Speed: 11554.089290 samples/sec	accuracy=0.567365	lr=0.541740
Epoch[252] Batch [449]	Speed: 11165.607566 samples/sec	accuracy=0.566669	lr=0.540993
Epoch[252] Batch [499]	Speed: 11254.400671 samples/sec	accuracy=0.566479	lr=0.540246
Epoch[252] Batch [549]	Speed: 11472.782589 samples/sec	accuracy=0.566100	lr=0.539499
Epoch[252] Batch [599]	Speed: 11219.499372 samples/sec	accuracy=0.565715	lr=0.538753
[Epoch 252] training: accuracy=0.565559
[Epoch 252] speed: 11227 samples/sec	time cost: 118.607109
[Epoch 252] validation: err-top1=0.411316 err-top5=0.180868
Epoch[253] Batch [49]	Speed: 10851.369557 samples/sec	accuracy=0.572334	lr=0.537620
Epoch[253] Batch [99]	Speed: 11565.664692 samples/sec	accuracy=0.571689	lr=0.536874
Epoch[253] Batch [149]	Speed: 10194.109175 samples/sec	accuracy=0.570973	lr=0.536129
Epoch[253] Batch [199]	Speed: 11241.911968 samples/sec	accuracy=0.570063	lr=0.535385
Epoch[253] Batch [249]	Speed: 11865.879389 samples/sec	accuracy=0.569613	lr=0.534641
Epoch[253] Batch [299]	Speed: 11118.046796 samples/sec	accuracy=0.568236	lr=0.533897
Epoch[253] Batch [349]	Speed: 11400.264079 samples/sec	accuracy=0.567412	lr=0.533154
Epoch[253] Batch [399]	Speed: 11437.250442 samples/sec	accuracy=0.567131	lr=0.532411
Epoch[253] Batch [449]	Speed: 11151.984915 samples/sec	accuracy=0.566954	lr=0.531668
Epoch[253] Batch [499]	Speed: 11409.583927 samples/sec	accuracy=0.566828	lr=0.530926
Epoch[253] Batch [549]	Speed: 11225.121129 samples/sec	accuracy=0.566357	lr=0.530184
Epoch[253] Batch [599]	Speed: 11244.826414 samples/sec	accuracy=0.565923	lr=0.529442
[Epoch 253] training: accuracy=0.565748
[Epoch 253] speed: 11190 samples/sec	time cost: 118.951128
[Epoch 253] validation: err-top1=0.407051 err-top5=0.176367
Epoch[254] Batch [49]	Speed: 11289.753927 samples/sec	accuracy=0.574492	lr=0.528331
Epoch[254] Batch [99]	Speed: 10746.349800 samples/sec	accuracy=0.573945	lr=0.527591
Epoch[254] Batch [149]	Speed: 10667.246691 samples/sec	accuracy=0.572516	lr=0.526850
Epoch[254] Batch [199]	Speed: 11325.167239 samples/sec	accuracy=0.572114	lr=0.526111
Epoch[254] Batch [249]	Speed: 11477.823356 samples/sec	accuracy=0.571141	lr=0.525371
Epoch[254] Batch [299]	Speed: 11617.084499 samples/sec	accuracy=0.569824	lr=0.524633
Epoch[254] Batch [349]	Speed: 11237.606052 samples/sec	accuracy=0.569499	lr=0.523894
Epoch[254] Batch [399]	Speed: 11363.961748 samples/sec	accuracy=0.569241	lr=0.523156
Epoch[254] Batch [449]	Speed: 11138.644466 samples/sec	accuracy=0.569025	lr=0.522418
Epoch[254] Batch [499]	Speed: 11705.483051 samples/sec	accuracy=0.568468	lr=0.521681
Epoch[254] Batch [549]	Speed: 10688.017156 samples/sec	accuracy=0.567991	lr=0.520944
Epoch[254] Batch [599]	Speed: 11334.862701 samples/sec	accuracy=0.567725	lr=0.520207
[Epoch 254] training: accuracy=0.567587
[Epoch 254] speed: 11186 samples/sec	time cost: 118.906035
[Epoch 254] validation: err-top1=0.406128 err-top5=0.173971
Epoch[255] Batch [49]	Speed: 10627.407305 samples/sec	accuracy=0.573984	lr=0.519088
Epoch[255] Batch [99]	Speed: 11564.851878 samples/sec	accuracy=0.571719	lr=0.518353
Epoch[255] Batch [149]	Speed: 11483.384034 samples/sec	accuracy=0.571351	lr=0.517617
Epoch[255] Batch [199]	Speed: 11639.351595 samples/sec	accuracy=0.571260	lr=0.516883
Epoch[255] Batch [249]	Speed: 11352.864398 samples/sec	accuracy=0.570619	lr=0.516148
Epoch[255] Batch [299]	Speed: 11660.478320 samples/sec	accuracy=0.570444	lr=0.515414
Epoch[255] Batch [349]	Speed: 11589.817128 samples/sec	accuracy=0.570173	lr=0.514681
Epoch[255] Batch [399]	Speed: 11258.195927 samples/sec	accuracy=0.569481	lr=0.513947
Epoch[255] Batch [449]	Speed: 11142.101309 samples/sec	accuracy=0.569182	lr=0.513214
Epoch[255] Batch [499]	Speed: 11305.931345 samples/sec	accuracy=0.568872	lr=0.512482
Epoch[255] Batch [549]	Speed: 11268.792221 samples/sec	accuracy=0.568574	lr=0.511750
Epoch[255] Batch [599]	Speed: 11245.572486 samples/sec	accuracy=0.568132	lr=0.511018
[Epoch 255] training: accuracy=0.567863
[Epoch 255] speed: 11311 samples/sec	time cost: 117.336583
[Epoch 255] validation: err-top1=0.419027 err-top5=0.189840
Epoch[256] Batch [49]	Speed: 11393.401899 samples/sec	accuracy=0.575371	lr=0.509921
Epoch[256] Batch [99]	Speed: 10209.927686 samples/sec	accuracy=0.574814	lr=0.509191
Epoch[256] Batch [149]	Speed: 10707.451029 samples/sec	accuracy=0.573981	lr=0.508460
Epoch[256] Batch [199]	Speed: 11451.752509 samples/sec	accuracy=0.573110	lr=0.507731
Epoch[256] Batch [249]	Speed: 11816.341500 samples/sec	accuracy=0.572012	lr=0.507001
Epoch[256] Batch [299]	Speed: 11360.044681 samples/sec	accuracy=0.571654	lr=0.506272
Epoch[256] Batch [349]	Speed: 10391.624215 samples/sec	accuracy=0.571228	lr=0.505543
Epoch[256] Batch [399]	Speed: 10895.307037 samples/sec	accuracy=0.570836	lr=0.504815
Epoch[256] Batch [449]	Speed: 11324.671540 samples/sec	accuracy=0.570239	lr=0.504087
Epoch[256] Batch [499]	Speed: 11336.522865 samples/sec	accuracy=0.569846	lr=0.503360
Epoch[256] Batch [549]	Speed: 11635.170862 samples/sec	accuracy=0.569417	lr=0.502632
Epoch[256] Batch [599]	Speed: 11436.239976 samples/sec	accuracy=0.569023	lr=0.501906
[Epoch 256] training: accuracy=0.568856
[Epoch 256] speed: 11125 samples/sec	time cost: 119.964773
[Epoch 256] validation: err-top1=0.407617 err-top5=0.176504
Epoch[257] Batch [49]	Speed: 11226.956192 samples/sec	accuracy=0.575547	lr=0.500802
Epoch[257] Batch [99]	Speed: 11124.074667 samples/sec	accuracy=0.576011	lr=0.500076
Epoch[257] Batch [149]	Speed: 11261.079259 samples/sec	accuracy=0.575374	lr=0.499351
Epoch[257] Batch [199]	Speed: 9911.848485 samples/sec	accuracy=0.574529	lr=0.498626
Epoch[257] Batch [249]	Speed: 10839.144697 samples/sec	accuracy=0.573947	lr=0.497902
Epoch[257] Batch [299]	Speed: 11364.239279 samples/sec	accuracy=0.573239	lr=0.497178
Epoch[257] Batch [349]	Speed: 11490.093474 samples/sec	accuracy=0.572649	lr=0.496454
Epoch[257] Batch [399]	Speed: 11606.283482 samples/sec	accuracy=0.571859	lr=0.495731
Epoch[257] Batch [449]	Speed: 11184.271259 samples/sec	accuracy=0.571540	lr=0.495008
Epoch[257] Batch [499]	Speed: 11215.317791 samples/sec	accuracy=0.571113	lr=0.494285
Epoch[257] Batch [549]	Speed: 11567.437397 samples/sec	accuracy=0.570859	lr=0.493563
Epoch[257] Batch [599]	Speed: 10329.615604 samples/sec	accuracy=0.570667	lr=0.492841
[Epoch 257] training: accuracy=0.570559
[Epoch 257] speed: 11042 samples/sec	time cost: 120.504763
[Epoch 257] validation: err-top1=0.423238 err-top5=0.192424
Epoch[258] Batch [49]	Speed: 11245.453826 samples/sec	accuracy=0.576465	lr=0.491745
Epoch[258] Batch [99]	Speed: 11289.424234 samples/sec	accuracy=0.574736	lr=0.491025
Epoch[258] Batch [149]	Speed: 11113.252607 samples/sec	accuracy=0.573021	lr=0.490304
Epoch[258] Batch [199]	Speed: 11421.101964 samples/sec	accuracy=0.573157	lr=0.489585
Epoch[258] Batch [249]	Speed: 11508.051453 samples/sec	accuracy=0.573189	lr=0.488865
Epoch[258] Batch [299]	Speed: 10886.999075 samples/sec	accuracy=0.572780	lr=0.488146
Epoch[258] Batch [349]	Speed: 10177.359423 samples/sec	accuracy=0.572337	lr=0.487427
Epoch[258] Batch [399]	Speed: 11602.188522 samples/sec	accuracy=0.571885	lr=0.486709
Epoch[258] Batch [449]	Speed: 10980.609364 samples/sec	accuracy=0.571631	lr=0.485991
Epoch[258] Batch [499]	Speed: 11289.216516 samples/sec	accuracy=0.571334	lr=0.485274
Epoch[258] Batch [549]	Speed: 10601.085629 samples/sec	accuracy=0.570911	lr=0.484557
Epoch[258] Batch [599]	Speed: 11209.098214 samples/sec	accuracy=0.570661	lr=0.483841
[Epoch 258] training: accuracy=0.570572
[Epoch 258] speed: 11097 samples/sec	time cost: 119.873285
[Epoch 258] validation: err-top1=0.399199 err-top5=0.172891
Epoch[259] Batch [49]	Speed: 10263.935701 samples/sec	accuracy=0.579717	lr=0.482766
Epoch[259] Batch [99]	Speed: 11169.162495 samples/sec	accuracy=0.577876	lr=0.482051
Epoch[259] Batch [149]	Speed: 11286.156531 samples/sec	accuracy=0.576777	lr=0.481336
Epoch[259] Batch [199]	Speed: 10273.146253 samples/sec	accuracy=0.575791	lr=0.480621
Epoch[259] Batch [249]	Speed: 11365.282473 samples/sec	accuracy=0.574830	lr=0.479907
Epoch[259] Batch [299]	Speed: 10822.985576 samples/sec	accuracy=0.574385	lr=0.479193
Epoch[259] Batch [349]	Speed: 11586.422315 samples/sec	accuracy=0.573503	lr=0.478479
Epoch[259] Batch [399]	Speed: 11647.541401 samples/sec	accuracy=0.573053	lr=0.477766
Epoch[259] Batch [449]	Speed: 11477.503137 samples/sec	accuracy=0.572788	lr=0.477053
Epoch[259] Batch [499]	Speed: 11330.015735 samples/sec	accuracy=0.572542	lr=0.476341
Epoch[259] Batch [549]	Speed: 10240.946377 samples/sec	accuracy=0.572267	lr=0.475629
Epoch[259] Batch [599]	Speed: 9940.248804 samples/sec	accuracy=0.572137	lr=0.474918
[Epoch 259] training: accuracy=0.572082
[Epoch 259] speed: 10854 samples/sec	time cost: 122.439667
[Epoch 259] validation: err-top1=0.406860 err-top5=0.180522
Epoch[260] Batch [49]	Speed: 10314.046734 samples/sec	accuracy=0.579072	lr=0.473837
Epoch[260] Batch [99]	Speed: 11190.100029 samples/sec	accuracy=0.578936	lr=0.473127
Epoch[260] Batch [149]	Speed: 11105.336655 samples/sec	accuracy=0.576836	lr=0.472417
Epoch[260] Batch [199]	Speed: 11553.614684 samples/sec	accuracy=0.576577	lr=0.471707
Epoch[260] Batch [249]	Speed: 10857.559580 samples/sec	accuracy=0.576336	lr=0.470998
Epoch[260] Batch [299]	Speed: 11267.106020 samples/sec	accuracy=0.575752	lr=0.470290
Epoch[260] Batch [349]	Speed: 10833.124562 samples/sec	accuracy=0.575297	lr=0.469581
Epoch[260] Batch [399]	Speed: 11097.405593 samples/sec	accuracy=0.574763	lr=0.468873
Epoch[260] Batch [449]	Speed: 11031.873279 samples/sec	accuracy=0.574160	lr=0.468166
Epoch[260] Batch [499]	Speed: 11102.088255 samples/sec	accuracy=0.573771	lr=0.467459
Epoch[260] Batch [549]	Speed: 10946.258066 samples/sec	accuracy=0.573410	lr=0.466752
Epoch[260] Batch [599]	Speed: 11266.550665 samples/sec	accuracy=0.573125	lr=0.466046
[Epoch 260] training: accuracy=0.573017
[Epoch 260] speed: 11027 samples/sec	time cost: 120.633424
[Epoch 260] validation: err-top1=0.397363 err-top5=0.170156
Epoch[261] Batch [49]	Speed: 10690.816703 samples/sec	accuracy=0.583604	lr=0.464987
Epoch[261] Batch [99]	Speed: 11116.014406 samples/sec	accuracy=0.580571	lr=0.464282
Epoch[261] Batch [149]	Speed: 11299.989934 samples/sec	accuracy=0.579430	lr=0.463577
Epoch[261] Batch [199]	Speed: 10413.354139 samples/sec	accuracy=0.579124	lr=0.462873
Epoch[261] Batch [249]	Speed: 10600.819263 samples/sec	accuracy=0.578119	lr=0.462169
Epoch[261] Batch [299]	Speed: 11360.779376 samples/sec	accuracy=0.577194	lr=0.461466
Epoch[261] Batch [349]	Speed: 11228.103485 samples/sec	accuracy=0.576338	lr=0.460762
Epoch[261] Batch [399]	Speed: 11474.127198 samples/sec	accuracy=0.575460	lr=0.460060
Epoch[261] Batch [449]	Speed: 11381.604537 samples/sec	accuracy=0.574854	lr=0.459358
Epoch[261] Batch [499]	Speed: 11560.213238 samples/sec	accuracy=0.574806	lr=0.458656
Epoch[261] Batch [549]	Speed: 11524.805948 samples/sec	accuracy=0.574522	lr=0.457954
Epoch[261] Batch [599]	Speed: 11646.872426 samples/sec	accuracy=0.574214	lr=0.457253
[Epoch 261] training: accuracy=0.573866
[Epoch 261] speed: 11159 samples/sec	time cost: 119.264878
[Epoch 261] validation: err-top1=0.397624 err-top5=0.168009
Epoch[262] Batch [49]	Speed: 11080.391252 samples/sec	accuracy=0.580020	lr=0.456189
Epoch[262] Batch [99]	Speed: 10961.773684 samples/sec	accuracy=0.578677	lr=0.455489
Epoch[262] Batch [149]	Speed: 10793.678942 samples/sec	accuracy=0.578841	lr=0.454789
Epoch[262] Batch [199]	Speed: 11161.853002 samples/sec	accuracy=0.578274	lr=0.454090
Epoch[262] Batch [249]	Speed: 10589.480288 samples/sec	accuracy=0.577311	lr=0.453391
Epoch[262] Batch [299]	Speed: 11328.575303 samples/sec	accuracy=0.577160	lr=0.452693
Epoch[262] Batch [349]	Speed: 10306.804583 samples/sec	accuracy=0.576602	lr=0.451995
Epoch[262] Batch [399]	Speed: 11140.308326 samples/sec	accuracy=0.576488	lr=0.451298
Epoch[262] Batch [449]	Speed: 11327.352119 samples/sec	accuracy=0.576011	lr=0.450601
Epoch[262] Batch [499]	Speed: 11218.174511 samples/sec	accuracy=0.575424	lr=0.449905
Epoch[262] Batch [549]	Speed: 11579.868121 samples/sec	accuracy=0.575164	lr=0.449209
Epoch[262] Batch [599]	Speed: 11126.731447 samples/sec	accuracy=0.575072	lr=0.448513
[Epoch 262] training: accuracy=0.574625
[Epoch 262] speed: 11033 samples/sec	time cost: 120.482853
[Epoch 262] validation: err-top1=0.425334 err-top5=0.190857
Epoch[263] Batch [49]	Speed: 11140.362939 samples/sec	accuracy=0.580244	lr=0.447470
Epoch[263] Batch [99]	Speed: 11157.369059 samples/sec	accuracy=0.580747	lr=0.446776
Epoch[263] Batch [149]	Speed: 11236.241049 samples/sec	accuracy=0.580368	lr=0.446081
Epoch[263] Batch [199]	Speed: 10988.646725 samples/sec	accuracy=0.580200	lr=0.445388
Epoch[263] Batch [249]	Speed: 11538.219107 samples/sec	accuracy=0.579689	lr=0.444694
Epoch[263] Batch [299]	Speed: 10960.681010 samples/sec	accuracy=0.578875	lr=0.444001
Epoch[263] Batch [349]	Speed: 10275.914339 samples/sec	accuracy=0.578291	lr=0.443309
Epoch[263] Batch [399]	Speed: 11244.858798 samples/sec	accuracy=0.577817	lr=0.442617
Epoch[263] Batch [449]	Speed: 10841.157274 samples/sec	accuracy=0.577484	lr=0.441925
Epoch[263] Batch [499]	Speed: 11126.808987 samples/sec	accuracy=0.577017	lr=0.441234
Epoch[263] Batch [549]	Speed: 11128.905883 samples/sec	accuracy=0.576566	lr=0.440543
Epoch[263] Batch [599]	Speed: 11391.973406 samples/sec	accuracy=0.576310	lr=0.439853
[Epoch 263] training: accuracy=0.575987
[Epoch 263] speed: 11053 samples/sec	time cost: 120.673259
[Epoch 263] validation: err-top1=0.396895 err-top5=0.173906
Epoch[264] Batch [49]	Speed: 10998.460654 samples/sec	accuracy=0.581963	lr=0.438805
Epoch[264] Batch [99]	Speed: 10978.633642 samples/sec	accuracy=0.582085	lr=0.438115
Epoch[264] Batch [149]	Speed: 11592.509881 samples/sec	accuracy=0.580052	lr=0.437427
Epoch[264] Batch [199]	Speed: 11248.465557 samples/sec	accuracy=0.580261	lr=0.436738
Epoch[264] Batch [249]	Speed: 11336.507604 samples/sec	accuracy=0.580004	lr=0.436050
Epoch[264] Batch [299]	Speed: 11325.065707 samples/sec	accuracy=0.579582	lr=0.435363
Epoch[264] Batch [349]	Speed: 11403.026585 samples/sec	accuracy=0.579021	lr=0.434676
Epoch[264] Batch [399]	Speed: 10963.139694 samples/sec	accuracy=0.578680	lr=0.433989
Epoch[264] Batch [449]	Speed: 11401.908044 samples/sec	accuracy=0.578112	lr=0.433303
Epoch[264] Batch [499]	Speed: 10503.004569 samples/sec	accuracy=0.577838	lr=0.432617
Epoch[264] Batch [549]	Speed: 11155.825276 samples/sec	accuracy=0.577387	lr=0.431932
Epoch[264] Batch [599]	Speed: 10600.338374 samples/sec	accuracy=0.576972	lr=0.431247
[Epoch 264] training: accuracy=0.576861
[Epoch 264] speed: 11104 samples/sec	time cost: 119.848677
[Epoch 264] validation: err-top1=0.393860 err-top5=0.168844
Epoch[265] Batch [49]	Speed: 10411.525770 samples/sec	accuracy=0.585820	lr=0.430207
Epoch[265] Batch [99]	Speed: 11081.833880 samples/sec	accuracy=0.582812	lr=0.429523
Epoch[265] Batch [149]	Speed: 10950.969817 samples/sec	accuracy=0.581403	lr=0.428840
Epoch[265] Batch [199]	Speed: 11203.100311 samples/sec	accuracy=0.581252	lr=0.428157
Epoch[265] Batch [249]	Speed: 11363.540516 samples/sec	accuracy=0.580285	lr=0.427474
Epoch[265] Batch [299]	Speed: 11316.310020 samples/sec	accuracy=0.579897	lr=0.426792
Epoch[265] Batch [349]	Speed: 11687.488640 samples/sec	accuracy=0.579597	lr=0.426111
Epoch[265] Batch [399]	Speed: 11310.737764 samples/sec	accuracy=0.579056	lr=0.425429
Epoch[265] Batch [449]	Speed: 10983.784527 samples/sec	accuracy=0.578593	lr=0.424749
Epoch[265] Batch [499]	Speed: 11590.490199 samples/sec	accuracy=0.578253	lr=0.424068
Epoch[265] Batch [549]	Speed: 11270.344954 samples/sec	accuracy=0.577884	lr=0.423389
Epoch[265] Batch [599]	Speed: 11137.180080 samples/sec	accuracy=0.577663	lr=0.422709
[Epoch 265] training: accuracy=0.577788
[Epoch 265] speed: 11156 samples/sec	time cost: 119.347055
