{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Automated ML for time series predicion"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We used one of the dataset in [Numenta Anomaly Benchmark (NAB)](https://github.com/numenta/NAB) for demo, i.e. NYC taxi passengers dataset, which contains 10320 records, each indicating the total number of taxi passengers in NYC at a corresonponding time spot. "
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"%load_ext autoreload\n",
"%autoreload 2"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 0. Helper function definations"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"# plot the predicted values and actual values (for the test data)\n",
"def plot_result(test_df, pred_df, dt_col=\"datetime\", value_col=\"value\", past_seq_len=1):\n",
" # target column of dataframe is \"value\"\n",
" # past sequence length is 50\n",
" pred_value = pred_df[value_col].values\n",
" true_value = test_df[value_col].values[past_seq_len:]\n",
" fig, axs = plt.subplots(figsize=(12, 5))\n",
"\n",
" axs.plot(pred_df[dt_col], pred_value, color='red', label='predicted values')\n",
" axs.plot(test_df[dt_col][past_seq_len:], true_value, color='blue', label='actual values')\n",
" axs.set_title('the predicted values and actual values (for the test data)')\n",
"\n",
" plt.xlabel(dt_col)\n",
" plt.xticks(rotation=45)\n",
" plt.ylabel('number of taxi passengers')\n",
" plt.legend(loc='upper left')\n",
" plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"# plot results of multi step forecasting\n",
"# plot at most five values for better view\n",
"# plot the predicted values and actual values (for the test data)\n",
"def plot_less_five_step_result(test_df, pred_df, dt_col=\"datetime\", value_col=\"value\", past_seq_len=1):\n",
" fig, axs = plt.subplots(figsize=(12, 5))\n",
" target_value = test_df[value_col].values[past_seq_len:]\n",
" axs.plot(test_df[dt_col][past_seq_len:], target_value, color='blue', label='actual values')\n",
"\n",
" value_cols=[\"{}_{}\".format(value_col, i) for i in range(min(pred_df.shape[1] - 1, 5))]\n",
" time_delta = pred_df[dt_col][1] - pred_df[dt_col][0]\n",
" plot_color = [\"g\", \"r\", \"c\", \"m\", \"y\"]\n",
" for i in range(len(value_cols)):\n",
" pred_value = pred_df[value_cols[i]].values\n",
" pred_dt = pred_df[dt_col].values + time_delta * i\n",
" axs.plot(pred_dt, pred_value, color=plot_color[i], label='predicted values' + str(i))\n",
"\n",
" axs.set_title('the predicted values and actual values (for the test data)')\n",
"\n",
" plt.xlabel(dt_col)\n",
" plt.xticks(rotation=45)\n",
" plt.ylabel('number of taxi passengers')\n",
" plt.legend(loc='upper left')\n",
" plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"# plot results of multi step forecasting\n",
"# plot result of multi step forecasting\n",
"# plot the predicted values and actual values (for the test data)\n",
"def plot_first_last_step_result(test_df, pred_df, dt_col=\"datetime\", value_col=\"value\", past_seq_len=1):\n",
" fig, axs = plt.subplots(figsize=(12, 5))\n",
" target_value = test_df[value_col].values[past_seq_len:]\n",
" axs.plot(test_df[dt_col][past_seq_len:], target_value, color='blue', label='actual values')\n",
"\n",
" value_cols=[\"{}_{}\".format(value_col, i) for i in range(pred_df.shape[1] - 1)]\n",
" time_delta = pred_df[dt_col][1] - pred_df[dt_col][0]\n",
" \n",
" pred_value_first = pred_df[value_cols[0]].values\n",
" pred_dt_first = pred_df[dt_col].values\n",
" axs.plot(pred_dt_first, pred_value_first, color=\"g\", label='first predicted values')\n",
" \n",
" pred_value_last = pred_df[value_cols[-1]].values\n",
" pred_dt_last = pred_df[dt_col].values + time_delta * (len(value_cols)-1)\n",
" axs.plot(pred_dt_last, pred_value_last, color=\"r\", label='last predicted values')\n",
"\n",
" axs.set_title('the predicted values and actual values (for the test data)')\n",
"\n",
" plt.xlabel(dt_col)\n",
" plt.xticks(rotation=45)\n",
" plt.ylabel('number of taxi passengers')\n",
" plt.legend(loc='upper left')\n",
" plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 1. load data"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Populating the interactive namespace from numpy and matplotlib\n"
]
}
],
"source": [
"import os\n",
"import pandas as pd\n",
"import numpy as np\n",
"\n",
"import matplotlib\n",
"matplotlib.use('Agg')\n",
"%pylab inline\n",
"import matplotlib.dates as md\n",
"from matplotlib import pyplot as plt"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"# load nyc taxi data\n",
"try:\n",
" dataset_path = os.getenv(\"ANALYTICS_ZOO_HOME\")+\"/bin/data/NAB/nyc_taxi/nyc_taxi.csv\"\n",
" df = pd.read_csv(dataset_path)\n",
"except Exception as e:\n",
" print(\"nyc_taxi.csv doesn't exist\")\n",
" print(\"you can run $ANALYTICS_ZOO_HOME/bin/data/NAB/nyc_taxi/get_nyc_taxi.sh to download nyc_taxi.csv\")"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"# download nyc_taxi.csv\n",