[Epoch 265] validation: err-top1=0.402988 err-top5=0.176230
Epoch[266] Batch [49]	Speed: 11476.537373 samples/sec	accuracy=0.584902	lr=0.421691
Epoch[266] Batch [99]	Speed: 11386.503600 samples/sec	accuracy=0.583604	lr=0.421013
Epoch[266] Batch [149]	Speed: 10535.041153 samples/sec	accuracy=0.582894	lr=0.420335
Epoch[266] Batch [199]	Speed: 11248.231948 samples/sec	accuracy=0.582549	lr=0.419657
Epoch[266] Batch [249]	Speed: 11397.609679 samples/sec	accuracy=0.581785	lr=0.418980
Epoch[266] Batch [299]	Speed: 11812.179839 samples/sec	accuracy=0.580731	lr=0.418304
Epoch[266] Batch [349]	Speed: 11301.031178 samples/sec	accuracy=0.580262	lr=0.417628
Epoch[266] Batch [399]	Speed: 11279.365529 samples/sec	accuracy=0.580190	lr=0.416952
Epoch[266] Batch [449]	Speed: 11453.028973 samples/sec	accuracy=0.580007	lr=0.416277
Epoch[266] Batch [499]	Speed: 11106.379957 samples/sec	accuracy=0.579767	lr=0.415602
Epoch[266] Batch [549]	Speed: 10639.884019 samples/sec	accuracy=0.579400	lr=0.414928
Epoch[266] Batch [599]	Speed: 11125.465579 samples/sec	accuracy=0.579163	lr=0.414254
[Epoch 266] training: accuracy=0.578858
[Epoch 266] speed: 11179 samples/sec	time cost: 119.075070
[Epoch 266] validation: err-top1=0.420003 err-top5=0.188314
Epoch[267] Batch [49]	Speed: 10838.056914 samples/sec	accuracy=0.586387	lr=0.413230
Epoch[267] Batch [99]	Speed: 11426.158510 samples/sec	accuracy=0.585034	lr=0.412558
Epoch[267] Batch [149]	Speed: 11461.805799 samples/sec	accuracy=0.583600	lr=0.411885
Epoch[267] Batch [199]	Speed: 11300.009853 samples/sec	accuracy=0.583059	lr=0.411213
Epoch[267] Batch [249]	Speed: 11451.261848 samples/sec	accuracy=0.582059	lr=0.410542
Epoch[267] Batch [299]	Speed: 11493.336720 samples/sec	accuracy=0.581523	lr=0.409871
Epoch[267] Batch [349]	Speed: 10502.464971 samples/sec	accuracy=0.581014	lr=0.409200
Epoch[267] Batch [399]	Speed: 11101.634561 samples/sec	accuracy=0.580692	lr=0.408530
Epoch[267] Batch [449]	Speed: 11862.508683 samples/sec	accuracy=0.580369	lr=0.407861
Epoch[267] Batch [499]	Speed: 11107.626834 samples/sec	accuracy=0.580178	lr=0.407191
Epoch[267] Batch [549]	Speed: 11630.313143 samples/sec	accuracy=0.579697	lr=0.406523
Epoch[267] Batch [599]	Speed: 10293.866841 samples/sec	accuracy=0.579419	lr=0.405854
[Epoch 267] training: accuracy=0.579302
[Epoch 267] speed: 11124 samples/sec	time cost: 119.522131
[Epoch 267] validation: err-top1=0.418681 err-top5=0.190999
Epoch[268] Batch [49]	Speed: 11424.808707 samples/sec	accuracy=0.588008	lr=0.404853
Epoch[268] Batch [99]	Speed: 11147.679893 samples/sec	accuracy=0.586372	lr=0.404186
Epoch[268] Batch [149]	Speed: 11339.517116 samples/sec	accuracy=0.585742	lr=0.403519
Epoch[268] Batch [199]	Speed: 11514.001054 samples/sec	accuracy=0.584304	lr=0.402852
Epoch[268] Batch [249]	Speed: 10367.124759 samples/sec	accuracy=0.584025	lr=0.402187
Epoch[268] Batch [299]	Speed: 10573.189643 samples/sec	accuracy=0.583271	lr=0.401521
Epoch[268] Batch [349]	Speed: 11438.333586 samples/sec	accuracy=0.582843	lr=0.400856
Epoch[268] Batch [399]	Speed: 11612.818638 samples/sec	accuracy=0.582321	lr=0.400192
Epoch[268] Batch [449]	Speed: 11254.455229 samples/sec	accuracy=0.581839	lr=0.399528
Epoch[268] Batch [499]	Speed: 10333.981414 samples/sec	accuracy=0.581438	lr=0.398864
Epoch[268] Batch [549]	Speed: 11242.696795 samples/sec	accuracy=0.581093	lr=0.398201
Epoch[268] Batch [599]	Speed: 11135.365872 samples/sec	accuracy=0.580806	lr=0.397538
[Epoch 268] training: accuracy=0.580665
[Epoch 268] speed: 11082 samples/sec	time cost: 120.305553
[Epoch 268] validation: err-top1=0.388613 err-top5=0.164258
Epoch[269] Batch [49]	Speed: 10536.270562 samples/sec	accuracy=0.587393	lr=0.396532
Epoch[269] Batch [99]	Speed: 11081.560249 samples/sec	accuracy=0.586060	lr=0.395870
Epoch[269] Batch [149]	Speed: 11268.087999 samples/sec	accuracy=0.584368	lr=0.395209
Epoch[269] Batch [199]	Speed: 10948.079265 samples/sec	accuracy=0.584507	lr=0.394549
Epoch[269] Batch [249]	Speed: 10849.009251 samples/sec	accuracy=0.583744	lr=0.393889
Epoch[269] Batch [299]	Speed: 10577.728911 samples/sec	accuracy=0.583464	lr=0.393229
Epoch[269] Batch [349]	Speed: 10759.019392 samples/sec	accuracy=0.582733	lr=0.392570
Epoch[269] Batch [399]	Speed: 11303.199317 samples/sec	accuracy=0.582291	lr=0.391911
Epoch[269] Batch [449]	Speed: 11458.300303 samples/sec	accuracy=0.582211	lr=0.391252
Epoch[269] Batch [499]	Speed: 11577.727690 samples/sec	accuracy=0.581941	lr=0.390594
Epoch[269] Batch [549]	Speed: 11347.520218 samples/sec	accuracy=0.581728	lr=0.389937
Epoch[269] Batch [599]	Speed: 11292.886532 samples/sec	accuracy=0.581381	lr=0.389280
[Epoch 269] training: accuracy=0.581222
[Epoch 269] speed: 11063 samples/sec	time cost: 120.100864
[Epoch 269] validation: err-top1=0.391276 err-top5=0.170369
Epoch[270] Batch [49]	Speed: 11133.129751 samples/sec	accuracy=0.589834	lr=0.388295
Epoch[270] Batch [99]	Speed: 10632.132222 samples/sec	accuracy=0.589102	lr=0.387639
Epoch[270] Batch [149]	Speed: 11352.232745 samples/sec	accuracy=0.587936	lr=0.386984
Epoch[270] Batch [199]	Speed: 11553.527662 samples/sec	accuracy=0.587378	lr=0.386329
Epoch[270] Batch [249]	Speed: 11356.955494 samples/sec	accuracy=0.586160	lr=0.385675
Epoch[270] Batch [299]	Speed: 11312.755873 samples/sec	accuracy=0.585392	lr=0.385021
Epoch[270] Batch [349]	Speed: 10906.832986 samples/sec	accuracy=0.584654	lr=0.384367
Epoch[270] Batch [399]	Speed: 10735.395691 samples/sec	accuracy=0.584268	lr=0.383714
Epoch[270] Batch [449]	Speed: 10910.679559 samples/sec	accuracy=0.583819	lr=0.383061
Epoch[270] Batch [499]	Speed: 11356.141723 samples/sec	accuracy=0.583443	lr=0.382409
Epoch[270] Batch [549]	Speed: 11246.549243 samples/sec	accuracy=0.583136	lr=0.381757
Epoch[270] Batch [599]	Speed: 11052.049363 samples/sec	accuracy=0.582681	lr=0.381106
[Epoch 270] training: accuracy=0.582496
[Epoch 270] speed: 11056 samples/sec	time cost: 120.716274
[Epoch 270] validation: err-top1=0.395605 err-top5=0.169805
Epoch[271] Batch [49]	Speed: 10040.964163 samples/sec	accuracy=0.589082	lr=0.380117
Epoch[271] Batch [99]	Speed: 10397.825267 samples/sec	accuracy=0.587261	lr=0.379467
Epoch[271] Batch [149]	Speed: 11432.894656 samples/sec	accuracy=0.586885	lr=0.378817
Epoch[271] Batch [199]	Speed: 11301.840935 samples/sec	accuracy=0.586272	lr=0.378168
Epoch[271] Batch [249]	Speed: 11274.457578 samples/sec	accuracy=0.586303	lr=0.377519
Epoch[271] Batch [299]	Speed: 11628.466964 samples/sec	accuracy=0.585622	lr=0.376871
Epoch[271] Batch [349]	Speed: 10928.795103 samples/sec	accuracy=0.585416	lr=0.376223
Epoch[271] Batch [399]	Speed: 10846.787757 samples/sec	accuracy=0.585016	lr=0.375576
Epoch[271] Batch [449]	Speed: 11422.896856 samples/sec	accuracy=0.584714	lr=0.374929
Epoch[271] Batch [499]	Speed: 11166.898264 samples/sec	accuracy=0.584353	lr=0.374283
Epoch[271] Batch [549]	Speed: 11480.950728 samples/sec	accuracy=0.584011	lr=0.373637
Epoch[271] Batch [599]	Speed: 11264.933385 samples/sec	accuracy=0.583564	lr=0.372991
[Epoch 271] training: accuracy=0.583370
[Epoch 271] speed: 11047 samples/sec	time cost: 120.163487
[Epoch 271] validation: err-top1=0.397481 err-top5=0.171346
Epoch[272] Batch [49]	Speed: 11199.248072 samples/sec	accuracy=0.591299	lr=0.372024
Epoch[272] Batch [99]	Speed: 11146.897282 samples/sec	accuracy=0.589712	lr=0.371380
Epoch[272] Batch [149]	Speed: 10957.532048 samples/sec	accuracy=0.588376	lr=0.370736
Epoch[272] Batch [199]	Speed: 10964.197312 samples/sec	accuracy=0.587134	lr=0.370093
Epoch[272] Batch [249]	Speed: 10904.923591 samples/sec	accuracy=0.586385	lr=0.369450
Epoch[272] Batch [299]	Speed: 10606.792409 samples/sec	accuracy=0.586060	lr=0.368807
Epoch[272] Batch [349]	Speed: 11159.230456 samples/sec	accuracy=0.585533	lr=0.368165
Epoch[272] Batch [399]	Speed: 10435.681721 samples/sec	accuracy=0.585455	lr=0.367524
Epoch[272] Batch [449]	Speed: 11278.308135 samples/sec	accuracy=0.584639	lr=0.366883
Epoch[272] Batch [499]	Speed: 10969.982136 samples/sec	accuracy=0.584730	lr=0.366242
Epoch[272] Batch [549]	Speed: 11298.903998 samples/sec	accuracy=0.584375	lr=0.365602
Epoch[272] Batch [599]	Speed: 11667.126327 samples/sec	accuracy=0.584100	lr=0.364962
[Epoch 272] training: accuracy=0.584044
[Epoch 272] speed: 11049 samples/sec	time cost: 120.708889
[Epoch 272] validation: err-top1=0.397188 err-top5=0.173105
Epoch[273] Batch [49]	Speed: 11006.284998 samples/sec	accuracy=0.591240	lr=0.363991
Epoch[273] Batch [99]	Speed: 11231.283319 samples/sec	accuracy=0.589302	lr=0.363352
Epoch[273] Batch [149]	Speed: 11404.687700 samples/sec	accuracy=0.588548	lr=0.362714
Epoch[273] Batch [199]	Speed: 10561.505020 samples/sec	accuracy=0.588418	lr=0.362077
Epoch[273] Batch [249]	Speed: 10908.407251 samples/sec	accuracy=0.587709	lr=0.361440
Epoch[273] Batch [299]	Speed: 10945.827340 samples/sec	accuracy=0.587544	lr=0.360803
Epoch[273] Batch [349]	Speed: 11022.449413 samples/sec	accuracy=0.587274	lr=0.360167
Epoch[273] Batch [399]	Speed: 11217.862169 samples/sec	accuracy=0.586771	lr=0.359532
Epoch[273] Batch [449]	Speed: 11407.818365 samples/sec	accuracy=0.586421	lr=0.358896
Epoch[273] Batch [499]	Speed: 11321.588726 samples/sec	accuracy=0.586042	lr=0.358262
Epoch[273] Batch [549]	Speed: 11169.142164 samples/sec	accuracy=0.585670	lr=0.357627
Epoch[273] Batch [599]	Speed: 11174.989190 samples/sec	accuracy=0.585436	lr=0.356994
[Epoch 273] training: accuracy=0.585628
[Epoch 273] speed: 11079 samples/sec	time cost: 120.194088
[Epoch 273] validation: err-top1=0.407023 err-top5=0.178935
Epoch[274] Batch [49]	Speed: 11074.063971 samples/sec	accuracy=0.593350	lr=0.356031
Epoch[274] Batch [99]	Speed: 10876.456332 samples/sec	accuracy=0.591504	lr=0.355399
Epoch[274] Batch [149]	Speed: 11177.602859 samples/sec	accuracy=0.589941	lr=0.354767
Epoch[274] Batch [199]	Speed: 10627.795453 samples/sec	accuracy=0.589885	lr=0.354135
Epoch[274] Batch [249]	Speed: 10696.076567 samples/sec	accuracy=0.589830	lr=0.353504
Epoch[274] Batch [299]	Speed: 11264.179720 samples/sec	accuracy=0.589321	lr=0.352873
Epoch[274] Batch [349]	Speed: 10495.939216 samples/sec	accuracy=0.588537	lr=0.352243
Epoch[274] Batch [399]	Speed: 11307.545236 samples/sec	accuracy=0.587911	lr=0.351613
Epoch[274] Batch [449]	Speed: 11563.179272 samples/sec	accuracy=0.587431	lr=0.350984
Epoch[274] Batch [499]	Speed: 11059.525254 samples/sec	accuracy=0.586972	lr=0.350355
Epoch[274] Batch [549]	Speed: 10743.725075 samples/sec	accuracy=0.586627	lr=0.349727
Epoch[274] Batch [599]	Speed: 11629.002840 samples/sec	accuracy=0.586199	lr=0.349099
[Epoch 274] training: accuracy=0.586109
[Epoch 274] speed: 11019 samples/sec	time cost: 120.594903
[Epoch 274] validation: err-top1=0.404521 err-top5=0.178304
Epoch[275] Batch [49]	Speed: 10872.741468 samples/sec	accuracy=0.593105	lr=0.348158
Epoch[275] Batch [99]	Speed: 11259.248077 samples/sec	accuracy=0.592832	lr=0.347532
Epoch[275] Batch [149]	Speed: 11092.243870 samples/sec	accuracy=0.590605	lr=0.346906
Epoch[275] Batch [199]	Speed: 11434.050943 samples/sec	accuracy=0.590330	lr=0.346280
Epoch[275] Batch [249]	Speed: 11280.673774 samples/sec	accuracy=0.589779	lr=0.345655
Epoch[275] Batch [299]	Speed: 11527.146815 samples/sec	accuracy=0.589282	lr=0.345030
Epoch[275] Batch [349]	Speed: 11486.967878 samples/sec	accuracy=0.588823	lr=0.344406
Epoch[275] Batch [399]	Speed: 11538.758787 samples/sec	accuracy=0.588260	lr=0.343782
Epoch[275] Batch [449]	Speed: 11318.019930 samples/sec	accuracy=0.587726	lr=0.343159
Epoch[275] Batch [499]	Speed: 11291.895179 samples/sec	accuracy=0.587439	lr=0.342536
Epoch[275] Batch [549]	Speed: 11407.817153 samples/sec	accuracy=0.587203	lr=0.341914
Epoch[275] Batch [599]	Speed: 10863.771507 samples/sec	accuracy=0.587226	lr=0.341292
[Epoch 275] training: accuracy=0.587299
[Epoch 275] speed: 11256 samples/sec	time cost: 118.541960
[Epoch 275] validation: err-top1=0.401953 err-top5=0.172500
Epoch[276] Batch [49]	Speed: 10836.167968 samples/sec	accuracy=0.596924	lr=0.340348
Epoch[276] Batch [99]	Speed: 11631.225901 samples/sec	accuracy=0.594814	lr=0.339727
Epoch[276] Batch [149]	Speed: 10901.955734 samples/sec	accuracy=0.593278	lr=0.339107
Epoch[276] Batch [199]	Speed: 11301.948000 samples/sec	accuracy=0.592393	lr=0.338487
Epoch[276] Batch [249]	Speed: 11572.716366 samples/sec	accuracy=0.591932	lr=0.337868
Epoch[276] Batch [299]	Speed: 11124.202881 samples/sec	accuracy=0.591235	lr=0.337249
Epoch[276] Batch [349]	Speed: 11225.712896 samples/sec	accuracy=0.590820	lr=0.336631
Epoch[276] Batch [399]	Speed: 11137.877565 samples/sec	accuracy=0.590364	lr=0.336013
Epoch[276] Batch [449]	Speed: 11506.491723 samples/sec	accuracy=0.590098	lr=0.335396
Epoch[276] Batch [499]	Speed: 11371.500800 samples/sec	accuracy=0.589593	lr=0.334779
Epoch[276] Batch [549]	Speed: 11436.759805 samples/sec	accuracy=0.589345	lr=0.334163
Epoch[276] Batch [599]	Speed: 11267.418449 samples/sec	accuracy=0.588915	lr=0.333547
[Epoch 276] training: accuracy=0.588762
[Epoch 276] speed: 11259 samples/sec	time cost: 118.079997
[Epoch 276] validation: err-top1=0.415222 err-top5=0.183960
Epoch[277] Batch [49]	Speed: 10918.832458 samples/sec	accuracy=0.596816	lr=0.332624
Epoch[277] Batch [99]	Speed: 11435.169711 samples/sec	accuracy=0.596084	lr=0.332010
Epoch[277] Batch [149]	Speed: 10928.796216 samples/sec	accuracy=0.594906	lr=0.331396
Epoch[277] Batch [199]	Speed: 11282.785498 samples/sec	accuracy=0.594277	lr=0.330782
Epoch[277] Batch [249]	Speed: 11383.295016 samples/sec	accuracy=0.593504	lr=0.330169
Epoch[277] Batch [299]	Speed: 11755.554147 samples/sec	accuracy=0.592760	lr=0.329556
Epoch[277] Batch [349]	Speed: 11089.795666 samples/sec	accuracy=0.592238	lr=0.328944
Epoch[277] Batch [399]	Speed: 11189.502099 samples/sec	accuracy=0.591592	lr=0.328333
Epoch[277] Batch [449]	Speed: 11637.761434 samples/sec	accuracy=0.590837	lr=0.327721
Epoch[277] Batch [499]	Speed: 11400.463497 samples/sec	accuracy=0.590520	lr=0.327111
Epoch[277] Batch [549]	Speed: 11169.732691 samples/sec	accuracy=0.590294	lr=0.326500
Epoch[277] Batch [599]	Speed: 11736.424450 samples/sec	accuracy=0.589998	lr=0.325891
[Epoch 277] training: accuracy=0.589744
[Epoch 277] speed: 11290 samples/sec	time cost: 118.126209
[Epoch 277] validation: err-top1=0.392324 err-top5=0.168633
Epoch[278] Batch [49]	Speed: 11208.601507 samples/sec	accuracy=0.594600	lr=0.324965
Epoch[278] Batch [99]	Speed: 10948.687676 samples/sec	accuracy=0.594814	lr=0.324356
Epoch[278] Batch [149]	Speed: 11212.463697 samples/sec	accuracy=0.593688	lr=0.323748
Epoch[278] Batch [199]	Speed: 11321.078121 samples/sec	accuracy=0.593245	lr=0.323141
Epoch[278] Batch [249]	Speed: 10553.143470 samples/sec	accuracy=0.592771	lr=0.322534
Epoch[278] Batch [299]	Speed: 10730.854981 samples/sec	accuracy=0.592376	lr=0.321927
Epoch[278] Batch [349]	Speed: 11226.766320 samples/sec	accuracy=0.592062	lr=0.321321
Epoch[278] Batch [399]	Speed: 11244.185530 samples/sec	accuracy=0.591497	lr=0.320716
Epoch[278] Batch [449]	Speed: 11550.803594 samples/sec	accuracy=0.591259	lr=0.320111
Epoch[278] Batch [499]	Speed: 11483.613390 samples/sec	accuracy=0.590847	lr=0.319506
Epoch[278] Batch [549]	Speed: 11332.370227 samples/sec	accuracy=0.590632	lr=0.318902
Epoch[278] Batch [599]	Speed: 11441.832971 samples/sec	accuracy=0.590420	lr=0.318299
[Epoch 278] training: accuracy=0.590376
[Epoch 278] speed: 11171 samples/sec	time cost: 118.930521
[Epoch 278] validation: err-top1=0.407145 err-top5=0.178426
Epoch[279] Batch [49]	Speed: 11133.448070 samples/sec	accuracy=0.595674	lr=0.317394
Epoch[279] Batch [99]	Speed: 11048.048770 samples/sec	accuracy=0.595571	lr=0.316792
Epoch[279] Batch [149]	Speed: 10764.965518 samples/sec	accuracy=0.595430	lr=0.316190
Epoch[279] Batch [199]	Speed: 10971.551141 samples/sec	accuracy=0.594011	lr=0.315588
Epoch[279] Batch [249]	Speed: 11171.719099 samples/sec	accuracy=0.593729	lr=0.314988
Epoch[279] Batch [299]	Speed: 11120.620937 samples/sec	accuracy=0.593341	lr=0.314387
Epoch[279] Batch [349]	Speed: 11631.644847 samples/sec	accuracy=0.593288	lr=0.313787
Epoch[279] Batch [399]	Speed: 11126.110294 samples/sec	accuracy=0.592897	lr=0.313188
Epoch[279] Batch [449]	Speed: 11311.730939 samples/sec	accuracy=0.592729	lr=0.312589
Epoch[279] Batch [499]	Speed: 10734.638237 samples/sec	accuracy=0.592244	lr=0.311991
Epoch[279] Batch [549]	Speed: 11111.047202 samples/sec	accuracy=0.591881	lr=0.311393
Epoch[279] Batch [599]	Speed: 11292.541513 samples/sec	accuracy=0.591700	lr=0.310795
[Epoch 279] training: accuracy=0.591596
[Epoch 279] speed: 11095 samples/sec	time cost: 120.029788
[Epoch 279] validation: err-top1=0.382935 err-top5=0.162638
Epoch[280] Batch [49]	Speed: 11211.964058 samples/sec	accuracy=0.600410	lr=0.309888
Epoch[280] Batch [99]	Speed: 11708.364836 samples/sec	accuracy=0.598130	lr=0.309292
Epoch[280] Batch [149]	Speed: 10719.499257 samples/sec	accuracy=0.596816	lr=0.308696
Epoch[280] Batch [199]	Speed: 10799.914103 samples/sec	accuracy=0.596782	lr=0.308101
Epoch[280] Batch [249]	Speed: 11280.670218 samples/sec	accuracy=0.595990	lr=0.307507
Epoch[280] Batch [299]	Speed: 11350.701464 samples/sec	accuracy=0.594777	lr=0.306912
Epoch[280] Batch [349]	Speed: 11259.828689 samples/sec	accuracy=0.594160	lr=0.306319
Epoch[280] Batch [399]	Speed: 10619.328913 samples/sec	accuracy=0.593579	lr=0.305726
Epoch[280] Batch [449]	Speed: 11128.431540 samples/sec	accuracy=0.592916	lr=0.305133
Epoch[280] Batch [499]	Speed: 10986.754391 samples/sec	accuracy=0.592849	lr=0.304541
Epoch[280] Batch [549]	Speed: 11586.833976 samples/sec	accuracy=0.592473	lr=0.303949
Epoch[280] Batch [599]	Speed: 11365.141124 samples/sec	accuracy=0.592306	lr=0.303358
[Epoch 280] training: accuracy=0.592128
[Epoch 280] speed: 11108 samples/sec	time cost: 120.125522
[Epoch 280] validation: err-top1=0.391504 err-top5=0.167363
Epoch[281] Batch [49]	Speed: 10905.361350 samples/sec	accuracy=0.601250	lr=0.302460
Epoch[281] Batch [99]	Speed: 11301.853426 samples/sec	accuracy=0.598911	lr=0.301870
Epoch[281] Batch [149]	Speed: 10703.864835 samples/sec	accuracy=0.597243	lr=0.301281
Epoch[281] Batch [199]	Speed: 11849.137825 samples/sec	accuracy=0.597007	lr=0.300692
Epoch[281] Batch [249]	Speed: 11390.049874 samples/sec	accuracy=0.595498	lr=0.300103
Epoch[281] Batch [299]	Speed: 11515.429442 samples/sec	accuracy=0.595208	lr=0.299516
Epoch[281] Batch [349]	Speed: 11534.827188 samples/sec	accuracy=0.595025	lr=0.298928
Epoch[281] Batch [399]	Speed: 11502.959459 samples/sec	accuracy=0.594423	lr=0.298341
Epoch[281] Batch [449]	Speed: 11029.099032 samples/sec	accuracy=0.594104	lr=0.297755
Epoch[281] Batch [499]	Speed: 10840.298910 samples/sec	accuracy=0.593733	lr=0.297169
Epoch[281] Batch [549]	Speed: 11252.176038 samples/sec	accuracy=0.593590	lr=0.296583
Epoch[281] Batch [599]	Speed: 11011.522283 samples/sec	accuracy=0.593110	lr=0.295999
[Epoch 281] training: accuracy=0.593134
[Epoch 281] speed: 11198 samples/sec	time cost: 118.773718
[Epoch 281] validation: err-top1=0.385050 err-top5=0.165039
Epoch[282] Batch [49]	Speed: 10719.836903 samples/sec	accuracy=0.598623	lr=0.295122
Epoch[282] Batch [99]	Speed: 11156.513217 samples/sec	accuracy=0.599014	lr=0.294539
Epoch[282] Batch [149]	Speed: 11028.401511 samples/sec	accuracy=0.598512	lr=0.293955
Epoch[282] Batch [199]	Speed: 11091.619973 samples/sec	accuracy=0.597637	lr=0.293373
Epoch[282] Batch [249]	Speed: 10923.338930 samples/sec	accuracy=0.596801	lr=0.292791
Epoch[282] Batch [299]	Speed: 11078.101432 samples/sec	accuracy=0.596367	lr=0.292209
Epoch[282] Batch [349]	Speed: 11309.830833 samples/sec	accuracy=0.595678	lr=0.291628
Epoch[282] Batch [399]	Speed: 10728.347958 samples/sec	accuracy=0.595315	lr=0.291047
Epoch[282] Batch [449]	Speed: 10975.263729 samples/sec	accuracy=0.594988	lr=0.290467
Epoch[282] Batch [499]	Speed: 11576.005182 samples/sec	accuracy=0.594603	lr=0.289888
Epoch[282] Batch [549]	Speed: 11097.068688 samples/sec	accuracy=0.594401	lr=0.289308
Epoch[282] Batch [599]	Speed: 10695.056458 samples/sec	accuracy=0.594227	lr=0.288730
[Epoch 282] training: accuracy=0.594073
[Epoch 282] speed: 11005 samples/sec	time cost: 121.251163
[Epoch 282] validation: err-top1=0.377539 err-top5=0.156602
Epoch[283] Batch [49]	Speed: 11453.683806 samples/sec	accuracy=0.601777	lr=0.287851
Epoch[283] Batch [99]	Speed: 11424.914467 samples/sec	accuracy=0.601045	lr=0.287274
Epoch[283] Batch [149]	Speed: 11204.031708 samples/sec	accuracy=0.600632	lr=0.286697
Epoch[283] Batch [199]	Speed: 11346.348392 samples/sec	accuracy=0.600452	lr=0.286121
Epoch[283] Batch [249]	Speed: 10272.904221 samples/sec	accuracy=0.599697	lr=0.285545
Epoch[283] Batch [299]	Speed: 11043.672248 samples/sec	accuracy=0.599181	lr=0.284970
Epoch[283] Batch [349]	Speed: 10665.314581 samples/sec	accuracy=0.598662	lr=0.284395
Epoch[283] Batch [399]	Speed: 11069.303351 samples/sec	accuracy=0.598070	lr=0.283821
Epoch[283] Batch [449]	Speed: 11487.552857 samples/sec	accuracy=0.597550	lr=0.283247
Epoch[283] Batch [499]	Speed: 10934.904529 samples/sec	accuracy=0.597055	lr=0.282674
Epoch[283] Batch [549]	Speed: 11019.910901 samples/sec	accuracy=0.596645	lr=0.282101
Epoch[283] Batch [599]	Speed: 11215.178976 samples/sec	accuracy=0.596204	lr=0.281529
[Epoch 283] training: accuracy=0.596073
[Epoch 283] speed: 11063 samples/sec	time cost: 119.977812
[Epoch 283] validation: err-top1=0.382812 err-top5=0.161560
Epoch[284] Batch [49]	Speed: 11577.005235 samples/sec	accuracy=0.603135	lr=0.280671
Epoch[284] Batch [99]	Speed: 11476.944943 samples/sec	accuracy=0.601914	lr=0.280100
Epoch[284] Batch [149]	Speed: 11391.835018 samples/sec	accuracy=0.601152	lr=0.279530
Epoch[284] Batch [199]	Speed: 11265.336997 samples/sec	accuracy=0.600693	lr=0.278960
Epoch[284] Batch [249]	Speed: 10836.539799 samples/sec	accuracy=0.600068	lr=0.278391
Epoch[284] Batch [299]	Speed: 10134.065376 samples/sec	accuracy=0.599775	lr=0.277822
Epoch[284] Batch [349]	Speed: 10777.404566 samples/sec	accuracy=0.599302	lr=0.277253
Epoch[284] Batch [399]	Speed: 11261.335547 samples/sec	accuracy=0.598385	lr=0.276685
Epoch[284] Batch [449]	Speed: 11212.897222 samples/sec	accuracy=0.598223	lr=0.276118
Epoch[284] Batch [499]	Speed: 11099.504615 samples/sec	accuracy=0.597842	lr=0.275551
Epoch[284] Batch [549]	Speed: 11311.812271 samples/sec	accuracy=0.597498	lr=0.274985
Epoch[284] Batch [599]	Speed: 11227.154288 samples/sec	accuracy=0.597103	lr=0.274419
[Epoch 284] training: accuracy=0.597077
[Epoch 284] speed: 11095 samples/sec	time cost: 120.047155
[Epoch 284] validation: err-top1=0.385703 err-top5=0.164746
Epoch[285] Batch [49]	Speed: 10366.133153 samples/sec	accuracy=0.603135	lr=0.273560
Epoch[285] Batch [99]	Speed: 10939.766729 samples/sec	accuracy=0.603047	lr=0.272995
Epoch[285] Batch [149]	Speed: 11077.668838 samples/sec	accuracy=0.602327	lr=0.272431
Epoch[285] Batch [199]	Speed: 10938.429660 samples/sec	accuracy=0.601289	lr=0.271868
Epoch[285] Batch [249]	Speed: 11282.421239 samples/sec	accuracy=0.600420	lr=0.271305
Epoch[285] Batch [299]	Speed: 11302.343263 samples/sec	accuracy=0.599818	lr=0.270743
Epoch[285] Batch [349]	Speed: 11436.422383 samples/sec	accuracy=0.599229	lr=0.270181
Epoch[285] Batch [399]	Speed: 11171.199840 samples/sec	accuracy=0.598695	lr=0.269619
Epoch[285] Batch [449]	Speed: 11205.837662 samples/sec	accuracy=0.598346	lr=0.269058
Epoch[285] Batch [499]	Speed: 11285.303651 samples/sec	accuracy=0.597976	lr=0.268498
Epoch[285] Batch [549]	Speed: 10119.444986 samples/sec	accuracy=0.597702	lr=0.267938
Epoch[285] Batch [599]	Speed: 11074.284691 samples/sec	accuracy=0.597507	lr=0.267379
[Epoch 285] training: accuracy=0.597419
[Epoch 285] speed: 10984 samples/sec	time cost: 120.927065
[Epoch 285] validation: err-top1=0.383464 err-top5=0.160339
Epoch[286] Batch [49]	Speed: 11147.224779 samples/sec	accuracy=0.603750	lr=0.266540
Epoch[286] Batch [99]	Speed: 11618.670271 samples/sec	accuracy=0.603354	lr=0.265982
Epoch[286] Batch [149]	Speed: 11492.076778 samples/sec	accuracy=0.603262	lr=0.265425
Epoch[286] Batch [199]	Speed: 11432.082746 samples/sec	accuracy=0.602109	lr=0.264868
Epoch[286] Batch [249]	Speed: 11247.464610 samples/sec	accuracy=0.601437	lr=0.264311
Epoch[286] Batch [299]	Speed: 10801.734467 samples/sec	accuracy=0.601331	lr=0.263755
Epoch[286] Batch [349]	Speed: 10840.573342 samples/sec	accuracy=0.601112	lr=0.263200
Epoch[286] Batch [399]	Speed: 11291.336487 samples/sec	accuracy=0.600487	lr=0.262645
Epoch[286] Batch [449]	Speed: 11210.548510 samples/sec	accuracy=0.600471	lr=0.262090
Epoch[286] Batch [499]	Speed: 11553.549728 samples/sec	accuracy=0.600095	lr=0.261536
Epoch[286] Batch [549]	Speed: 11734.619774 samples/sec	accuracy=0.599860	lr=0.260983
Epoch[286] Batch [599]	Speed: 10817.651681 samples/sec	accuracy=0.599446	lr=0.260430
[Epoch 286] training: accuracy=0.599178
[Epoch 286] speed: 11186 samples/sec	time cost: 119.138738
[Epoch 286] validation: err-top1=0.383443 err-top5=0.162048
Epoch[287] Batch [49]	Speed: 10439.990480 samples/sec	accuracy=0.603301	lr=0.259591
Epoch[287] Batch [99]	Speed: 11037.927471 samples/sec	accuracy=0.603179	lr=0.259039
Epoch[287] Batch [149]	Speed: 10983.562905 samples/sec	accuracy=0.603187	lr=0.258488
Epoch[287] Batch [199]	Speed: 11315.983544 samples/sec	accuracy=0.603132	lr=0.257938
Epoch[287] Batch [249]	Speed: 11432.861179 samples/sec	accuracy=0.602301	lr=0.257388
Epoch[287] Batch [299]	Speed: 10457.628571 samples/sec	accuracy=0.601776	lr=0.256838
Epoch[287] Batch [349]	Speed: 11261.926415 samples/sec	accuracy=0.601265	lr=0.256289
Epoch[287] Batch [399]	Speed: 10859.907407 samples/sec	accuracy=0.600826	lr=0.255741
Epoch[287] Batch [449]	Speed: 11457.831088 samples/sec	accuracy=0.600547	lr=0.255193
Epoch[287] Batch [499]	Speed: 11250.645400 samples/sec	accuracy=0.600092	lr=0.254645
Epoch[287] Batch [549]	Speed: 11220.983435 samples/sec	accuracy=0.599636	lr=0.254099
Epoch[287] Batch [599]	Speed: 11236.224881 samples/sec	accuracy=0.599443	lr=0.253552
[Epoch 287] training: accuracy=0.599286
[Epoch 287] speed: 11053 samples/sec	time cost: 120.690993
[Epoch 287] validation: err-top1=0.375313 err-top5=0.156055
Epoch[288] Batch [49]	Speed: 10700.841160 samples/sec	accuracy=0.607607	lr=0.252723
Epoch[288] Batch [99]	Speed: 11269.856408 samples/sec	accuracy=0.607817	lr=0.252178
Epoch[288] Batch [149]	Speed: 11698.741881 samples/sec	accuracy=0.606725	lr=0.251633
Epoch[288] Batch [199]	Speed: 11227.672599 samples/sec	accuracy=0.606094	lr=0.251089
Epoch[288] Batch [249]	Speed: 10518.564001 samples/sec	accuracy=0.605066	lr=0.250546
Epoch[288] Batch [299]	Speed: 9824.047108 samples/sec	accuracy=0.604134	lr=0.250003
Epoch[288] Batch [349]	Speed: 11551.329851 samples/sec	accuracy=0.603772	lr=0.249461
Epoch[288] Batch [399]	Speed: 11333.921688 samples/sec	accuracy=0.603218	lr=0.248919
Epoch[288] Batch [449]	Speed: 10870.623325 samples/sec	accuracy=0.602568	lr=0.248377
Epoch[288] Batch [499]	Speed: 10800.171285 samples/sec	accuracy=0.602256	lr=0.247837
Epoch[288] Batch [549]	Speed: 11276.376903 samples/sec	accuracy=0.601855	lr=0.247296
Epoch[288] Batch [599]	Speed: 10858.583198 samples/sec	accuracy=0.601491	lr=0.246757
[Epoch 288] training: accuracy=0.601304
[Epoch 288] speed: 10954 samples/sec	time cost: 121.286881
[Epoch 288] validation: err-top1=0.395264 err-top5=0.171753
Epoch[289] Batch [49]	Speed: 10491.081868 samples/sec	accuracy=0.608262	lr=0.245948
Epoch[289] Batch [99]	Speed: 10960.454725 samples/sec	accuracy=0.607588	lr=0.245410
Epoch[289] Batch [149]	Speed: 11279.155219 samples/sec	accuracy=0.606546	lr=0.244872
Epoch[289] Batch [199]	Speed: 11195.479317 samples/sec	accuracy=0.605723	lr=0.244334
Epoch[289] Batch [249]	Speed: 11212.371493 samples/sec	accuracy=0.605227	lr=0.243797
Epoch[289] Batch [299]	Speed: 11371.941291 samples/sec	accuracy=0.604679	lr=0.243261
Epoch[289] Batch [349]	Speed: 11350.164234 samples/sec	accuracy=0.604268	lr=0.242725
Epoch[289] Batch [399]	Speed: 11200.746354 samples/sec	accuracy=0.603507	lr=0.242190
Epoch[289] Batch [449]	Speed: 10515.741673 samples/sec	accuracy=0.603163	lr=0.241655
Epoch[289] Batch [499]	Speed: 11385.170394 samples/sec	accuracy=0.602686	lr=0.241121
Epoch[289] Batch [549]	Speed: 10906.123150 samples/sec	accuracy=0.602409	lr=0.240587
Epoch[289] Batch [599]	Speed: 10673.552240 samples/sec	accuracy=0.602051	lr=0.240054
[Epoch 289] training: accuracy=0.601882
[Epoch 289] speed: 11004 samples/sec	time cost: 121.114900
[Epoch 289] validation: err-top1=0.373711 err-top5=0.155254
Epoch[290] Batch [49]	Speed: 10924.472799 samples/sec	accuracy=0.607979	lr=0.239245
Epoch[290] Batch [99]	Speed: 11206.018056 samples/sec	accuracy=0.607544	lr=0.238713
Epoch[290] Batch [149]	Speed: 10903.444167 samples/sec	accuracy=0.606989	lr=0.238182
Epoch[290] Batch [199]	Speed: 11122.706859 samples/sec	accuracy=0.606584	lr=0.237651
Epoch[290] Batch [249]	Speed: 10983.537064 samples/sec	accuracy=0.605818	lr=0.237121
Epoch[290] Batch [299]	Speed: 11174.995296 samples/sec	accuracy=0.605451	lr=0.236591
Epoch[290] Batch [349]	Speed: 11398.382214 samples/sec	accuracy=0.605061	lr=0.236062
Epoch[290] Batch [399]	Speed: 11552.166555 samples/sec	accuracy=0.604636	lr=0.235534
Epoch[290] Batch [449]	Speed: 11127.306831 samples/sec	accuracy=0.604326	lr=0.235006
Epoch[290] Batch [499]	Speed: 10923.893471 samples/sec	accuracy=0.604027	lr=0.234478
Epoch[290] Batch [549]	Speed: 10622.131205 samples/sec	accuracy=0.603608	lr=0.233951
Epoch[290] Batch [599]	Speed: 11114.343126 samples/sec	accuracy=0.603047	lr=0.233425
[Epoch 290] training: accuracy=0.602874
[Epoch 290] speed: 11082 samples/sec	time cost: 119.921388
[Epoch 290] validation: err-top1=0.376872 err-top5=0.158162
Epoch[291] Batch [49]	Speed: 10701.110709 samples/sec	accuracy=0.610771	lr=0.232636
Epoch[291] Batch [99]	Speed: 11087.318203 samples/sec	accuracy=0.609126	lr=0.232111
Epoch[291] Batch [149]	Speed: 11245.497109 samples/sec	accuracy=0.608473	lr=0.231586
Epoch[291] Batch [199]	Speed: 11280.937177 samples/sec	accuracy=0.607468	lr=0.231062
Epoch[291] Batch [249]	Speed: 11088.727994 samples/sec	accuracy=0.607432	lr=0.230539
Epoch[291] Batch [299]	Speed: 10953.062127 samples/sec	accuracy=0.606927	lr=0.230016
Epoch[291] Batch [349]	Speed: 11359.845774 samples/sec	accuracy=0.606483	lr=0.229493
Epoch[291] Batch [399]	Speed: 11550.622491 samples/sec	accuracy=0.606157	lr=0.228971
Epoch[291] Batch [449]	Speed: 11316.826457 samples/sec	accuracy=0.605563	lr=0.228450
Epoch[291] Batch [499]	Speed: 11067.733644 samples/sec	accuracy=0.605324	lr=0.227929
Epoch[291] Batch [549]	Speed: 10715.517330 samples/sec	accuracy=0.605066	lr=0.227409
Epoch[291] Batch [599]	Speed: 11342.065450 samples/sec	accuracy=0.604708	lr=0.226889
[Epoch 291] training: accuracy=0.604738
[Epoch 291] speed: 11116 samples/sec	time cost: 119.811211
[Epoch 291] validation: err-top1=0.375448 err-top5=0.154724
Epoch[292] Batch [49]	Speed: 10216.538740 samples/sec	accuracy=0.612061	lr=0.226100
Epoch[292] Batch [99]	Speed: 10658.299815 samples/sec	accuracy=0.610415	lr=0.225582
Epoch[292] Batch [149]	Speed: 10832.278671 samples/sec	accuracy=0.609202	lr=0.225064
Epoch[292] Batch [199]	Speed: 11345.748634 samples/sec	accuracy=0.607900	lr=0.224546
Epoch[292] Batch [249]	Speed: 11758.143243 samples/sec	accuracy=0.607395	lr=0.224030
Epoch[292] Batch [299]	Speed: 10596.757935 samples/sec	accuracy=0.607360	lr=0.223513
Epoch[292] Batch [349]	Speed: 11666.565382 samples/sec	accuracy=0.606811	lr=0.222998
Epoch[292] Batch [399]	Speed: 11311.883476 samples/sec	accuracy=0.606599	lr=0.222482
Epoch[292] Batch [449]	Speed: 10990.359158 samples/sec	accuracy=0.606128	lr=0.221968
Epoch[292] Batch [499]	Speed: 11550.429589 samples/sec	accuracy=0.605674	lr=0.221454
Epoch[292] Batch [549]	Speed: 11431.297726 samples/sec	accuracy=0.605313	lr=0.220940