"#! $ANALYTICS_ZOO_HOME/dist/bin/data/NAB/nyc_taxi/get_nyc_taxi.sh"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"You can use split_input_df to split the whole dataset into train/val/test sets. There will be two columns in the output dataframe: \"datetime\" and \"value\", where the data type of \"datetime\" column is datetime64."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"from zoo.automl.common.util import split_input_df\n",
"train_df, val_df, test_df = split_input_df(df, val_split_ratio=0.1, test_split_ratio=0.1)"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" value | \n",
"
\n",
" \n",
" \n",
" \n",
" | count | \n",
" 8256.000000 | \n",
"
\n",
" \n",
" | mean | \n",
" 15421.585514 | \n",
"
\n",
" \n",
" | std | \n",
" 6871.989592 | \n",
"
\n",
" \n",
" | min | \n",
" 1431.000000 | \n",
"
\n",
" \n",
" | 25% | \n",
" 11040.000000 | \n",
"
\n",
" \n",
" | 50% | \n",
" 17014.500000 | \n",
"
\n",
" \n",
" | 75% | \n",
" 20032.250000 | \n",
"
\n",
" \n",
" | max | \n",
" 39197.000000 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" datetime | \n",
" value | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 2014-07-01 00:00:00 | \n",
" 10844 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2014-07-01 00:30:00 | \n",
" 8127 | \n",
"
\n",
" \n",
" | 2 | \n",
" 2014-07-01 01:00:00 | \n",
" 6210 | \n",
"
\n",
" \n",
" | 3 | \n",
" 2014-07-01 01:30:00 | \n",
" 4656 | \n",
"
\n",
" \n",
" | 4 | \n",
" 2014-07-01 02:00:00 | \n",
" 3820 | \n",
"
\n",
" \n",
"
\n",
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"# visualisation of anomaly throughout time in train_df\n",
"from pandas.plotting import register_matplotlib_converters\n",
"register_matplotlib_converters()\n",
"\n",
"fig, ax = plt.subplots(figsize=(12, 5))\n",
"# pd.plotting.deregister_matplotlib_converters()\n",
"\n",
"ax.plot(train_df['datetime'], train_df['value'], color='blue', linewidth=0.6)\n",
"ax.set_title('NYC taxi passengers throughout time')\n",
"\n",
"plt.xlabel('datetime')\n",
"plt.xticks(rotation=45) \n",
"plt.ylabel('The Number of NYC taxi passengers')\n",
"plt.legend(loc='upper left')\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 2. Train and validation"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"You can use analytices zoo automl to predict time series data by simply define a `TimeSequencePredictor`. \n",
"\n",
"We use feature tools to generate features from the given datetime. The generated features are \\['HOUR', 'DAY', 'MONTH'. 'IS_AWAKE', 'IS_BUSY_HOURS'\\]. Our feature space comprises these generated features as well as the original inputs such as \\['datetime','value','extra_features'\\]. \n",
"\n",
"Currently, We use RNN to learn from 50 previous values, and predict just the 1 next value. You can specify the sequence length to predict while creating `TimeSequencePredictor` with arg: `future_seq_len`."
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"# build time sequence predictor\n",
"from zoo.automl.regression.time_sequence_predictor import *\n",
"\n",
"# you need to specify the name of datetime column and target column\n",
"# The default names are \"datetime\" and \"value\" respectively.\n",
"tsp = TimeSequencePredictor(dt_col=\"datetime\",\n",
" target_col=\"value\",\n",
" extra_features_col=None)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Current pyspark location is : /home/shan/Applications/spark-2.4.3-bin-hadoop2.7/python/lib/pyspark.zip/pyspark/__init__.py\n",
"Start to getOrCreate SparkContext\n",
"Successfully got a SparkContext\n",
"Start to launch the JVM guarding process\n",
"JVM guarding process has been successfully launched\n",
"Start to launch ray on cluster\n",
"Start to launch ray on local\n"
]
}
],
"source": [
"from zoo import init_spark_on_local\n",
"from zoo.ray import RayContext\n",
"sc = init_spark_on_local(cores=4)\n",
"ray_ctx = RayContext(sc=sc, object_store_memory=\"1g\")\n",
"ray_ctx.init()"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"2019-12-18 15:48:56,962\tINFO tune.py:60 -- Tip: to resume incomplete experiments, pass resume='prompt' or resume=True to run()\n",
"2019-12-18 15:48:56,962\tINFO tune.py:223 -- Starting a new experiment.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model selection: Vanilla LSTM model is selected.\n",
"== Status ==\n",
"Using FIFO scheduling algorithm.\n",
"Resources requested: 0/4 CPUs, 0/0 GPUs (0/1.0 ps, 0/1.0 trainer)\n",
"Memory usage on this node: 28.9/67.4 GB\n",
"\n",
"== Status ==\n",
"Using FIFO scheduling algorithm.\n",
"Resources requested: 2/4 CPUs, 0/0 GPUs (0/1.0 ps, 0/1.0 trainer)\n",
"Memory usage on this node: 28.9/67.4 GB\n",
"Result logdir: /home/shan/ray_results/automl\n",
"Number of trials: 1 ({'RUNNING': 1})\n",
"RUNNING trials:\n",
" - train_func_0_batch_size=1024,dropout=0.48112,dropout_1=0.26833,dropout_2=0.39814,latent_dim=32,lr=0.0080377,lstm_1_units=64,lstm_2_units=8,past_seq_len=3,selected_features=['HOUR(datetime)' 'IS_BUSY_HOURS(datetime)' 'WEEKDAY(datetime)']:\tRUNNING\n",
"\n",
"\u001b[2m\u001b[36m(pid=7206)\u001b[0m Prepending /home/shan/anaconda2/envs/ray36/lib/python3.6/site-packages/bigdl/share/conf/spark-bigdl.conf to sys.path\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Prepending /home/shan/anaconda2/envs/ray36/lib/python3.6/site-packages/bigdl/share/conf/spark-bigdl.conf to sys.path\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Prepending /home/shan/anaconda2/envs/ray36/lib/python3.6/site-packages/bigdl/share/conf/spark-bigdl.conf to sys.path\n",