Epoch[292] Batch [599]	Speed: 11247.574475 samples/sec	accuracy=0.605062	lr=0.220427
[Epoch 292] training: accuracy=0.604784
[Epoch 292] speed: 11123 samples/sec	time cost: 119.892882
[Epoch 292] validation: err-top1=0.384238 err-top5=0.161445
Epoch[293] Batch [49]	Speed: 11437.048517 samples/sec	accuracy=0.612578	lr=0.219659
Epoch[293] Batch [99]	Speed: 11128.609450 samples/sec	accuracy=0.612222	lr=0.219147
Epoch[293] Batch [149]	Speed: 11203.938474 samples/sec	accuracy=0.611084	lr=0.218636
Epoch[293] Batch [199]	Speed: 11259.569812 samples/sec	accuracy=0.610471	lr=0.218125
Epoch[293] Batch [249]	Speed: 11513.695480 samples/sec	accuracy=0.609680	lr=0.217615
Epoch[293] Batch [299]	Speed: 11183.059528 samples/sec	accuracy=0.609243	lr=0.217106
Epoch[293] Batch [349]	Speed: 11090.857242 samples/sec	accuracy=0.608488	lr=0.216597
Epoch[293] Batch [399]	Speed: 11651.388707 samples/sec	accuracy=0.608174	lr=0.216088
Epoch[293] Batch [449]	Speed: 11002.792953 samples/sec	accuracy=0.607766	lr=0.215581
Epoch[293] Batch [499]	Speed: 11168.409683 samples/sec	accuracy=0.607097	lr=0.215073
Epoch[293] Batch [549]	Speed: 11218.155758 samples/sec	accuracy=0.606809	lr=0.214567
Epoch[293] Batch [599]	Speed: 11257.772466 samples/sec	accuracy=0.606471	lr=0.214060
[Epoch 293] training: accuracy=0.606454
[Epoch 293] speed: 11233 samples/sec	time cost: 118.648521
[Epoch 293] validation: err-top1=0.372457 err-top5=0.156311
Epoch[294] Batch [49]	Speed: 10704.747618 samples/sec	accuracy=0.612119	lr=0.213292
Epoch[294] Batch [99]	Speed: 11007.788796 samples/sec	accuracy=0.611235	lr=0.212787
Epoch[294] Batch [149]	Speed: 11752.259648 samples/sec	accuracy=0.610872	lr=0.212283
Epoch[294] Batch [199]	Speed: 11223.584351 samples/sec	accuracy=0.609934	lr=0.211779
Epoch[294] Batch [249]	Speed: 11128.136574 samples/sec	accuracy=0.609705	lr=0.211276
Epoch[294] Batch [299]	Speed: 11629.531209 samples/sec	accuracy=0.609479	lr=0.210773
Epoch[294] Batch [349]	Speed: 11348.020618 samples/sec	accuracy=0.609189	lr=0.210271
Epoch[294] Batch [399]	Speed: 11202.143066 samples/sec	accuracy=0.608934	lr=0.209769
Epoch[294] Batch [449]	Speed: 10785.666300 samples/sec	accuracy=0.608437	lr=0.209268
Epoch[294] Batch [499]	Speed: 11173.416110 samples/sec	accuracy=0.608229	lr=0.208768
Epoch[294] Batch [549]	Speed: 11108.546158 samples/sec	accuracy=0.607907	lr=0.208268
Epoch[294] Batch [599]	Speed: 11575.920318 samples/sec	accuracy=0.607629	lr=0.207769
[Epoch 294] training: accuracy=0.607627
[Epoch 294] speed: 11138 samples/sec	time cost: 119.823490
[Epoch 294] validation: err-top1=0.367129 err-top5=0.150000
Epoch[295] Batch [49]	Speed: 10686.190247 samples/sec	accuracy=0.616865	lr=0.207010
Epoch[295] Batch [99]	Speed: 11332.942256 samples/sec	accuracy=0.614614	lr=0.206512
Epoch[295] Batch [149]	Speed: 11460.607194 samples/sec	accuracy=0.613070	lr=0.206015
Epoch[295] Batch [199]	Speed: 11178.138715 samples/sec	accuracy=0.612515	lr=0.205518
Epoch[295] Batch [249]	Speed: 10368.683739 samples/sec	accuracy=0.611400	lr=0.205022
Epoch[295] Batch [299]	Speed: 11145.367096 samples/sec	accuracy=0.611170	lr=0.204526
Epoch[295] Batch [349]	Speed: 10830.299156 samples/sec	accuracy=0.610598	lr=0.204031
Epoch[295] Batch [399]	Speed: 11053.860134 samples/sec	accuracy=0.610226	lr=0.203536
Epoch[295] Batch [449]	Speed: 10929.318493 samples/sec	accuracy=0.609807	lr=0.203042
Epoch[295] Batch [499]	Speed: 10179.437465 samples/sec	accuracy=0.609406	lr=0.202548
Epoch[295] Batch [549]	Speed: 11647.855068 samples/sec	accuracy=0.609202	lr=0.202055
Epoch[295] Batch [599]	Speed: 10977.314551 samples/sec	accuracy=0.608841	lr=0.201563
[Epoch 295] training: accuracy=0.608792
[Epoch 295] speed: 10925 samples/sec	time cost: 121.407791
[Epoch 295] validation: err-top1=0.378113 err-top5=0.158671
Epoch[296] Batch [49]	Speed: 11389.235582 samples/sec	accuracy=0.616699	lr=0.200825
Epoch[296] Batch [99]	Speed: 11385.747766 samples/sec	accuracy=0.614985	lr=0.200334
Epoch[296] Batch [149]	Speed: 11354.266891 samples/sec	accuracy=0.613841	lr=0.199843
Epoch[296] Batch [199]	Speed: 10892.102433 samples/sec	accuracy=0.613340	lr=0.199353
Epoch[296] Batch [249]	Speed: 11233.257013 samples/sec	accuracy=0.612842	lr=0.198863
Epoch[296] Batch [299]	Speed: 11020.969038 samples/sec	accuracy=0.612314	lr=0.198375
Epoch[296] Batch [349]	Speed: 10977.866449 samples/sec	accuracy=0.611703	lr=0.197886
Epoch[296] Batch [399]	Speed: 11443.449309 samples/sec	accuracy=0.611229	lr=0.197398
Epoch[296] Batch [449]	Speed: 11199.473519 samples/sec	accuracy=0.610980	lr=0.196911
Epoch[296] Batch [499]	Speed: 10987.830625 samples/sec	accuracy=0.610390	lr=0.196424
Epoch[296] Batch [549]	Speed: 10919.773267 samples/sec	accuracy=0.609998	lr=0.195938
Epoch[296] Batch [599]	Speed: 11030.282159 samples/sec	accuracy=0.609879	lr=0.195453
[Epoch 296] training: accuracy=0.609760
[Epoch 296] speed: 11119 samples/sec	time cost: 119.673316
[Epoch 296] validation: err-top1=0.365377 err-top5=0.150228
Epoch[297] Batch [49]	Speed: 11098.244934 samples/sec	accuracy=0.614395	lr=0.194715
Epoch[297] Batch [99]	Speed: 11374.046965 samples/sec	accuracy=0.614346	lr=0.194231
Epoch[297] Batch [149]	Speed: 10642.527074 samples/sec	accuracy=0.613151	lr=0.193748
Epoch[297] Batch [199]	Speed: 11501.195989 samples/sec	accuracy=0.612432	lr=0.193264
Epoch[297] Batch [249]	Speed: 11106.645911 samples/sec	accuracy=0.612488	lr=0.192782
Epoch[297] Batch [299]	Speed: 10868.337692 samples/sec	accuracy=0.612157	lr=0.192300
Epoch[297] Batch [349]	Speed: 11315.402195 samples/sec	accuracy=0.611851	lr=0.191818
Epoch[297] Batch [399]	Speed: 11262.808254 samples/sec	accuracy=0.611422	lr=0.191338
Epoch[297] Batch [449]	Speed: 10780.312294 samples/sec	accuracy=0.611393	lr=0.190857
Epoch[297] Batch [499]	Speed: 10803.658169 samples/sec	accuracy=0.611146	lr=0.190377
Epoch[297] Batch [549]	Speed: 11234.847870 samples/sec	accuracy=0.610847	lr=0.189898
Epoch[297] Batch [599]	Speed: 11289.455392 samples/sec	accuracy=0.610474	lr=0.189419
[Epoch 297] training: accuracy=0.610391
[Epoch 297] speed: 11100 samples/sec	time cost: 119.761548
[Epoch 297] validation: err-top1=0.370488 err-top5=0.154961
Epoch[298] Batch [49]	Speed: 11005.964602 samples/sec	accuracy=0.618623	lr=0.188703
Epoch[298] Batch [99]	Speed: 11236.945117 samples/sec	accuracy=0.617764	lr=0.188225
Epoch[298] Batch [149]	Speed: 11324.519853 samples/sec	accuracy=0.616683	lr=0.187749
Epoch[298] Batch [199]	Speed: 11519.302719 samples/sec	accuracy=0.615613	lr=0.187272
Epoch[298] Batch [249]	Speed: 11198.449445 samples/sec	accuracy=0.614816	lr=0.186797
Epoch[298] Batch [299]	Speed: 11190.333271 samples/sec	accuracy=0.614285	lr=0.186322
Epoch[298] Batch [349]	Speed: 11274.679552 samples/sec	accuracy=0.613761	lr=0.185847
Epoch[298] Batch [399]	Speed: 11486.257627 samples/sec	accuracy=0.613232	lr=0.185373
Epoch[298] Batch [449]	Speed: 11223.686712 samples/sec	accuracy=0.612870	lr=0.184900
Epoch[298] Batch [499]	Speed: 11159.059104 samples/sec	accuracy=0.612349	lr=0.184427
Epoch[298] Batch [549]	Speed: 10774.849529 samples/sec	accuracy=0.611814	lr=0.183955
Epoch[298] Batch [599]	Speed: 11527.036060 samples/sec	accuracy=0.611392	lr=0.183483
[Epoch 298] training: accuracy=0.611270
[Epoch 298] speed: 11217 samples/sec	time cost: 118.668172
[Epoch 298] validation: err-top1=0.384705 err-top5=0.163940
Epoch[299] Batch [49]	Speed: 10355.654949 samples/sec	accuracy=0.617100	lr=0.182767
Epoch[299] Batch [99]	Speed: 11562.386728 samples/sec	accuracy=0.615640	lr=0.182297
Epoch[299] Batch [149]	Speed: 10944.522813 samples/sec	accuracy=0.615003	lr=0.181827
Epoch[299] Batch [199]	Speed: 11444.600109 samples/sec	accuracy=0.614475	lr=0.181358
Epoch[299] Batch [249]	Speed: 11370.933903 samples/sec	accuracy=0.614070	lr=0.180889
Epoch[299] Batch [299]	Speed: 11345.619159 samples/sec	accuracy=0.614064	lr=0.180421
Epoch[299] Batch [349]	Speed: 11431.462328 samples/sec	accuracy=0.613810	lr=0.179954
Epoch[299] Batch [399]	Speed: 11196.990018 samples/sec	accuracy=0.613589	lr=0.179487
Epoch[299] Batch [449]	Speed: 11418.204415 samples/sec	accuracy=0.613360	lr=0.179021
Epoch[299] Batch [499]	Speed: 11402.047887 samples/sec	accuracy=0.613354	lr=0.178555
Epoch[299] Batch [549]	Speed: 11476.494747 samples/sec	accuracy=0.613134	lr=0.178090
Epoch[299] Batch [599]	Speed: 10823.463694 samples/sec	accuracy=0.612927	lr=0.177625
[Epoch 299] training: accuracy=0.612654
[Epoch 299] speed: 11203 samples/sec	time cost: 118.853085
[Epoch 299] validation: err-top1=0.375820 err-top5=0.157793
Epoch[300] Batch [49]	Speed: 10529.872608 samples/sec	accuracy=0.620801	lr=0.176929
Epoch[300] Batch [99]	Speed: 11249.247469 samples/sec	accuracy=0.619165	lr=0.176466
Epoch[300] Batch [149]	Speed: 11027.871987 samples/sec	accuracy=0.618734	lr=0.176003
Epoch[300] Batch [199]	Speed: 11354.167537 samples/sec	accuracy=0.617039	lr=0.175541
Epoch[300] Batch [249]	Speed: 11291.719728 samples/sec	accuracy=0.616543	lr=0.175079
Epoch[300] Batch [299]	Speed: 11232.828081 samples/sec	accuracy=0.616061	lr=0.174618
Epoch[300] Batch [349]	Speed: 10826.316372 samples/sec	accuracy=0.615770	lr=0.174158
Epoch[300] Batch [399]	Speed: 11180.229393 samples/sec	accuracy=0.615284	lr=0.173698
Epoch[300] Batch [449]	Speed: 11322.371286 samples/sec	accuracy=0.615082	lr=0.173239
Epoch[300] Batch [499]	Speed: 10869.870878 samples/sec	accuracy=0.614677	lr=0.172780
Epoch[300] Batch [549]	Speed: 10970.339389 samples/sec	accuracy=0.614517	lr=0.172322
Epoch[300] Batch [599]	Speed: 11627.678667 samples/sec	accuracy=0.614360	lr=0.171864
[Epoch 300] training: accuracy=0.614319
[Epoch 300] speed: 11092 samples/sec	time cost: 119.953837
[Epoch 300] validation: err-top1=0.372945 err-top5=0.155029
Epoch[301] Batch [49]	Speed: 10476.355067 samples/sec	accuracy=0.623799	lr=0.171169
Epoch[301] Batch [99]	Speed: 11255.892802 samples/sec	accuracy=0.622192	lr=0.170713
Epoch[301] Batch [149]	Speed: 10827.037691 samples/sec	accuracy=0.620846	lr=0.170258
Epoch[301] Batch [199]	Speed: 11387.165943 samples/sec	accuracy=0.619917	lr=0.169803
Epoch[301] Batch [249]	Speed: 11350.266217 samples/sec	accuracy=0.618676	lr=0.169348
Epoch[301] Batch [299]	Speed: 10817.946765 samples/sec	accuracy=0.618071	lr=0.168894
Epoch[301] Batch [349]	Speed: 11740.112805 samples/sec	accuracy=0.617411	lr=0.168441
Epoch[301] Batch [399]	Speed: 11753.342814 samples/sec	accuracy=0.616842	lr=0.167988
Epoch[301] Batch [449]	Speed: 11093.890170 samples/sec	accuracy=0.616206	lr=0.167536
Epoch[301] Batch [499]	Speed: 9929.608357 samples/sec	accuracy=0.615688	lr=0.167084
Epoch[301] Batch [549]	Speed: 10997.402048 samples/sec	accuracy=0.615095	lr=0.166633
Epoch[301] Batch [599]	Speed: 11255.198745 samples/sec	accuracy=0.614599	lr=0.166182
[Epoch 301] training: accuracy=0.614539
[Epoch 301] speed: 11016 samples/sec	time cost: 121.047364
[Epoch 301] validation: err-top1=0.362090 err-top5=0.147090
Epoch[302] Batch [49]	Speed: 10321.821205 samples/sec	accuracy=0.621240	lr=0.165499
Epoch[302] Batch [99]	Speed: 11199.794475 samples/sec	accuracy=0.620005	lr=0.165050
Epoch[302] Batch [149]	Speed: 10913.661042 samples/sec	accuracy=0.619222	lr=0.164601
Epoch[302] Batch [199]	Speed: 11249.530623 samples/sec	accuracy=0.618384	lr=0.164153
Epoch[302] Batch [249]	Speed: 11353.317852 samples/sec	accuracy=0.618311	lr=0.163706
Epoch[302] Batch [299]	Speed: 11211.072897 samples/sec	accuracy=0.617404	lr=0.163259
Epoch[302] Batch [349]	Speed: 11455.188312 samples/sec	accuracy=0.617287	lr=0.162813
Epoch[302] Batch [399]	Speed: 10934.835487 samples/sec	accuracy=0.616984	lr=0.162367
Epoch[302] Batch [449]	Speed: 11780.259530 samples/sec	accuracy=0.616697	lr=0.161922
Epoch[302] Batch [499]	Speed: 11535.912162 samples/sec	accuracy=0.616449	lr=0.161477
Epoch[302] Batch [549]	Speed: 10840.608639 samples/sec	accuracy=0.616290	lr=0.161033
Epoch[302] Batch [599]	Speed: 11153.606419 samples/sec	accuracy=0.616065	lr=0.160590
[Epoch 302] training: accuracy=0.615968
[Epoch 302] speed: 11130 samples/sec	time cost: 119.418175
[Epoch 302] validation: err-top1=0.357218 err-top5=0.145203
Epoch[303] Batch [49]	Speed: 10744.695622 samples/sec	accuracy=0.624092	lr=0.159926
Epoch[303] Batch [99]	Speed: 10670.212175 samples/sec	accuracy=0.622207	lr=0.159484
Epoch[303] Batch [149]	Speed: 11246.746853 samples/sec	accuracy=0.621383	lr=0.159043
Epoch[303] Batch [199]	Speed: 11156.053904 samples/sec	accuracy=0.620220	lr=0.158602
Epoch[303] Batch [249]	Speed: 11153.522422 samples/sec	accuracy=0.620248	lr=0.158161
Epoch[303] Batch [299]	Speed: 11431.338800 samples/sec	accuracy=0.619582	lr=0.157722
Epoch[303] Batch [349]	Speed: 11239.098439 samples/sec	accuracy=0.619074	lr=0.157283
Epoch[303] Batch [399]	Speed: 11445.572400 samples/sec	accuracy=0.618645	lr=0.156844
Epoch[303] Batch [449]	Speed: 10851.075114 samples/sec	accuracy=0.618465	lr=0.156406
Epoch[303] Batch [499]	Speed: 10826.584365 samples/sec	accuracy=0.618022	lr=0.155969
Epoch[303] Batch [549]	Speed: 11227.321868 samples/sec	accuracy=0.617812	lr=0.155532
Epoch[303] Batch [599]	Speed: 11541.394989 samples/sec	accuracy=0.617536	lr=0.155096
[Epoch 303] training: accuracy=0.617404
[Epoch 303] speed: 11091 samples/sec	time cost: 120.163292
[Epoch 303] validation: err-top1=0.367106 err-top5=0.151103
Epoch[304] Batch [49]	Speed: 10246.414171 samples/sec	accuracy=0.624268	lr=0.154434
Epoch[304] Batch [99]	Speed: 11283.117472 samples/sec	accuracy=0.622700	lr=0.153999
Epoch[304] Batch [149]	Speed: 11723.597918 samples/sec	accuracy=0.622246	lr=0.153565
Epoch[304] Batch [199]	Speed: 11301.538787 samples/sec	accuracy=0.622251	lr=0.153131
Epoch[304] Batch [249]	Speed: 10953.993479 samples/sec	accuracy=0.621533	lr=0.152698
Epoch[304] Batch [299]	Speed: 11598.612617 samples/sec	accuracy=0.621077	lr=0.152266
Epoch[304] Batch [349]	Speed: 10874.770121 samples/sec	accuracy=0.620310	lr=0.151834
Epoch[304] Batch [399]	Speed: 11319.381308 samples/sec	accuracy=0.620305	lr=0.151402
Epoch[304] Batch [449]	Speed: 11476.894647 samples/sec	accuracy=0.620181	lr=0.150971
Epoch[304] Batch [499]	Speed: 11150.772932 samples/sec	accuracy=0.619744	lr=0.150541
Epoch[304] Batch [549]	Speed: 10824.960235 samples/sec	accuracy=0.619399	lr=0.150112
Epoch[304] Batch [599]	Speed: 11244.357151 samples/sec	accuracy=0.619090	lr=0.149683
[Epoch 304] training: accuracy=0.619082
[Epoch 304] speed: 11132 samples/sec	time cost: 119.694591
[Epoch 304] validation: err-top1=0.359629 err-top5=0.146328
Epoch[305] Batch [49]	Speed: 11171.704570 samples/sec	accuracy=0.627402	lr=0.149040
Epoch[305] Batch [99]	Speed: 11156.480180 samples/sec	accuracy=0.624526	lr=0.148613
Epoch[305] Batch [149]	Speed: 11269.859069 samples/sec	accuracy=0.623304	lr=0.148185
Epoch[305] Batch [199]	Speed: 11365.059023 samples/sec	accuracy=0.623291	lr=0.147759
Epoch[305] Batch [249]	Speed: 11196.819547 samples/sec	accuracy=0.622531	lr=0.147333
Epoch[305] Batch [299]	Speed: 11166.268845 samples/sec	accuracy=0.622121	lr=0.146908
Epoch[305] Batch [349]	Speed: 11322.788277 samples/sec	accuracy=0.621678	lr=0.146483
Epoch[305] Batch [399]	Speed: 11092.390258 samples/sec	accuracy=0.621443	lr=0.146059
Epoch[305] Batch [449]	Speed: 10973.330021 samples/sec	accuracy=0.620920	lr=0.145635
Epoch[305] Batch [499]	Speed: 11624.257235 samples/sec	accuracy=0.620514	lr=0.145212
Epoch[305] Batch [549]	Speed: 11175.205810 samples/sec	accuracy=0.620045	lr=0.144790
Epoch[305] Batch [599]	Speed: 11107.197677 samples/sec	accuracy=0.619707	lr=0.144368
[Epoch 305] training: accuracy=0.619460
[Epoch 305] speed: 11195 samples/sec	time cost: 118.834785
[Epoch 305] validation: err-top1=0.365214 err-top5=0.149597
Epoch[306] Batch [49]	Speed: 11292.513009 samples/sec	accuracy=0.623984	lr=0.143728
Epoch[306] Batch [99]	Speed: 11444.092375 samples/sec	accuracy=0.624829	lr=0.143308
Epoch[306] Batch [149]	Speed: 11759.111590 samples/sec	accuracy=0.623040	lr=0.142888
Epoch[306] Batch [199]	Speed: 10953.063244 samples/sec	accuracy=0.622144	lr=0.142469
Epoch[306] Batch [249]	Speed: 11495.318683 samples/sec	accuracy=0.622420	lr=0.142050
Epoch[306] Batch [299]	Speed: 11307.912608 samples/sec	accuracy=0.622171	lr=0.141632
Epoch[306] Batch [349]	Speed: 11387.592552 samples/sec	accuracy=0.621844	lr=0.141215
Epoch[306] Batch [399]	Speed: 11644.966365 samples/sec	accuracy=0.621794	lr=0.140798
Epoch[306] Batch [449]	Speed: 11194.717992 samples/sec	accuracy=0.621802	lr=0.140382
Epoch[306] Batch [499]	Speed: 10271.858085 samples/sec	accuracy=0.621805	lr=0.139966
Epoch[306] Batch [549]	Speed: 11442.759977 samples/sec	accuracy=0.621605	lr=0.139551
Epoch[306] Batch [599]	Speed: 11520.868088 samples/sec	accuracy=0.621311	lr=0.139136
[Epoch 306] training: accuracy=0.621119
[Epoch 306] speed: 11291 samples/sec	time cost: 117.915999
[Epoch 306] validation: err-top1=0.358086 err-top5=0.145664
Epoch[307] Batch [49]	Speed: 11437.516031 samples/sec	accuracy=0.625156	lr=0.138515
Epoch[307] Batch [99]	Speed: 10925.187248 samples/sec	accuracy=0.627354	lr=0.138102
Epoch[307] Batch [149]	Speed: 10948.971252 samples/sec	accuracy=0.626533	lr=0.137690
Epoch[307] Batch [199]	Speed: 11530.282244 samples/sec	accuracy=0.625793	lr=0.137278
Epoch[307] Batch [249]	Speed: 11403.674199 samples/sec	accuracy=0.624568	lr=0.136866
Epoch[307] Batch [299]	Speed: 11028.196208 samples/sec	accuracy=0.624295	lr=0.136456
Epoch[307] Batch [349]	Speed: 11692.117963 samples/sec	accuracy=0.623894	lr=0.136046
Epoch[307] Batch [399]	Speed: 10919.424018 samples/sec	accuracy=0.623439	lr=0.135636
Epoch[307] Batch [449]	Speed: 10947.761970 samples/sec	accuracy=0.623046	lr=0.135227
Epoch[307] Batch [499]	Speed: 11868.594388 samples/sec	accuracy=0.622839	lr=0.134818
Epoch[307] Batch [549]	Speed: 11586.255720 samples/sec	accuracy=0.622531	lr=0.134411
Epoch[307] Batch [599]	Speed: 11320.075546 samples/sec	accuracy=0.622263	lr=0.134003
[Epoch 307] training: accuracy=0.622097
[Epoch 307] speed: 11193 samples/sec	time cost: 118.946464
[Epoch 307] validation: err-top1=0.354411 err-top5=0.143555
Epoch[308] Batch [49]	Speed: 10970.543944 samples/sec	accuracy=0.626455	lr=0.133385
Epoch[308] Batch [99]	Speed: 11106.169155 samples/sec	accuracy=0.626479	lr=0.132980
Epoch[308] Batch [149]	Speed: 11135.201315 samples/sec	accuracy=0.626527	lr=0.132574
Epoch[308] Batch [199]	Speed: 11374.735574 samples/sec	accuracy=0.626008	lr=0.132170
Epoch[308] Batch [249]	Speed: 11520.356037 samples/sec	accuracy=0.625887	lr=0.131766
Epoch[308] Batch [299]	Speed: 11303.423613 samples/sec	accuracy=0.625983	lr=0.131362
Epoch[308] Batch [349]	Speed: 10828.495088 samples/sec	accuracy=0.625459	lr=0.130959
Epoch[308] Batch [399]	Speed: 11818.440016 samples/sec	accuracy=0.625255	lr=0.130557
Epoch[308] Batch [449]	Speed: 11070.048854 samples/sec	accuracy=0.624895	lr=0.130155
Epoch[308] Batch [499]	Speed: 11423.046026 samples/sec	accuracy=0.624513	lr=0.129754
Epoch[308] Batch [549]	Speed: 11287.983126 samples/sec	accuracy=0.624181	lr=0.129354
Epoch[308] Batch [599]	Speed: 11577.729250 samples/sec	accuracy=0.623921	lr=0.128954
[Epoch 308] training: accuracy=0.623670
[Epoch 308] speed: 11242 samples/sec	time cost: 118.521160
[Epoch 308] validation: err-top1=0.352926 err-top5=0.143005
Epoch[309] Batch [49]	Speed: 10222.293123 samples/sec	accuracy=0.628203	lr=0.128347
Epoch[309] Batch [99]	Speed: 11332.052690 samples/sec	accuracy=0.627764	lr=0.127949
Epoch[309] Batch [149]	Speed: 11444.921545 samples/sec	accuracy=0.627725	lr=0.127551
Epoch[309] Batch [199]	Speed: 11218.493902 samples/sec	accuracy=0.626897	lr=0.127154
Epoch[309] Batch [249]	Speed: 11257.788400 samples/sec	accuracy=0.626219	lr=0.126757
Epoch[309] Batch [299]	Speed: 11363.287070 samples/sec	accuracy=0.626115	lr=0.126361
Epoch[309] Batch [349]	Speed: 11463.576180 samples/sec	accuracy=0.625790	lr=0.125965
Epoch[309] Batch [399]	Speed: 11304.437224 samples/sec	accuracy=0.625359	lr=0.125570
Epoch[309] Batch [449]	Speed: 11477.135091 samples/sec	accuracy=0.625133	lr=0.125176
Epoch[309] Batch [499]	Speed: 11548.470499 samples/sec	accuracy=0.624842	lr=0.124782
Epoch[309] Batch [549]	Speed: 11357.423991 samples/sec	accuracy=0.624153	lr=0.124389
Epoch[309] Batch [599]	Speed: 10610.330093 samples/sec	accuracy=0.624344	lr=0.123996
[Epoch 309] training: accuracy=0.624216
[Epoch 309] speed: 11180 samples/sec	time cost: 119.267902
[Epoch 309] validation: err-top1=0.354121 err-top5=0.142754
Epoch[310] Batch [49]	Speed: 11059.534652 samples/sec	accuracy=0.630078	lr=0.123409
Epoch[310] Batch [99]	Speed: 11354.377652 samples/sec	accuracy=0.628276	lr=0.123018
Epoch[310] Batch [149]	Speed: 11645.844478 samples/sec	accuracy=0.628005	lr=0.122627
Epoch[310] Batch [199]	Speed: 11229.675854 samples/sec	accuracy=0.628398	lr=0.122237
Epoch[310] Batch [249]	Speed: 11481.493966 samples/sec	accuracy=0.627756	lr=0.121848
Epoch[310] Batch [299]	Speed: 11127.212563 samples/sec	accuracy=0.627467	lr=0.121459
Epoch[310] Batch [349]	Speed: 10643.272640 samples/sec	accuracy=0.627073	lr=0.121071
Epoch[310] Batch [399]	Speed: 11325.375386 samples/sec	accuracy=0.627020	lr=0.120683
Epoch[310] Batch [449]	Speed: 11469.963212 samples/sec	accuracy=0.626406	lr=0.120296
Epoch[310] Batch [499]	Speed: 11050.043303 samples/sec	accuracy=0.626116	lr=0.119910
Epoch[310] Batch [549]	Speed: 11409.336000 samples/sec	accuracy=0.625495	lr=0.119524
Epoch[310] Batch [599]	Speed: 11019.802045 samples/sec	accuracy=0.625123	lr=0.119139
[Epoch 310] training: accuracy=0.624998
[Epoch 310] speed: 11182 samples/sec	time cost: 119.045657
[Epoch 310] validation: err-top1=0.349467 err-top5=0.141215
Epoch[311] Batch [49]	Speed: 10564.935369 samples/sec	accuracy=0.632441	lr=0.118555
Epoch[311] Batch [99]	Speed: 11130.945298 samples/sec	accuracy=0.631982	lr=0.118171
Epoch[311] Batch [149]	Speed: 10314.644926 samples/sec	accuracy=0.630710	lr=0.117788
Epoch[311] Batch [199]	Speed: 11247.311449 samples/sec	accuracy=0.629902	lr=0.117405
Epoch[311] Batch [249]	Speed: 10822.027012 samples/sec	accuracy=0.629230	lr=0.117024
Epoch[311] Batch [299]	Speed: 11283.812605 samples/sec	accuracy=0.628908	lr=0.116642
Epoch[311] Batch [349]	Speed: 10952.956264 samples/sec	accuracy=0.628584	lr=0.116261
Epoch[311] Batch [399]	Speed: 10875.481664 samples/sec	accuracy=0.628177	lr=0.115881
Epoch[311] Batch [449]	Speed: 11240.683891 samples/sec	accuracy=0.628019	lr=0.115502
Epoch[311] Batch [499]	Speed: 11293.691262 samples/sec	accuracy=0.627685	lr=0.115123
Epoch[311] Batch [549]	Speed: 11028.527528 samples/sec	accuracy=0.627488	lr=0.114744
Epoch[311] Batch [599]	Speed: 11082.871340 samples/sec	accuracy=0.627459	lr=0.114367
[Epoch 311] training: accuracy=0.627280
[Epoch 311] speed: 10976 samples/sec	time cost: 121.404795
[Epoch 311] validation: err-top1=0.348164 err-top5=0.140859
Epoch[312] Batch [49]	Speed: 10686.954175 samples/sec	accuracy=0.631846	lr=0.113801
Epoch[312] Batch [99]	Speed: 11377.131897 samples/sec	accuracy=0.631050	lr=0.113425
Epoch[312] Batch [149]	Speed: 11159.141735 samples/sec	accuracy=0.630742	lr=0.113049
Epoch[312] Batch [199]	Speed: 11064.984093 samples/sec	accuracy=0.630488	lr=0.112674
Epoch[312] Batch [249]	Speed: 11490.294816 samples/sec	accuracy=0.629852	lr=0.112300
Epoch[312] Batch [299]	Speed: 11330.884356 samples/sec	accuracy=0.629277	lr=0.111926
Epoch[312] Batch [349]	Speed: 11220.121323 samples/sec	accuracy=0.628650	lr=0.111552
Epoch[312] Batch [399]	Speed: 10806.054516 samples/sec	accuracy=0.628304	lr=0.111180
Epoch[312] Batch [449]	Speed: 11199.470891 samples/sec	accuracy=0.628107	lr=0.110808
Epoch[312] Batch [499]	Speed: 10135.405556 samples/sec	accuracy=0.628168	lr=0.110436
Epoch[312] Batch [549]	Speed: 10725.640950 samples/sec	accuracy=0.628014	lr=0.110065
Epoch[312] Batch [599]	Speed: 10768.967276 samples/sec	accuracy=0.627721	lr=0.109695
[Epoch 312] training: accuracy=0.627578
[Epoch 312] speed: 10979 samples/sec	time cost: 120.998596
[Epoch 312] validation: err-top1=0.353923 err-top5=0.143107
Epoch[313] Batch [49]	Speed: 10452.910148 samples/sec	accuracy=0.634492	lr=0.109133
Epoch[313] Batch [99]	Speed: 11350.157635 samples/sec	accuracy=0.633276	lr=0.108764
Epoch[313] Batch [149]	Speed: 11518.455323 samples/sec	accuracy=0.631823	lr=0.108396
Epoch[313] Batch [199]	Speed: 11487.972886 samples/sec	accuracy=0.631624	lr=0.108028
Epoch[313] Batch [249]	Speed: 11385.285683 samples/sec	accuracy=0.631311	lr=0.107661
Epoch[313] Batch [299]	Speed: 11242.311871 samples/sec	accuracy=0.630907	lr=0.107295
Epoch[313] Batch [349]	Speed: 10944.516120 samples/sec	accuracy=0.630107	lr=0.106929
Epoch[313] Batch [399]	Speed: 11504.704056 samples/sec	accuracy=0.629719	lr=0.106564
Epoch[313] Batch [449]	Speed: 10924.645358 samples/sec	accuracy=0.629542	lr=0.106199
Epoch[313] Batch [499]	Speed: 11289.451831 samples/sec	accuracy=0.629321	lr=0.105835
Epoch[313] Batch [549]	Speed: 11188.446912 samples/sec	accuracy=0.629232	lr=0.105472
Epoch[313] Batch [599]	Speed: 11143.483433 samples/sec	accuracy=0.629041	lr=0.105109
[Epoch 313] training: accuracy=0.628985
[Epoch 313] speed: 11174 samples/sec	time cost: 119.036475
[Epoch 313] validation: err-top1=0.348906 err-top5=0.139551
Epoch[314] Batch [49]	Speed: 11209.381690 samples/sec	accuracy=0.634482	lr=0.104566
Epoch[314] Batch [99]	Speed: 11324.844134 samples/sec	accuracy=0.633276	lr=0.104204
Epoch[314] Batch [149]	Speed: 10809.658376 samples/sec	accuracy=0.632624	lr=0.103844
Epoch[314] Batch [199]	Speed: 11082.198456 samples/sec	accuracy=0.633291	lr=0.103484
Epoch[314] Batch [249]	Speed: 10990.216294 samples/sec	accuracy=0.632676	lr=0.103124
Epoch[314] Batch [299]	Speed: 11337.843205 samples/sec	accuracy=0.632572	lr=0.102765
Epoch[314] Batch [349]	Speed: 11448.386512 samples/sec	accuracy=0.632330	lr=0.102407
Epoch[314] Batch [399]	Speed: 11429.413205 samples/sec	accuracy=0.631896	lr=0.102049
Epoch[314] Batch [449]	Speed: 11501.372774 samples/sec	accuracy=0.631674	lr=0.101692
Epoch[314] Batch [499]	Speed: 10707.848249 samples/sec	accuracy=0.631482	lr=0.101335
Epoch[314] Batch [549]	Speed: 11218.448483 samples/sec	accuracy=0.630988	lr=0.100979
Epoch[314] Batch [599]	Speed: 11145.275125 samples/sec	accuracy=0.630717	lr=0.100624
[Epoch 314] training: accuracy=0.630561
[Epoch 314] speed: 11135 samples/sec	time cost: 119.511103
[Epoch 314] validation: err-top1=0.351420 err-top5=0.142090
Epoch[315] Batch [49]	Speed: 11449.219661 samples/sec	accuracy=0.634629	lr=0.100085
Epoch[315] Batch [99]	Speed: 11505.365736 samples/sec	accuracy=0.634961	lr=0.099731
Epoch[315] Batch [149]	Speed: 11239.583440 samples/sec	accuracy=0.634665	lr=0.099378
Epoch[315] Batch [199]	Speed: 11324.979406 samples/sec	accuracy=0.634194	lr=0.099025
Epoch[315] Batch [249]	Speed: 11551.292259 samples/sec	accuracy=0.633184	lr=0.098673
Epoch[315] Batch [299]	Speed: 11460.744506 samples/sec	accuracy=0.632811	lr=0.098322
Epoch[315] Batch [349]	Speed: 11520.878905 samples/sec	accuracy=0.632239	lr=0.097971
Epoch[315] Batch [399]	Speed: 11259.030253 samples/sec	accuracy=0.632195	lr=0.097621
Epoch[315] Batch [449]	Speed: 11637.364749 samples/sec	accuracy=0.631788	lr=0.097271
Epoch[315] Batch [499]	Speed: 10753.371135 samples/sec	accuracy=0.631472	lr=0.096922
Epoch[315] Batch [549]	Speed: 11380.433201 samples/sec	accuracy=0.631322	lr=0.096574
Epoch[315] Batch [599]	Speed: 11009.280025 samples/sec	accuracy=0.631178	lr=0.096226
[Epoch 315] training: accuracy=0.631085
[Epoch 315] speed: 11294 samples/sec	time cost: 117.956525
[Epoch 315] validation: err-top1=0.348938 err-top5=0.139913
Epoch[316] Batch [49]	Speed: 11128.861186 samples/sec	accuracy=0.634600	lr=0.095698
Epoch[316] Batch [99]	Speed: 11116.442806 samples/sec	accuracy=0.634736	lr=0.095352
Epoch[316] Batch [149]	Speed: 11355.565248 samples/sec	accuracy=0.634915	lr=0.095007
Epoch[316] Batch [199]	Speed: 11210.073619 samples/sec	accuracy=0.635210	lr=0.094661
Epoch[316] Batch [249]	Speed: 11340.932483 samples/sec	accuracy=0.634881	lr=0.094317
Epoch[316] Batch [299]	Speed: 11371.328889 samples/sec	accuracy=0.634657	lr=0.093973
Epoch[316] Batch [349]	Speed: 11151.121213 samples/sec	accuracy=0.634174	lr=0.093630
Epoch[316] Batch [399]	Speed: 10920.840028 samples/sec	accuracy=0.633818	lr=0.093287
Epoch[316] Batch [449]	Speed: 10618.290314 samples/sec	accuracy=0.633456	lr=0.092945
Epoch[316] Batch [499]	Speed: 10927.911962 samples/sec	accuracy=0.633099	lr=0.092604
Epoch[316] Batch [549]	Speed: 11176.998992 samples/sec	accuracy=0.632615	lr=0.092263
Epoch[316] Batch [599]	Speed: 11573.142958 samples/sec	accuracy=0.632452	lr=0.091923
[Epoch 316] training: accuracy=0.632410
[Epoch 316] speed: 11135 samples/sec	time cost: 119.557078
[Epoch 316] validation: err-top1=0.352773 err-top5=0.142969
Epoch[317] Batch [49]	Speed: 10676.230892 samples/sec	accuracy=0.639258	lr=0.091413
Epoch[317] Batch [99]	Speed: 11127.866995 samples/sec	accuracy=0.638057	lr=0.091075
Epoch[317] Batch [149]	Speed: 11191.377732 samples/sec	accuracy=0.637985	lr=0.090737
Epoch[317] Batch [199]	Speed: 11691.260864 samples/sec	accuracy=0.636924	lr=0.090399
Epoch[317] Batch [249]	Speed: 10979.498898 samples/sec	accuracy=0.636344	lr=0.090062
Epoch[317] Batch [299]	Speed: 10209.654889 samples/sec	accuracy=0.635840	lr=0.089726
Epoch[317] Batch [349]	Speed: 11453.947715 samples/sec	accuracy=0.635412	lr=0.089390
Epoch[317] Batch [399]	Speed: 11754.624669 samples/sec	accuracy=0.635232	lr=0.089055
Epoch[317] Batch [449]	Speed: 11140.914304 samples/sec	accuracy=0.635199	lr=0.088721
Epoch[317] Batch [499]	Speed: 11433.689939 samples/sec	accuracy=0.634764	lr=0.088387
Epoch[317] Batch [549]	Speed: 11509.320453 samples/sec	accuracy=0.634400	lr=0.088053
Epoch[317] Batch [599]	Speed: 11288.224324 samples/sec	accuracy=0.634134	lr=0.087721
[Epoch 317] training: accuracy=0.634085
[Epoch 317] speed: 11166 samples/sec	time cost: 119.202431
[Epoch 317] validation: err-top1=0.342590 err-top5=0.136963
Epoch[318] Batch [49]	Speed: 11086.435012 samples/sec	accuracy=0.640146	lr=0.087216
Epoch[318] Batch [99]	Speed: 10785.185557 samples/sec	accuracy=0.638481	lr=0.086885
Epoch[318] Batch [149]	Speed: 11262.373519 samples/sec	accuracy=0.637949	lr=0.086555
Epoch[318] Batch [199]	Speed: 10981.907064 samples/sec	accuracy=0.636909	lr=0.086225
Epoch[318] Batch [249]	Speed: 11659.061831 samples/sec	accuracy=0.636885	lr=0.085895
Epoch[318] Batch [299]	Speed: 11402.644531 samples/sec	accuracy=0.636273	lr=0.085567
Epoch[318] Batch [349]	Speed: 11332.199795 samples/sec	accuracy=0.636020	lr=0.085239
Epoch[318] Batch [399]	Speed: 11313.878151 samples/sec	accuracy=0.635336	lr=0.084911
Epoch[318] Batch [449]	Speed: 11322.193096 samples/sec	accuracy=0.635443	lr=0.084584
Epoch[318] Batch [499]	Speed: 11394.014264 samples/sec	accuracy=0.635038	lr=0.084258
Epoch[318] Batch [549]	Speed: 11535.795041 samples/sec	accuracy=0.634674	lr=0.083932
Epoch[318] Batch [599]	Speed: 11305.235864 samples/sec	accuracy=0.634531	lr=0.083607
[Epoch 318] training: accuracy=0.634448
[Epoch 318] speed: 11242 samples/sec	time cost: 118.422402
[Epoch 318] validation: err-top1=0.347246 err-top5=0.140234
Epoch[319] Batch [49]	Speed: 11320.994268 samples/sec	accuracy=0.640869	lr=0.083121
Epoch[319] Batch [99]	Speed: 11009.750191 samples/sec	accuracy=0.639839	lr=0.082797
Epoch[319] Batch [149]	Speed: 11630.059625 samples/sec	accuracy=0.638994	lr=0.082475
Epoch[319] Batch [199]	Speed: 11303.649704 samples/sec	accuracy=0.638533	lr=0.082152
Epoch[319] Batch [249]	Speed: 11197.833689 samples/sec	accuracy=0.637967	lr=0.081831
Epoch[319] Batch [299]	Speed: 11298.906376 samples/sec	accuracy=0.637170	lr=0.081510
Epoch[319] Batch [349]	Speed: 10930.233294 samples/sec	accuracy=0.637280	lr=0.081189
Epoch[319] Batch [399]	Speed: 10238.869510 samples/sec	accuracy=0.636862	lr=0.080869
Epoch[319] Batch [449]	Speed: 11169.231044 samples/sec	accuracy=0.636470	lr=0.080550
Epoch[319] Batch [499]	Speed: 11512.133294 samples/sec	accuracy=0.636406	lr=0.080231
Epoch[319] Batch [549]	Speed: 11320.349446 samples/sec	accuracy=0.636233	lr=0.079913
Epoch[319] Batch [599]	Speed: 11317.979964 samples/sec	accuracy=0.636015	lr=0.079596
[Epoch 319] training: accuracy=0.636071
[Epoch 319] speed: 11161 samples/sec	time cost: 119.243648
[Epoch 319] validation: err-top1=0.345174 err-top5=0.137655
Epoch[320] Batch [49]	Speed: 9907.285546 samples/sec	accuracy=0.638359	lr=0.079115
Epoch[320] Batch [99]	Speed: 11366.695255 samples/sec	accuracy=0.639287	lr=0.078799
Epoch[320] Batch [149]	Speed: 11345.659320 samples/sec	accuracy=0.639746	lr=0.078484
Epoch[320] Batch [199]	Speed: 11337.090226 samples/sec	accuracy=0.639204	lr=0.078169
Epoch[320] Batch [249]	Speed: 10570.074928 samples/sec	accuracy=0.639303	lr=0.077855
Epoch[320] Batch [299]	Speed: 11569.486762 samples/sec	accuracy=0.638823	lr=0.077541
Epoch[320] Batch [349]	Speed: 11460.218825 samples/sec	accuracy=0.638350	lr=0.077229
Epoch[320] Batch [399]	Speed: 10408.025315 samples/sec	accuracy=0.638079	lr=0.076916
Epoch[320] Batch [449]	Speed: 11392.826470 samples/sec	accuracy=0.638003	lr=0.076605
Epoch[320] Batch [499]	Speed: 10292.749339 samples/sec	accuracy=0.637766	lr=0.076294
Epoch[320] Batch [549]	Speed: 11051.629324 samples/sec	accuracy=0.637686	lr=0.075984
Epoch[320] Batch [599]	Speed: 10695.516682 samples/sec	accuracy=0.637485	lr=0.075674
[Epoch 320] training: accuracy=0.637260
[Epoch 320] speed: 10882 samples/sec	time cost: 122.232774
[Epoch 320] validation: err-top1=0.345011 err-top5=0.139669
Epoch[321] Batch [49]	Speed: 11053.761416 samples/sec	accuracy=0.641670	lr=0.075210
Epoch[321] Batch [99]	Speed: 10810.508357 samples/sec	accuracy=0.642329	lr=0.074902
Epoch[321] Batch [149]	Speed: 10428.657499 samples/sec	accuracy=0.641540	lr=0.074595
Epoch[321] Batch [199]	Speed: 11357.854681 samples/sec	accuracy=0.641228	lr=0.074288
Epoch[321] Batch [249]	Speed: 11095.948302 samples/sec	accuracy=0.640568	lr=0.073981
Epoch[321] Batch [299]	Speed: 11053.615762 samples/sec	accuracy=0.640013	lr=0.073676
Epoch[321] Batch [349]	Speed: 10896.469925 samples/sec	accuracy=0.639947	lr=0.073370
Epoch[321] Batch [399]	Speed: 11189.916357 samples/sec	accuracy=0.639570	lr=0.073066
Epoch[321] Batch [449]	Speed: 11342.559079 samples/sec	accuracy=0.639317	lr=0.072762
Epoch[321] Batch [499]	Speed: 10932.598745 samples/sec	accuracy=0.639143	lr=0.072459
Epoch[321] Batch [549]	Speed: 11591.137071 samples/sec	accuracy=0.638726	lr=0.072156
Epoch[321] Batch [599]	Speed: 10836.450393 samples/sec	accuracy=0.638598	lr=0.071854
[Epoch 321] training: accuracy=0.638674
[Epoch 321] speed: 11022 samples/sec	time cost: 120.948632
[Epoch 321] validation: err-top1=0.346836 err-top5=0.139453
Epoch[322] Batch [49]	Speed: 11169.322249 samples/sec	accuracy=0.643555	lr=0.071396
Epoch[322] Batch [99]	Speed: 11523.427484 samples/sec	accuracy=0.641929	lr=0.071096
Epoch[322] Batch [149]	Speed: 11269.613333 samples/sec	accuracy=0.642301	lr=0.070796
Epoch[322] Batch [199]	Speed: 10977.329702 samples/sec	accuracy=0.642139	lr=0.070496
Epoch[322] Batch [249]	Speed: 10573.635273 samples/sec	accuracy=0.641729	lr=0.070198
Epoch[322] Batch [299]	Speed: 11103.883310 samples/sec	accuracy=0.641009	lr=0.069900
Epoch[322] Batch [349]	Speed: 11531.748423 samples/sec	accuracy=0.640338	lr=0.069602
Epoch[322] Batch [399]	Speed: 11241.959931 samples/sec	accuracy=0.640061	lr=0.069306
Epoch[322] Batch [449]	Speed: 11320.194593 samples/sec	accuracy=0.639834	lr=0.069009
Epoch[322] Batch [499]	Speed: 11563.909343 samples/sec	accuracy=0.639950	lr=0.068714
Epoch[322] Batch [549]	Speed: 11279.356643 samples/sec	accuracy=0.639496	lr=0.068419
Epoch[322] Batch [599]	Speed: 11478.657115 samples/sec	accuracy=0.639385	lr=0.068124
[Epoch 322] training: accuracy=0.639212
[Epoch 322] speed: 11198 samples/sec	time cost: 118.941724
[Epoch 322] validation: err-top1=0.345459 err-top5=0.138713
Epoch[323] Batch [49]	Speed: 11269.333012 samples/sec	accuracy=0.643252	lr=0.067678
Epoch[323] Batch [99]	Speed: 11283.522684 samples/sec	accuracy=0.644028	lr=0.067385
Epoch[323] Batch [149]	Speed: 11315.658577 samples/sec	accuracy=0.643210	lr=0.067093
Epoch[323] Batch [199]	Speed: 11524.354772 samples/sec	accuracy=0.642715	lr=0.066802
Epoch[323] Batch [249]	Speed: 10622.501365 samples/sec	accuracy=0.642436	lr=0.066511
Epoch[323] Batch [299]	Speed: 9898.254836 samples/sec	accuracy=0.641794	lr=0.066221
Epoch[323] Batch [349]	Speed: 10422.966715 samples/sec	accuracy=0.641751	lr=0.065931
Epoch[323] Batch [399]	Speed: 11043.687866 samples/sec	accuracy=0.641472	lr=0.065642
Epoch[323] Batch [449]	Speed: 11126.639783 samples/sec	accuracy=0.641018	lr=0.065353
Epoch[323] Batch [499]	Speed: 11326.897451 samples/sec	accuracy=0.640877	lr=0.065066
Epoch[323] Batch [549]	Speed: 11664.671878 samples/sec	accuracy=0.640596	lr=0.064778
Epoch[323] Batch [599]	Speed: 10844.772815 samples/sec	accuracy=0.640309	lr=0.064492
[Epoch 323] training: accuracy=0.640340
[Epoch 323] speed: 11001 samples/sec	time cost: 120.873407
[Epoch 323] validation: err-top1=0.340215 err-top5=0.135684
Epoch[324] Batch [49]	Speed: 10812.822267 samples/sec	accuracy=0.646133	lr=0.064063
Epoch[324] Batch [99]	Speed: 11094.211121 samples/sec	accuracy=0.646387	lr=0.063778
Epoch[324] Batch [149]	Speed: 11091.141929 samples/sec	accuracy=0.645221	lr=0.063493
Epoch[324] Batch [199]	Speed: 11139.002101 samples/sec	accuracy=0.644167	lr=0.063210
Epoch[324] Batch [249]	Speed: 11330.647310 samples/sec	accuracy=0.643865	lr=0.062927
Epoch[324] Batch [299]	Speed: 11531.899211 samples/sec	accuracy=0.643301	lr=0.062644
Epoch[324] Batch [349]	Speed: 11413.160971 samples/sec	accuracy=0.643101	lr=0.062362
Epoch[324] Batch [399]	Speed: 11618.941838 samples/sec	accuracy=0.642401	lr=0.062081
Epoch[324] Batch [449]	Speed: 11383.148090 samples/sec	accuracy=0.642192	lr=0.061800
Epoch[324] Batch [499]	Speed: 11004.241100 samples/sec	accuracy=0.642051	lr=0.061520
Epoch[324] Batch [549]	Speed: 11383.689654 samples/sec	accuracy=0.641786	lr=0.061240
Epoch[324] Batch [599]	Speed: 11746.866731 samples/sec	accuracy=0.641691	lr=0.060962
[Epoch 324] training: accuracy=0.641537
[Epoch 324] speed: 11253 samples/sec	time cost: 118.295020
[Epoch 324] validation: err-top1=0.340332 err-top5=0.133565
Epoch[325] Batch [49]	Speed: 11292.466989 samples/sec	accuracy=0.647920	lr=0.060539
Epoch[325] Batch [99]	Speed: 11073.543186 samples/sec	accuracy=0.645786	lr=0.060262
Epoch[325] Batch [149]	Speed: 11712.879430 samples/sec	accuracy=0.645182	lr=0.059985
Epoch[325] Batch [199]	Speed: 11431.646104 samples/sec	accuracy=0.645115	lr=0.059709
Epoch[325] Batch [249]	Speed: 10671.028437 samples/sec	accuracy=0.645000	lr=0.059434
Epoch[325] Batch [299]	Speed: 11536.964800 samples/sec	accuracy=0.644295	lr=0.059159
Epoch[325] Batch [349]	Speed: 11123.912748 samples/sec	accuracy=0.643970	lr=0.058885
Epoch[325] Batch [399]	Speed: 11321.385791 samples/sec	accuracy=0.643789	lr=0.058611
Epoch[325] Batch [449]	Speed: 10303.346001 samples/sec	accuracy=0.643517	lr=0.058338
Epoch[325] Batch [499]	Speed: 11156.525388 samples/sec	accuracy=0.643176	lr=0.058066
Epoch[325] Batch [549]	Speed: 11464.727665 samples/sec	accuracy=0.643374	lr=0.057794
Epoch[325] Batch [599]	Speed: 11294.712633 samples/sec	accuracy=0.643245	lr=0.057523
[Epoch 325] training: accuracy=0.643176
[Epoch 325] speed: 11136 samples/sec	time cost: 119.263533
[Epoch 325] validation: err-top1=0.337606 err-top5=0.134399
Epoch[326] Batch [49]	Speed: 10639.553499 samples/sec	accuracy=0.648418	lr=0.057118
Epoch[326] Batch [99]	Speed: 11487.952606 samples/sec	accuracy=0.647007	lr=0.056848
Epoch[326] Batch [149]	Speed: 10820.022348 samples/sec	accuracy=0.646387	lr=0.056579
Epoch[326] Batch [199]	Speed: 10690.905850 samples/sec	accuracy=0.646685	lr=0.056311
Epoch[326] Batch [249]	Speed: 11111.188338 samples/sec	accuracy=0.646012	lr=0.056043
Epoch[326] Batch [299]	Speed: 11821.548190 samples/sec	accuracy=0.645882	lr=0.055776
Epoch[326] Batch [349]	Speed: 11221.311489 samples/sec	accuracy=0.645476	lr=0.055510
Epoch[326] Batch [399]	Speed: 10104.220585 samples/sec	accuracy=0.645242	lr=0.055244
Epoch[326] Batch [449]	Speed: 11077.507410 samples/sec	accuracy=0.645329	lr=0.054979
Epoch[326] Batch [499]	Speed: 11626.520970 samples/sec	accuracy=0.644963	lr=0.054715
Epoch[326] Batch [549]	Speed: 11120.716245 samples/sec	accuracy=0.644648	lr=0.054451
Epoch[326] Batch [599]	Speed: 11508.847051 samples/sec	accuracy=0.644456	lr=0.054187
[Epoch 326] training: accuracy=0.644316
[Epoch 326] speed: 11062 samples/sec	time cost: 120.451903
[Epoch 326] validation: err-top1=0.339961 err-top5=0.135547
Epoch[327] Batch [49]	Speed: 11234.923399 samples/sec	accuracy=0.648223	lr=0.053788
Epoch[327] Batch [99]	Speed: 11365.665637 samples/sec	accuracy=0.647437	lr=0.053527
Epoch[327] Batch [149]	Speed: 11218.333911 samples/sec	accuracy=0.646885	lr=0.053266
Epoch[327] Batch [199]	Speed: 11371.042582 samples/sec	accuracy=0.647122	lr=0.053005
Epoch[327] Batch [249]	Speed: 10959.825989 samples/sec	accuracy=0.646469	lr=0.052745
Epoch[327] Batch [299]	Speed: 11306.077177 samples/sec	accuracy=0.646058	lr=0.052486
Epoch[327] Batch [349]	Speed: 10132.312724 samples/sec	accuracy=0.645879	lr=0.052228
Epoch[327] Batch [399]	Speed: 10861.292091 samples/sec	accuracy=0.645802	lr=0.051970
Epoch[327] Batch [449]	Speed: 11426.513262 samples/sec	accuracy=0.645367	lr=0.051712
Epoch[327] Batch [499]	Speed: 11089.550561 samples/sec	accuracy=0.645375	lr=0.051456
Epoch[327] Batch [549]	Speed: 11024.128822 samples/sec	accuracy=0.645277	lr=0.051200
Epoch[327] Batch [599]	Speed: 11638.264107 samples/sec	accuracy=0.645129	lr=0.050944
[Epoch 327] training: accuracy=0.644927
[Epoch 327] speed: 11103 samples/sec	time cost: 119.649743
[Epoch 327] validation: err-top1=0.339966 err-top5=0.134298
Epoch[328] Batch [49]	Speed: 9745.716761 samples/sec	accuracy=0.648779	lr=0.050562
Epoch[328] Batch [99]	Speed: 11510.405877 samples/sec	accuracy=0.649248	lr=0.050308
Epoch[328] Batch [149]	Speed: 11065.005758 samples/sec	accuracy=0.648066	lr=0.050055
Epoch[328] Batch [199]	Speed: 10752.268198 samples/sec	accuracy=0.647529	lr=0.049803
Epoch[328] Batch [249]	Speed: 10645.090443 samples/sec	accuracy=0.647527	lr=0.049551
Epoch[328] Batch [299]	Speed: 11245.493281 samples/sec	accuracy=0.647184	lr=0.049299
Epoch[328] Batch [349]	Speed: 11395.713269 samples/sec	accuracy=0.646830	lr=0.049048
Epoch[328] Batch [399]	Speed: 10709.118057 samples/sec	accuracy=0.646660	lr=0.048798
Epoch[328] Batch [449]	Speed: 11412.481954 samples/sec	accuracy=0.646396	lr=0.048549
Epoch[328] Batch [499]	Speed: 11365.026243 samples/sec	accuracy=0.646312	lr=0.048300
Epoch[328] Batch [549]	Speed: 11156.109541 samples/sec	accuracy=0.646307	lr=0.048052
Epoch[328] Batch [599]	Speed: 10459.911302 samples/sec	accuracy=0.646220	lr=0.047804
[Epoch 328] training: accuracy=0.646069
[Epoch 328] speed: 10905 samples/sec	time cost: 122.263293
[Epoch 328] validation: err-top1=0.343555 err-top5=0.137676
Epoch[329] Batch [49]	Speed: 11312.665886 samples/sec	accuracy=0.650068	lr=0.047429
Epoch[329] Batch [99]	Speed: 11333.714723 samples/sec	accuracy=0.649463	lr=0.047183
Epoch[329] Batch [149]	Speed: 11372.673912 samples/sec	accuracy=0.649339	lr=0.046938
Epoch[329] Batch [199]	Speed: 11415.329269 samples/sec	accuracy=0.648713	lr=0.046693
Epoch[329] Batch [249]	Speed: 10884.065513 samples/sec	accuracy=0.648334	lr=0.046449
Epoch[329] Batch [299]	Speed: 11576.293167 samples/sec	accuracy=0.648237	lr=0.046205
Epoch[329] Batch [349]	Speed: 10452.905823 samples/sec	accuracy=0.648209	lr=0.045962
Epoch[329] Batch [399]	Speed: 11286.829793 samples/sec	accuracy=0.647878	lr=0.045720
Epoch[329] Batch [449]	Speed: 9998.900685 samples/sec	accuracy=0.647988	lr=0.045478
Epoch[329] Batch [499]	Speed: 10853.314271 samples/sec	accuracy=0.647406	lr=0.045237
Epoch[329] Batch [549]	Speed: 10927.811867 samples/sec	accuracy=0.647165	lr=0.044997
Epoch[329] Batch [599]	Speed: 11374.290951 samples/sec	accuracy=0.647136	lr=0.044757
[Epoch 329] training: accuracy=0.647087
[Epoch 329] speed: 11009 samples/sec	time cost: 120.744901
[Epoch 329] validation: err-top1=0.335002 err-top5=0.132121
Epoch[330] Batch [49]	Speed: 11432.511814 samples/sec	accuracy=0.651875	lr=0.044394
Epoch[330] Batch [99]	Speed: 11711.728659 samples/sec	accuracy=0.650195	lr=0.044156
Epoch[330] Batch [149]	Speed: 11701.689849 samples/sec	accuracy=0.649567	lr=0.043918
Epoch[330] Batch [199]	Speed: 10276.282399 samples/sec	accuracy=0.649570	lr=0.043682
Epoch[330] Batch [249]	Speed: 11235.670509 samples/sec	accuracy=0.649568	lr=0.043445
Epoch[330] Batch [299]	Speed: 11308.979134 samples/sec	accuracy=0.649523	lr=0.043210
Epoch[330] Batch [349]	Speed: 11206.209566 samples/sec	accuracy=0.649443	lr=0.042975
Epoch[330] Batch [399]	Speed: 11419.556907 samples/sec	accuracy=0.649032	lr=0.042740
Epoch[330] Batch [449]	Speed: 10914.822027 samples/sec	accuracy=0.648974	lr=0.042507
Epoch[330] Batch [499]	Speed: 11048.566023 samples/sec	accuracy=0.648843	lr=0.042273
Epoch[330] Batch [549]	Speed: 11504.250447 samples/sec	accuracy=0.648601	lr=0.042041
Epoch[330] Batch [599]	Speed: 10911.852661 samples/sec	accuracy=0.648599	lr=0.041809
[Epoch 330] training: accuracy=0.648623
[Epoch 330] speed: 11185 samples/sec	time cost: 119.026364
[Epoch 330] validation: err-top1=0.338320 err-top5=0.134531
Epoch[331] Batch [49]	Speed: 10785.850484 samples/sec	accuracy=0.652148	lr=0.041463
Epoch[331] Batch [99]	Speed: 10922.266959 samples/sec	accuracy=0.651895	lr=0.041232
Epoch[331] Batch [149]	Speed: 11282.135242 samples/sec	accuracy=0.651374	lr=0.041003
Epoch[331] Batch [199]	Speed: 11028.197341 samples/sec	accuracy=0.651763	lr=0.040774
Epoch[331] Batch [249]	Speed: 10372.151766 samples/sec	accuracy=0.651363	lr=0.040545
Epoch[331] Batch [299]	Speed: 11278.200037 samples/sec	accuracy=0.651182	lr=0.040317
Epoch[331] Batch [349]	Speed: 11569.481463 samples/sec	accuracy=0.651057	lr=0.040090
Epoch[331] Batch [399]	Speed: 11307.459202 samples/sec	accuracy=0.650858	lr=0.039864
Epoch[331] Batch [449]	Speed: 10880.753397 samples/sec	accuracy=0.650602	lr=0.039638
Epoch[331] Batch [499]	Speed: 11508.894852 samples/sec	accuracy=0.650153	lr=0.039413
Epoch[331] Batch [549]	Speed: 11703.881786 samples/sec	accuracy=0.650041	lr=0.039188
Epoch[331] Batch [599]	Speed: 11174.779265 samples/sec	accuracy=0.649860	lr=0.038964
[Epoch 331] training: accuracy=0.649749
[Epoch 331] speed: 11087 samples/sec	time cost: 120.083104
[Epoch 331] validation: err-top1=0.332662 err-top5=0.130798
Epoch[332] Batch [49]	Speed: 11091.880924 samples/sec	accuracy=0.653564	lr=0.038625
Epoch[332] Batch [99]	Speed: 11557.634990 samples/sec	accuracy=0.654238	lr=0.038403
Epoch[332] Batch [149]	Speed: 11573.138592 samples/sec	accuracy=0.653076	lr=0.038181
Epoch[332] Batch [199]	Speed: 11436.464712 samples/sec	accuracy=0.652056	lr=0.037960
Epoch[332] Batch [249]	Speed: 11671.703689 samples/sec	accuracy=0.651441	lr=0.037739
Epoch[332] Batch [299]	Speed: 10515.077453 samples/sec	accuracy=0.651024	lr=0.037520
Epoch[332] Batch [349]	Speed: 10625.563465 samples/sec	accuracy=0.650612	lr=0.037300
Epoch[332] Batch [399]	Speed: 11296.682538 samples/sec	accuracy=0.650725	lr=0.037082
Epoch[332] Batch [449]	Speed: 11138.953856 samples/sec	accuracy=0.650672	lr=0.036864
Epoch[332] Batch [499]	Speed: 10246.053381 samples/sec	accuracy=0.650616	lr=0.036647
Epoch[332] Batch [549]	Speed: 11356.247717 samples/sec	accuracy=0.650597	lr=0.036430
Epoch[332] Batch [599]	Speed: 11278.683383 samples/sec	accuracy=0.650295	lr=0.036214
[Epoch 332] training: accuracy=0.650213
[Epoch 332] speed: 11110 samples/sec	time cost: 119.746468
[Epoch 332] validation: err-top1=0.333333 err-top5=0.131551
Epoch[333] Batch [49]	Speed: 11269.311722 samples/sec	accuracy=0.654492	lr=0.035891
Epoch[333] Batch [99]	Speed: 10906.575130 samples/sec	accuracy=0.654121	lr=0.035677
Epoch[333] Batch [149]	Speed: 11453.112656 samples/sec	accuracy=0.653590	lr=0.035463
Epoch[333] Batch [199]	Speed: 10568.676375 samples/sec	accuracy=0.652717	lr=0.035250
Epoch[333] Batch [249]	Speed: 11477.781027 samples/sec	accuracy=0.653119	lr=0.035037
Epoch[333] Batch [299]	Speed: 11378.990171 samples/sec	accuracy=0.652552	lr=0.034825
Epoch[333] Batch [349]	Speed: 10754.323230 samples/sec	accuracy=0.652708	lr=0.034614
Epoch[333] Batch [399]	Speed: 11295.511385 samples/sec	accuracy=0.652273	lr=0.034403
Epoch[333] Batch [449]	Speed: 11317.839789 samples/sec	accuracy=0.651950	lr=0.034193
Epoch[333] Batch [499]	Speed: 10941.028873 samples/sec	accuracy=0.651808	lr=0.033984
Epoch[333] Batch [549]	Speed: 11435.437639 samples/sec	accuracy=0.651839	lr=0.033775
Epoch[333] Batch [599]	Speed: 10792.469819 samples/sec	accuracy=0.651750	lr=0.033567
[Epoch 333] training: accuracy=0.651555
[Epoch 333] speed: 11119 samples/sec	time cost: 119.961435
[Epoch 333] validation: err-top1=0.332227 err-top5=0.130313
Epoch[334] Batch [49]	Speed: 10469.624866 samples/sec	accuracy=0.655127	lr=0.033252
Epoch[334] Batch [99]	Speed: 11179.129689 samples/sec	accuracy=0.655635	lr=0.033045
Epoch[334] Batch [149]	Speed: 11569.566545 samples/sec	accuracy=0.654613	lr=0.032839
Epoch[334] Batch [199]	Speed: 11751.997892 samples/sec	accuracy=0.654333	lr=0.032634
Epoch[334] Batch [249]	Speed: 11151.757612 samples/sec	accuracy=0.653943	lr=0.032429
Epoch[334] Batch [299]	Speed: 11234.211943 samples/sec	accuracy=0.653739	lr=0.032226
Epoch[334] Batch [349]	Speed: 11399.161509 samples/sec	accuracy=0.653357	lr=0.032022
Epoch[334] Batch [399]	Speed: 11374.376801 samples/sec	accuracy=0.653322	lr=0.031819
Epoch[334] Batch [449]	Speed: 10566.508401 samples/sec	accuracy=0.653287	lr=0.031617
Epoch[334] Batch [499]	Speed: 11431.193065 samples/sec	accuracy=0.653144	lr=0.031416
Epoch[334] Batch [549]	Speed: 10866.874226 samples/sec	accuracy=0.653095	lr=0.031215
Epoch[334] Batch [599]	Speed: 10902.984420 samples/sec	accuracy=0.653018	lr=0.031015
[Epoch 334] training: accuracy=0.653036
[Epoch 334] speed: 11099 samples/sec	time cost: 119.661546
[Epoch 334] validation: err-top1=0.333130 err-top5=0.131856
Epoch[335] Batch [49]	Speed: 10610.747927 samples/sec	accuracy=0.657197	lr=0.030716
Epoch[335] Batch [99]	Speed: 11722.232919 samples/sec	accuracy=0.655874	lr=0.030517
Epoch[335] Batch [149]	Speed: 11201.178681 samples/sec	accuracy=0.655550	lr=0.030320
Epoch[335] Batch [199]	Speed: 11308.736752 samples/sec	accuracy=0.655017	lr=0.030122
Epoch[335] Batch [249]	Speed: 10958.605081 samples/sec	accuracy=0.654930	lr=0.029926
Epoch[335] Batch [299]	Speed: 11364.752583 samples/sec	accuracy=0.654609	lr=0.029730
Epoch[335] Batch [349]	Speed: 11347.062431 samples/sec	accuracy=0.654294	lr=0.029534
Epoch[335] Batch [399]	Speed: 11137.572278 samples/sec	accuracy=0.653971	lr=0.029339
Epoch[335] Batch [449]	Speed: 11073.903504 samples/sec	accuracy=0.654071	lr=0.029145
Epoch[335] Batch [499]	Speed: 10256.893057 samples/sec	accuracy=0.653898	lr=0.028952
Epoch[335] Batch [549]	Speed: 10257.761222 samples/sec	accuracy=0.653862	lr=0.028759
Epoch[335] Batch [599]	Speed: 11007.984030 samples/sec	accuracy=0.653929	lr=0.028567
[Epoch 335] training: accuracy=0.653793
[Epoch 335] speed: 10998 samples/sec	time cost: 121.057656
[Epoch 335] validation: err-top1=0.334629 err-top5=0.130078
Epoch[336] Batch [49]	Speed: 11070.586146 samples/sec	accuracy=0.656680	lr=0.028276
Epoch[336] Batch [99]	Speed: 11516.413806 samples/sec	accuracy=0.655249	lr=0.028085
Epoch[336] Batch [149]	Speed: 11343.822102 samples/sec	accuracy=0.655114	lr=0.027895
Epoch[336] Batch [199]	Speed: 9968.957882 samples/sec	accuracy=0.654993	lr=0.027706
Epoch[336] Batch [249]	Speed: 11569.169198 samples/sec	accuracy=0.654934	lr=0.027517
Epoch[336] Batch [299]	Speed: 11182.623648 samples/sec	accuracy=0.654993	lr=0.027329
Epoch[336] Batch [349]	Speed: 11333.644739 samples/sec	accuracy=0.654940	lr=0.027142
Epoch[336] Batch [399]	Speed: 11593.957189 samples/sec	accuracy=0.654830	lr=0.026955
Epoch[336] Batch [449]	Speed: 11577.585064 samples/sec	accuracy=0.654869	lr=0.026769
Epoch[336] Batch [499]	Speed: 11402.745945 samples/sec	accuracy=0.654723	lr=0.026583
Epoch[336] Batch [549]	Speed: 11521.918292 samples/sec	accuracy=0.654710	lr=0.026399
Epoch[336] Batch [599]	Speed: 11372.977166 samples/sec	accuracy=0.654535	lr=0.026214
[Epoch 336] training: accuracy=0.654440
[Epoch 336] speed: 11239 samples/sec	time cost: 118.468859
[Epoch 336] validation: err-top1=0.329427 err-top5=0.127441
Epoch[337] Batch [49]	Speed: 11443.055395 samples/sec	accuracy=0.657783	lr=0.025936
Epoch[337] Batch [99]	Speed: 11179.049090 samples/sec	accuracy=0.657402	lr=0.025753
Epoch[337] Batch [149]	Speed: 11427.720557 samples/sec	accuracy=0.657520	lr=0.025571
Epoch[337] Batch [199]	Speed: 11400.842984 samples/sec	accuracy=0.657139	lr=0.025390
Epoch[337] Batch [249]	Speed: 10433.386995 samples/sec	accuracy=0.656906	lr=0.025209
Epoch[337] Batch [299]	Speed: 11483.631505 samples/sec	accuracy=0.656356	lr=0.025029
Epoch[337] Batch [349]	Speed: 10163.097052 samples/sec	accuracy=0.656356	lr=0.024850
Epoch[337] Batch [399]	Speed: 10909.160056 samples/sec	accuracy=0.656362	lr=0.024671
Epoch[337] Batch [449]	Speed: 11420.808590 samples/sec	accuracy=0.656238	lr=0.024493
Epoch[337] Batch [499]	Speed: 11380.018284 samples/sec	accuracy=0.656207	lr=0.024315
Epoch[337] Batch [549]	Speed: 11605.135372 samples/sec	accuracy=0.655997	lr=0.024138
Epoch[337] Batch [599]	Speed: 11088.126248 samples/sec	accuracy=0.655876	lr=0.023962
[Epoch 337] training: accuracy=0.655803
[Epoch 337] speed: 11137 samples/sec	time cost: 119.266647
[Epoch 337] validation: err-top1=0.329407 err-top5=0.128723
Epoch[338] Batch [49]	Speed: 11110.084068 samples/sec	accuracy=0.657285	lr=0.023699
Epoch[338] Batch [99]	Speed: 11506.335742 samples/sec	accuracy=0.657578	lr=0.023524
Epoch[338] Batch [149]	Speed: 11342.707356 samples/sec	accuracy=0.657585	lr=0.023350
Epoch[338] Batch [199]	Speed: 11512.784410 samples/sec	accuracy=0.657744	lr=0.023177
Epoch[338] Batch [249]	Speed: 11579.535938 samples/sec	accuracy=0.657596	lr=0.023004
Epoch[338] Batch [299]	Speed: 10833.494545 samples/sec	accuracy=0.657018	lr=0.022832
Epoch[338] Batch [349]	Speed: 11389.151622 samples/sec	accuracy=0.657130	lr=0.022661
Epoch[338] Batch [399]	Speed: 10882.938914 samples/sec	accuracy=0.656931	lr=0.022490
Epoch[338] Batch [449]	Speed: 11414.501955 samples/sec	accuracy=0.657025	lr=0.022320
Epoch[338] Batch [499]	Speed: 11340.319523 samples/sec	accuracy=0.656964	lr=0.022150
Epoch[338] Batch [549]	Speed: 11337.774368 samples/sec	accuracy=0.656866	lr=0.021982
Epoch[338] Batch [599]	Speed: 10953.261011 samples/sec	accuracy=0.656662	lr=0.021813
[Epoch 338] training: accuracy=0.656514
[Epoch 338] speed: 11242 samples/sec	time cost: 118.754470
[Epoch 338] validation: err-top1=0.330234 err-top5=0.129258
Epoch[339] Batch [49]	Speed: 10662.042928 samples/sec	accuracy=0.657725	lr=0.021559
Epoch[339] Batch [99]	Speed: 10423.498428 samples/sec	accuracy=0.657769	lr=0.021392
Epoch[339] Batch [149]	Speed: 11184.058655 samples/sec	accuracy=0.657734	lr=0.021226
Epoch[339] Batch [199]	Speed: 10682.975925 samples/sec	accuracy=0.657532	lr=0.021061
Epoch[339] Batch [249]	Speed: 11217.503262 samples/sec	accuracy=0.657611	lr=0.020896
Epoch[339] Batch [299]	Speed: 11384.607263 samples/sec	accuracy=0.657446	lr=0.020732
Epoch[339] Batch [349]	Speed: 11061.437614 samples/sec	accuracy=0.657239	lr=0.020569
Epoch[339] Batch [399]	Speed: 11017.520803 samples/sec	accuracy=0.657228	lr=0.020406
Epoch[339] Batch [449]	Speed: 11447.039995 samples/sec	accuracy=0.656915	lr=0.020244
Epoch[339] Batch [499]	Speed: 10967.488735 samples/sec	accuracy=0.656796	lr=0.020083
Epoch[339] Batch [549]	Speed: 11272.279445 samples/sec	accuracy=0.656923	lr=0.019922
Epoch[339] Batch [599]	Speed: 11180.447963 samples/sec	accuracy=0.656851	lr=0.019762
[Epoch 339] training: accuracy=0.656913
[Epoch 339] speed: 11002 samples/sec	time cost: 120.674935
[Epoch 339] validation: err-top1=0.328715 err-top5=0.127747
Epoch[340] Batch [49]	Speed: 11385.791834 samples/sec	accuracy=0.660791	lr=0.019523
Epoch[340] Batch [99]	Speed: 11169.366109 samples/sec	accuracy=0.660942	lr=0.019364
Epoch[340] Batch [149]	Speed: 11205.618391 samples/sec	accuracy=0.660544	lr=0.019206
Epoch[340] Batch [199]	Speed: 11674.012597 samples/sec	accuracy=0.659661	lr=0.019049
Epoch[340] Batch [249]	Speed: 11229.374028 samples/sec	accuracy=0.659680	lr=0.018892
Epoch[340] Batch [299]	Speed: 11368.275087 samples/sec	accuracy=0.659749	lr=0.018736
Epoch[340] Batch [349]	Speed: 11510.533896 samples/sec	accuracy=0.659477	lr=0.018581
Epoch[340] Batch [399]	Speed: 10227.053498 samples/sec	accuracy=0.659115	lr=0.018426
Epoch[340] Batch [449]	Speed: 11175.580627 samples/sec	accuracy=0.659240	lr=0.018272
Epoch[340] Batch [499]	Speed: 11017.070320 samples/sec	accuracy=0.659181	lr=0.018118
Epoch[340] Batch [549]	Speed: 10856.889625 samples/sec	accuracy=0.658928	lr=0.017966
Epoch[340] Batch [599]	Speed: 10805.396883 samples/sec	accuracy=0.658910	lr=0.017813
[Epoch 340] training: accuracy=0.658863
[Epoch 340] speed: 11093 samples/sec	time cost: 120.165758
[Epoch 340] validation: err-top1=0.327227 err-top5=0.127188
Epoch[341] Batch [49]	Speed: 10724.797299 samples/sec	accuracy=0.660791	lr=0.017583
Epoch[341] Batch [99]	Speed: 11207.389178 samples/sec	accuracy=0.659512	lr=0.017433
Epoch[341] Batch [149]	Speed: 11023.076019 samples/sec	accuracy=0.659499	lr=0.017283
Epoch[341] Batch [199]	Speed: 10915.005655 samples/sec	accuracy=0.659568	lr=0.017134
Epoch[341] Batch [249]	Speed: 10974.863808 samples/sec	accuracy=0.659473	lr=0.016985
Epoch[341] Batch [299]	Speed: 11511.645468 samples/sec	accuracy=0.659412	lr=0.016837
Epoch[341] Batch [349]	Speed: 11323.473082 samples/sec	accuracy=0.659365	lr=0.016690
Epoch[341] Batch [399]	Speed: 11455.912143 samples/sec	accuracy=0.659712	lr=0.016543
Epoch[341] Batch [449]	Speed: 11268.669523 samples/sec	accuracy=0.659283	lr=0.016397
Epoch[341] Batch [499]	Speed: 11494.263171 samples/sec	accuracy=0.659369	lr=0.016252
Epoch[341] Batch [549]	Speed: 10759.861698 samples/sec	accuracy=0.659158	lr=0.016107
Epoch[341] Batch [599]	Speed: 10970.344152 samples/sec	accuracy=0.658868	lr=0.015963
[Epoch 341] training: accuracy=0.658894
[Epoch 341] speed: 11125 samples/sec	time cost: 119.428010
[Epoch 341] validation: err-top1=0.328125 err-top5=0.127665
Epoch[342] Batch [49]	Speed: 11109.532015 samples/sec	accuracy=0.662676	lr=0.015748
Epoch[342] Batch [99]	Speed: 11088.133405 samples/sec	accuracy=0.660991	lr=0.015605
Epoch[342] Batch [149]	Speed: 10894.370163 samples/sec	accuracy=0.659792	lr=0.015464
Epoch[342] Batch [199]	Speed: 11285.196605 samples/sec	accuracy=0.659761	lr=0.015322
Epoch[342] Batch [249]	Speed: 11626.565662 samples/sec	accuracy=0.660758	lr=0.015182
Epoch[342] Batch [299]	Speed: 10771.776707 samples/sec	accuracy=0.660638	lr=0.015042
Epoch[342] Batch [349]	Speed: 11242.617630 samples/sec	accuracy=0.660834	lr=0.014903
Epoch[342] Batch [399]	Speed: 11367.115518 samples/sec	accuracy=0.660790	lr=0.014764
Epoch[342] Batch [449]	Speed: 11018.729998 samples/sec	accuracy=0.660903	lr=0.014626
Epoch[342] Batch [499]	Speed: 11665.389474 samples/sec	accuracy=0.660683	lr=0.014489
Epoch[342] Batch [549]	Speed: 11193.269164 samples/sec	accuracy=0.660551	lr=0.014352
Epoch[342] Batch [599]	Speed: 11833.564581 samples/sec	accuracy=0.660553	lr=0.014216
[Epoch 342] training: accuracy=0.660458
[Epoch 342] speed: 11231 samples/sec	time cost: 118.849264
[Epoch 342] validation: err-top1=0.326797 err-top5=0.127031
Epoch[343] Batch [49]	Speed: 10585.676276 samples/sec	accuracy=0.662197	lr=0.014010
Epoch[343] Batch [99]	Speed: 11614.781405 samples/sec	accuracy=0.661172	lr=0.013876
Epoch[343] Batch [149]	Speed: 10713.251555 samples/sec	accuracy=0.661497	lr=0.013742
Epoch[343] Batch [199]	Speed: 11345.884705 samples/sec	accuracy=0.661516	lr=0.013609
Epoch[343] Batch [249]	Speed: 10410.672768 samples/sec	accuracy=0.661316	lr=0.013476
Epoch[343] Batch [299]	Speed: 11477.654963 samples/sec	accuracy=0.661084	lr=0.013345
Epoch[343] Batch [349]	Speed: 11646.382273 samples/sec	accuracy=0.660956	lr=0.013213
Epoch[343] Batch [399]	Speed: 11612.093053 samples/sec	accuracy=0.660570	lr=0.013083
Epoch[343] Batch [449]	Speed: 11432.282061 samples/sec	accuracy=0.660250	lr=0.012953
Epoch[343] Batch [499]	Speed: 11435.080202 samples/sec	accuracy=0.660143	lr=0.012824
Epoch[343] Batch [549]	Speed: 11357.978428 samples/sec	accuracy=0.660162	lr=0.012695
Epoch[343] Batch [599]	Speed: 11365.750153 samples/sec	accuracy=0.660340	lr=0.012567
[Epoch 343] training: accuracy=0.660361
[Epoch 343] speed: 11208 samples/sec	time cost: 118.763587
[Epoch 343] validation: err-top1=0.325745 err-top5=0.126689
Epoch[344] Batch [49]	Speed: 10586.999475 samples/sec	accuracy=0.661338	lr=0.012374
Epoch[344] Batch [99]	Speed: 11040.968129 samples/sec	accuracy=0.663286	lr=0.012247
Epoch[344] Batch [149]	Speed: 11109.509026 samples/sec	accuracy=0.663346	lr=0.012122
Epoch[344] Batch [199]	Speed: 10957.901351 samples/sec	accuracy=0.662561	lr=0.011997
Epoch[344] Batch [249]	Speed: 11402.142934 samples/sec	accuracy=0.662273	lr=0.011872
Epoch[344] Batch [299]	Speed: 11207.593018 samples/sec	accuracy=0.662433	lr=0.011748
Epoch[344] Batch [349]	Speed: 11616.076253 samples/sec	accuracy=0.661922	lr=0.011625
Epoch[344] Batch [399]	Speed: 11390.461595 samples/sec	accuracy=0.661670	lr=0.011503
Epoch[344] Batch [449]	Speed: 10769.457916 samples/sec	accuracy=0.661540	lr=0.011381
Epoch[344] Batch [499]	Speed: 11545.425105 samples/sec	accuracy=0.661595	lr=0.011260
Epoch[344] Batch [549]	Speed: 10896.058311 samples/sec	accuracy=0.661653	lr=0.011139
Epoch[344] Batch [599]	Speed: 10994.478492 samples/sec	accuracy=0.661702	lr=0.011019
[Epoch 344] training: accuracy=0.661652
[Epoch 344] speed: 11104 samples/sec	time cost: 119.782283
[Epoch 344] validation: err-top1=0.324483 err-top5=0.126689
Epoch[345] Batch [49]	Speed: 11008.605607 samples/sec	accuracy=0.662002	lr=0.010841
Epoch[345] Batch [99]	Speed: 11305.557852 samples/sec	accuracy=0.663779	lr=0.010722
Epoch[345] Batch [149]	Speed: 11338.438536 samples/sec	accuracy=0.663226	lr=0.010605
Epoch[345] Batch [199]	Speed: 10617.919397 samples/sec	accuracy=0.662900	lr=0.010488
Epoch[345] Batch [249]	Speed: 11310.566493 samples/sec	accuracy=0.662816	lr=0.010371
Epoch[345] Batch [299]	Speed: 11056.273711 samples/sec	accuracy=0.662542	lr=0.010256
Epoch[345] Batch [349]	Speed: 10485.940483 samples/sec	accuracy=0.662448	lr=0.010141
Epoch[345] Batch [399]	Speed: 10887.019773 samples/sec	accuracy=0.662275	lr=0.010026
Epoch[345] Batch [449]	Speed: 11031.673514 samples/sec	accuracy=0.662127	lr=0.009912
Epoch[345] Batch [499]	Speed: 11039.317067 samples/sec	accuracy=0.662089	lr=0.009799
Epoch[345] Batch [549]	Speed: 11285.508852 samples/sec	accuracy=0.661950	lr=0.009687
Epoch[345] Batch [599]	Speed: 10499.031175 samples/sec	accuracy=0.662209	lr=0.009575
[Epoch 345] training: accuracy=0.662327
[Epoch 345] speed: 10958 samples/sec	time cost: 121.639188
[Epoch 345] validation: err-top1=0.324824 err-top5=0.126152
Epoch[346] Batch [49]	Speed: 10314.185687 samples/sec	accuracy=0.666006	lr=0.009406
Epoch[346] Batch [99]	Speed: 11111.196674 samples/sec	accuracy=0.665942	lr=0.009296
Epoch[346] Batch [149]	Speed: 10794.745354 samples/sec	accuracy=0.665433	lr=0.009187
Epoch[346] Batch [199]	Speed: 11347.135578 samples/sec	accuracy=0.664778	lr=0.009078
Epoch[346] Batch [249]	Speed: 11255.772745 samples/sec	accuracy=0.664908	lr=0.008969
Epoch[346] Batch [299]	Speed: 10703.260657 samples/sec	accuracy=0.664276	lr=0.008862
Epoch[346] Batch [349]	Speed: 11137.187300 samples/sec	accuracy=0.664234	lr=0.008755
Epoch[346] Batch [399]	Speed: 11397.137861 samples/sec	accuracy=0.663872	lr=0.008648
Epoch[346] Batch [449]	Speed: 10684.371670 samples/sec	accuracy=0.663528	lr=0.008543
Epoch[346] Batch [499]	Speed: 10680.008654 samples/sec	accuracy=0.663360	lr=0.008438
Epoch[346] Batch [549]	Speed: 11174.251290 samples/sec	accuracy=0.663213	lr=0.008333
Epoch[346] Batch [599]	Speed: 11249.161141 samples/sec	accuracy=0.663049	lr=0.008230
[Epoch 346] training: accuracy=0.663137
[Epoch 346] speed: 10968 samples/sec	time cost: 121.048728
[Epoch 346] validation: err-top1=0.323222 err-top5=0.125916
Epoch[347] Batch [49]	Speed: 10841.090778 samples/sec	accuracy=0.664834	lr=0.008075
Epoch[347] Batch [99]	Speed: 11268.536775 samples/sec	accuracy=0.664053	lr=0.007973
Epoch[347] Batch [149]	Speed: 11209.682442 samples/sec	accuracy=0.663447	lr=0.007872
Epoch[347] Batch [199]	Speed: 11360.949767 samples/sec	accuracy=0.663049	lr=0.007771
Epoch[347] Batch [249]	Speed: 10965.143435 samples/sec	accuracy=0.662998	lr=0.007671
Epoch[347] Batch [299]	Speed: 11231.528267 samples/sec	accuracy=0.662861	lr=0.007571
Epoch[347] Batch [349]	Speed: 11183.768013 samples/sec	accuracy=0.662923	lr=0.007472
Epoch[347] Batch [399]	Speed: 11174.074244 samples/sec	accuracy=0.663401	lr=0.007374
Epoch[347] Batch [449]	Speed: 11336.881948 samples/sec	accuracy=0.663621	lr=0.007277
Epoch[347] Batch [499]	Speed: 11109.404427 samples/sec	accuracy=0.663597	lr=0.007180
Epoch[347] Batch [549]	Speed: 11270.923161 samples/sec	accuracy=0.663836	lr=0.007083
Epoch[347] Batch [599]	Speed: 11370.292110 samples/sec	accuracy=0.663588	lr=0.006988
[Epoch 347] training: accuracy=0.663449
[Epoch 347] speed: 11166 samples/sec	time cost: 119.299458
[Epoch 347] validation: err-top1=0.323770 err-top5=0.125977
Epoch[348] Batch [49]	Speed: 11031.975857 samples/sec	accuracy=0.665498	lr=0.006844
Epoch[348] Batch [99]	Speed: 11535.044042 samples/sec	accuracy=0.665933	lr=0.006750
Epoch[348] Batch [149]	Speed: 11450.987683 samples/sec	accuracy=0.664323	lr=0.006656
Epoch[348] Batch [199]	Speed: 10774.891427 samples/sec	accuracy=0.664644	lr=0.006564
Epoch[348] Batch [249]	Speed: 11387.271007 samples/sec	accuracy=0.664260	lr=0.006472
Epoch[348] Batch [299]	Speed: 10902.970304 samples/sec	accuracy=0.664040	lr=0.006380
Epoch[348] Batch [349]	Speed: 10560.065114 samples/sec	accuracy=0.664240	lr=0.006290
Epoch[348] Batch [399]	Speed: 10412.066666 samples/sec	accuracy=0.664204	lr=0.006199
Epoch[348] Batch [449]	Speed: 11138.775327 samples/sec	accuracy=0.664186	lr=0.006110
Epoch[348] Batch [499]	Speed: 11565.207509 samples/sec	accuracy=0.663953	lr=0.006021
Epoch[348] Batch [549]	Speed: 11570.546783 samples/sec	accuracy=0.664205	lr=0.005933
Epoch[348] Batch [599]	Speed: 11303.612220 samples/sec	accuracy=0.664240	lr=0.005846
[Epoch 348] training: accuracy=0.664255
[Epoch 348] speed: 11082 samples/sec	time cost: 119.853851
[Epoch 348] validation: err-top1=0.324097 err-top5=0.125488
Epoch[349] Batch [49]	Speed: 11007.005393 samples/sec	accuracy=0.665137	lr=0.005715
Epoch[349] Batch [99]	Speed: 10891.320495 samples/sec	accuracy=0.665640	lr=0.005630
Epoch[349] Batch [149]	Speed: 10860.659575 samples/sec	accuracy=0.665876	lr=0.005544
Epoch[349] Batch [199]	Speed: 11014.784277 samples/sec	accuracy=0.665657	lr=0.005460
Epoch[349] Batch [249]	Speed: 11390.280651 samples/sec	accuracy=0.665064	lr=0.005376
Epoch[349] Batch [299]	Speed: 11508.465582 samples/sec	accuracy=0.665050	lr=0.005293
Epoch[349] Batch [349]	Speed: 11749.884972 samples/sec	accuracy=0.664989	lr=0.005210
Epoch[349] Batch [399]	Speed: 11790.783880 samples/sec	accuracy=0.665175	lr=0.005128
Epoch[349] Batch [449]	Speed: 11031.270605 samples/sec	accuracy=0.665027	lr=0.005047
Epoch[349] Batch [499]	Speed: 11434.560526 samples/sec	accuracy=0.664877	lr=0.004966
Epoch[349] Batch [549]	Speed: 11540.326038 samples/sec	accuracy=0.664737	lr=0.004886
Epoch[349] Batch [599]	Speed: 11099.209746 samples/sec	accuracy=0.664714	lr=0.004806
[Epoch 349] training: accuracy=0.664711
[Epoch 349] speed: 11268 samples/sec	time cost: 118.181545
[Epoch 349] validation: err-top1=0.323263 err-top5=0.125448
Epoch[350] Batch [49]	Speed: 11139.857859 samples/sec	accuracy=0.665312	lr=0.004687
Epoch[350] Batch [99]	Speed: 10989.924954 samples/sec	accuracy=0.664790	lr=0.004609
Epoch[350] Batch [149]	Speed: 11307.993886 samples/sec	accuracy=0.665059	lr=0.004532
Epoch[350] Batch [199]	Speed: 11465.275796 samples/sec	accuracy=0.665083	lr=0.004456
Epoch[350] Batch [249]	Speed: 11712.406383 samples/sec	accuracy=0.664945	lr=0.004380
Epoch[350] Batch [299]	Speed: 11456.166987 samples/sec	accuracy=0.664642	lr=0.004305
Epoch[350] Batch [349]	Speed: 11395.494063 samples/sec	accuracy=0.664750	lr=0.004230
Epoch[350] Batch [399]	Speed: 11217.363221 samples/sec	accuracy=0.665046	lr=0.004156
Epoch[350] Batch [449]	Speed: 11237.971246 samples/sec	accuracy=0.664587	lr=0.004083
Epoch[350] Batch [499]	Speed: 11403.876158 samples/sec	accuracy=0.664658	lr=0.004011
Epoch[350] Batch [549]	Speed: 11539.465005 samples/sec	accuracy=0.664712	lr=0.003939
Epoch[350] Batch [599]	Speed: 11666.048536 samples/sec	accuracy=0.664653	lr=0.003867
[Epoch 350] training: accuracy=0.664722
[Epoch 350] speed: 11330 samples/sec	time cost: 117.904824
[Epoch 350] validation: err-top1=0.322949 err-top5=0.124785
Epoch[351] Batch [49]	Speed: 9786.189155 samples/sec	accuracy=0.666094	lr=0.003760
Epoch[351] Batch [99]	Speed: 11242.694146 samples/sec	accuracy=0.665488	lr=0.003691
Epoch[351] Batch [149]	Speed: 11228.114346 samples/sec	accuracy=0.665309	lr=0.003622
Epoch[351] Batch [199]	Speed: 10530.512622 samples/sec	accuracy=0.665625	lr=0.003553
Epoch[351] Batch [249]	Speed: 11142.702857 samples/sec	accuracy=0.665740	lr=0.003486
Epoch[351] Batch [299]	Speed: 11565.493400 samples/sec	accuracy=0.666081	lr=0.003419
Epoch[351] Batch [349]	Speed: 11348.438302 samples/sec	accuracy=0.665986	lr=0.003352
Epoch[351] Batch [399]	Speed: 11518.295620 samples/sec	accuracy=0.665618	lr=0.003287
Epoch[351] Batch [449]	Speed: 11886.621568 samples/sec	accuracy=0.665890	lr=0.003221
Epoch[351] Batch [499]	Speed: 11125.616880 samples/sec	accuracy=0.665757	lr=0.003157
Epoch[351] Batch [549]	Speed: 11293.376186 samples/sec	accuracy=0.665699	lr=0.003093
Epoch[351] Batch [599]	Speed: 11241.011038 samples/sec	accuracy=0.665758	lr=0.003030
[Epoch 351] training: accuracy=0.665711
[Epoch 351] speed: 11112 samples/sec	time cost: 119.611105
[Epoch 351] validation: err-top1=0.322449 err-top5=0.125285
Epoch[352] Batch [49]	Speed: 10563.199133 samples/sec	accuracy=0.663652	lr=0.002937
Epoch[352] Batch [99]	Speed: 11577.857210 samples/sec	accuracy=0.664395	lr=0.002875
Epoch[352] Batch [149]	Speed: 11586.865547 samples/sec	accuracy=0.664408	lr=0.002814
Epoch[352] Batch [199]	Speed: 11429.914771 samples/sec	accuracy=0.664634	lr=0.002754
Epoch[352] Batch [249]	Speed: 10882.506538 samples/sec	accuracy=0.664672	lr=0.002695
Epoch[352] Batch [299]	Speed: 11342.037595 samples/sec	accuracy=0.664412	lr=0.002636
Epoch[352] Batch [349]	Speed: 10557.348670 samples/sec	accuracy=0.664694	lr=0.002577
Epoch[352] Batch [399]	Speed: 10995.724016 samples/sec	accuracy=0.664885	lr=0.002520
Epoch[352] Batch [449]	Speed: 11258.686119 samples/sec	accuracy=0.665317	lr=0.002463
Epoch[352] Batch [499]	Speed: 10650.773664 samples/sec	accuracy=0.665310	lr=0.002407
Epoch[352] Batch [549]	Speed: 10909.460154 samples/sec	accuracy=0.665541	lr=0.002351
Epoch[352] Batch [599]	Speed: 11655.562443 samples/sec	accuracy=0.665377	lr=0.002296
[Epoch 352] training: accuracy=0.665344
[Epoch 352] speed: 11096 samples/sec	time cost: 120.181184
[Epoch 352] validation: err-top1=0.322949 err-top5=0.124492
Epoch[353] Batch [49]	Speed: 11056.439359 samples/sec	accuracy=0.665508	lr=0.002214
Epoch[353] Batch [99]	Speed: 10723.518148 samples/sec	accuracy=0.665991	lr=0.002160
Epoch[353] Batch [149]	Speed: 10907.293889 samples/sec	accuracy=0.665973	lr=0.002107
Epoch[353] Batch [199]	Speed: 11334.049700 samples/sec	accuracy=0.665959	lr=0.002055
Epoch[353] Batch [249]	Speed: 11387.817795 samples/sec	accuracy=0.666229	lr=0.002004
Epoch[353] Batch [299]	Speed: 11299.642994 samples/sec	accuracy=0.665926	lr=0.001953
Epoch[353] Batch [349]	Speed: 11355.369800 samples/sec	accuracy=0.666165	lr=0.001903
Epoch[353] Batch [399]	Speed: 11388.076564 samples/sec	accuracy=0.665854	lr=0.001854
Epoch[353] Batch [449]	Speed: 11320.421056 samples/sec	accuracy=0.665930	lr=0.001805
Epoch[353] Batch [499]	Speed: 11468.176465 samples/sec	accuracy=0.665987	lr=0.001757
Epoch[353] Batch [549]	Speed: 11058.385957 samples/sec	accuracy=0.666049	lr=0.001709
Epoch[353] Batch [599]	Speed: 11417.960362 samples/sec	accuracy=0.666019	lr=0.001662
[Epoch 353] training: accuracy=0.666052
[Epoch 353] speed: 11187 samples/sec	time cost: 118.698162
[Epoch 353] validation: err-top1=0.322896 err-top5=0.125041
Epoch[354] Batch [49]	Speed: 11240.340878 samples/sec	accuracy=0.665117	lr=0.001593
Epoch[354] Batch [99]	Speed: 11223.936315 samples/sec	accuracy=0.664302	lr=0.001548
Epoch[354] Batch [149]	Speed: 11104.351543 samples/sec	accuracy=0.665228	lr=0.001504
Epoch[354] Batch [199]	Speed: 11519.880800 samples/sec	accuracy=0.665786	lr=0.001460
Epoch[354] Batch [249]	Speed: 11330.164581 samples/sec	accuracy=0.665695	lr=0.001416
Epoch[354] Batch [299]	Speed: 10532.582420 samples/sec	accuracy=0.665968	lr=0.001374
Epoch[354] Batch [349]	Speed: 10824.222279 samples/sec	accuracy=0.666112	lr=0.001332
Epoch[354] Batch [399]	Speed: 11463.204132 samples/sec	accuracy=0.666309	lr=0.001290
Epoch[354] Batch [449]	Speed: 11773.393901 samples/sec	accuracy=0.666290	lr=0.001250
Epoch[354] Batch [499]	Speed: 11572.508382 samples/sec	accuracy=0.666533	lr=0.001210
Epoch[354] Batch [549]	Speed: 10894.568579 samples/sec	accuracy=0.666560	lr=0.001170
Epoch[354] Batch [599]	Speed: 11426.111393 samples/sec	accuracy=0.666630	lr=0.001132
[Epoch 354] training: accuracy=0.666516
[Epoch 354] speed: 11224 samples/sec	time cost: 118.481293
[Epoch 354] validation: err-top1=0.322591 err-top5=0.124552
Epoch[355] Batch [49]	Speed: 11042.233577 samples/sec	accuracy=0.668984	lr=0.001074
Epoch[355] Batch [99]	Speed: 11684.835830 samples/sec	accuracy=0.668848	lr=0.001037
Epoch[355] Batch [149]	Speed: 10765.725909 samples/sec	accuracy=0.667220	lr=0.001001
Epoch[355] Batch [199]	Speed: 11553.647318 samples/sec	accuracy=0.667161	lr=0.000965
Epoch[355] Batch [249]	Speed: 11194.068219 samples/sec	accuracy=0.667098	lr=0.000930
Epoch[355] Batch [299]	Speed: 11665.649604 samples/sec	accuracy=0.667028	lr=0.000895
Epoch[355] Batch [349]	Speed: 11106.247272 samples/sec	accuracy=0.666500	lr=0.000861
Epoch[355] Batch [399]	Speed: 10860.037567 samples/sec	accuracy=0.666555	lr=0.000828
Epoch[355] Batch [449]	Speed: 10385.557625 samples/sec	accuracy=0.666702	lr=0.000796
Epoch[355] Batch [499]	Speed: 11363.490006 samples/sec	accuracy=0.666596	lr=0.000764
Epoch[355] Batch [549]	Speed: 10835.081056 samples/sec	accuracy=0.666424	lr=0.000733
Epoch[355] Batch [599]	Speed: 10315.003379 samples/sec	accuracy=0.666444	lr=0.000702
[Epoch 355] training: accuracy=0.666443
[Epoch 355] speed: 11028 samples/sec	time cost: 120.729466
[Epoch 355] validation: err-top1=0.322793 err-top5=0.124668
Epoch[356] Batch [49]	Speed: 10583.740737 samples/sec	accuracy=0.665947	lr=0.000657
Epoch[356] Batch [99]	Speed: 10951.913939 samples/sec	accuracy=0.667261	lr=0.000628
Epoch[356] Batch [149]	Speed: 11494.376681 samples/sec	accuracy=0.667171	lr=0.000600
Epoch[356] Batch [199]	Speed: 11338.345446 samples/sec	accuracy=0.666633	lr=0.000572
Epoch[356] Batch [249]	Speed: 10799.837521 samples/sec	accuracy=0.666426	lr=0.000545
Epoch[356] Batch [299]	Speed: 11642.374162 samples/sec	accuracy=0.666296	lr=0.000519
Epoch[356] Batch [349]	Speed: 11087.209156 samples/sec	accuracy=0.666469	lr=0.000493
Epoch[356] Batch [399]	Speed: 10613.696503 samples/sec	accuracy=0.666666	lr=0.000468
Epoch[356] Batch [449]	Speed: 10929.996304 samples/sec	accuracy=0.666924	lr=0.000444
Epoch[356] Batch [499]	Speed: 11349.284857 samples/sec	accuracy=0.666882	lr=0.000420
Epoch[356] Batch [549]	Speed: 11168.538920 samples/sec	accuracy=0.666955	lr=0.000397
Epoch[356] Batch [599]	Speed: 10854.199934 samples/sec	accuracy=0.666928	lr=0.000375
[Epoch 356] training: accuracy=0.666917
[Epoch 356] speed: 11053 samples/sec	time cost: 120.378562
[Epoch 356] validation: err-top1=0.322957 err-top5=0.125488
Epoch[357] Batch [49]	Speed: 10806.860695 samples/sec	accuracy=0.667207	lr=0.000342
Epoch[357] Batch [99]	Speed: 11206.956371 samples/sec	accuracy=0.666606	lr=0.000321
Epoch[357] Batch [149]	Speed: 11381.331284 samples/sec	accuracy=0.667243	lr=0.000301
Epoch[357] Batch [199]	Speed: 11509.127386 samples/sec	accuracy=0.667014	lr=0.000282
Epoch[357] Batch [249]	Speed: 11083.087263 samples/sec	accuracy=0.667092	lr=0.000263
Epoch[357] Batch [299]	Speed: 11496.289149 samples/sec	accuracy=0.667056	lr=0.000245
Epoch[357] Batch [349]	Speed: 10934.133693 samples/sec	accuracy=0.666982	lr=0.000227
Epoch[357] Batch [399]	Speed: 11103.779678 samples/sec	accuracy=0.666920	lr=0.000210
Epoch[357] Batch [449]	Speed: 11428.959431 samples/sec	accuracy=0.666976	lr=0.000194
Epoch[357] Batch [499]	Speed: 11629.108636 samples/sec	accuracy=0.667223	lr=0.000178
Epoch[357] Batch [549]	Speed: 11160.643804 samples/sec	accuracy=0.667246	lr=0.000163
Epoch[357] Batch [599]	Speed: 11441.367542 samples/sec	accuracy=0.667064	lr=0.000149
[Epoch 357] training: accuracy=0.666992
[Epoch 357] speed: 11238 samples/sec	time cost: 118.533092
[Epoch 357] validation: err-top1=0.322637 err-top5=0.124453
Epoch[358] Batch [49]	Speed: 11338.016500 samples/sec	accuracy=0.666641	lr=0.000129
Epoch[358] Batch [99]	Speed: 11149.301303 samples/sec	accuracy=0.667188	lr=0.000116
Epoch[358] Batch [149]	Speed: 11401.359599 samples/sec	accuracy=0.667337	lr=0.000104
Epoch[358] Batch [199]	Speed: 11380.659668 samples/sec	accuracy=0.667412	lr=0.000093
Epoch[358] Batch [249]	Speed: 11372.603145 samples/sec	accuracy=0.667082	lr=0.000082
Epoch[358] Batch [299]	Speed: 11468.406132 samples/sec	accuracy=0.666935	lr=0.000072
Epoch[358] Batch [349]	Speed: 10883.355328 samples/sec	accuracy=0.666943	lr=0.000063
Epoch[358] Batch [399]	Speed: 11249.099269 samples/sec	accuracy=0.667003	lr=0.000054
Epoch[358] Batch [449]	Speed: 11322.597837 samples/sec	accuracy=0.667068	lr=0.000046
Epoch[358] Batch [499]	Speed: 11293.311451 samples/sec	accuracy=0.667104	lr=0.000039
Epoch[358] Batch [549]	Speed: 11212.791546 samples/sec	accuracy=0.666944	lr=0.000032
Epoch[358] Batch [599]	Speed: 10997.262662 samples/sec	accuracy=0.667097	lr=0.000026
[Epoch 358] training: accuracy=0.667057
[Epoch 358] speed: 11204 samples/sec	time cost: 118.735709
[Epoch 358] validation: err-top1=0.322205 err-top5=0.124817
Epoch[359] Batch [49]	Speed: 10651.983474 samples/sec	accuracy=0.669355	lr=0.000018
Epoch[359] Batch [99]	Speed: 11229.428344 samples/sec	accuracy=0.668521	lr=0.000013
Epoch[359] Batch [149]	Speed: 11488.826252 samples/sec	accuracy=0.668708	lr=0.000009
Epoch[359] Batch [199]	Speed: 11704.143955 samples/sec	accuracy=0.667952	lr=0.000006
Epoch[359] Batch [249]	Speed: 11088.370145 samples/sec	accuracy=0.667311	lr=0.000004
Epoch[359] Batch [299]	Speed: 11181.861742 samples/sec	accuracy=0.666846	lr=0.000002
Epoch[359] Batch [349]	Speed: 10770.011526 samples/sec	accuracy=0.667028	lr=0.000001
Epoch[359] Batch [399]	Speed: 11216.086607 samples/sec	accuracy=0.667112	lr=0.000000
Epoch[359] Batch [449]	Speed: 11642.928681 samples/sec	accuracy=0.667087	lr=0.000000
Epoch[359] Batch [499]	Speed: 11100.137431 samples/sec	accuracy=0.667077	lr=0.000000
Epoch[359] Batch [549]	Speed: 10915.012867 samples/sec	accuracy=0.667051	lr=0.000000
Epoch[359] Batch [599]	Speed: 11199.644654 samples/sec	accuracy=0.666925	lr=0.000000
[Epoch 359] training: accuracy=0.667020
[Epoch 359] speed: 11103 samples/sec	time cost: 120.187334
[Epoch 359] validation: err-top1=0.322422 err-top5=0.124844