"\u001b[2m\u001b[36m(pid=7229)\u001b[0m Prepending /home/shan/anaconda2/envs/ray36/lib/python3.6/site-packages/bigdl/share/conf/spark-bigdl.conf to sys.path\n",
"\u001b[2m\u001b[36m(pid=7206)\u001b[0m Prepending /home/shan/anaconda2/envs/ray36/lib/python3.6/site-packages/bigdl/share/conf/spark-bigdl.conf to sys.path\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Prepending /home/shan/anaconda2/envs/ray36/lib/python3.6/site-packages/bigdl/share/conf/spark-bigdl.conf to sys.path\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Prepending /home/shan/anaconda2/envs/ray36/lib/python3.6/site-packages/bigdl/share/conf/spark-bigdl.conf to sys.path\n",
"\u001b[2m\u001b[36m(pid=7229)\u001b[0m Prepending /home/shan/anaconda2/envs/ray36/lib/python3.6/site-packages/bigdl/share/conf/spark-bigdl.conf to sys.path\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Model selection: Vanilla LSTM model is selected.\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Model selection: Vanilla LSTM model is selected.\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 2019-12-18 15:49:02.369651: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 2019-12-18 15:49:02.369651: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Train on 8253 samples, validate on 1029 samples\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Train on 8253 samples, validate on 1029 samples\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 6s - loss: 1.0319 - mean_squared_error: 1.0319\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 6s - loss: 1.0319 - mean_squared_error: 1.0319\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.5378 - mean_squared_error: 0.5378\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.5378 - mean_squared_error: 0.5378\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.4298 - mean_squared_error: 0.4298\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.4298 - mean_squared_error: 0.4298\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 1s 144us/step - loss: 0.4075 - mean_squared_error: 0.4075 - val_loss: 0.2391 - val_mean_squared_error: 0.2391\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.2575 - mean_squared_error: 0.2575\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 1s 144us/step - loss: 0.4075 - mean_squared_error: 0.4075 - val_loss: 0.2391 - val_mean_squared_error: 0.2391\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.2575 - mean_squared_error: 0.2575\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.2372 - mean_squared_error: 0.2372\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.2372 - mean_squared_error: 0.2372\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.2405 - mean_squared_error: 0.2405\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.2405 - mean_squared_error: 0.2405\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.2393 - mean_squared_error: 0.2393 - val_loss: 0.1187 - val_mean_squared_error: 0.1187\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.2393 - mean_squared_error: 0.2393 - val_loss: 0.1187 - val_mean_squared_error: 0.1187\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.2387 - mean_squared_error: 0.2387\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.2387 - mean_squared_error: 0.2387\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.2252 - mean_squared_error: 0.2252\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.2252 - mean_squared_error: 0.2252\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.2146 - mean_squared_error: 0.2146\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.2146 - mean_squared_error: 0.2146\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.2154 - mean_squared_error: 0.21\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.2152 - mean_squared_error: 0.2152 - val_loss: 0.2252 - val_mean_squared_error: 0.2252\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.2154 - mean_squared_error: 0.21\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.2152 - mean_squared_error: 0.2152 - val_loss: 0.2252 - val_mean_squared_error: 0.2252\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.2553 - mean_squared_error: 0.2553\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.2553 - mean_squared_error: 0.2553\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.2125 - mean_squared_error: 0.2125\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.2125 - mean_squared_error: 0.2125\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 5120/8253 [=================>............] - ETA: 0s - loss: 0.1989 - mean_squared_error: 0.1989\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 5120/8253 [=================>............] - ETA: 0s - loss: 0.1989 - mean_squared_error: 0.1989\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.2005 - mean_squared_error: 0.20\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 28us/step - loss: 0.2004 - mean_squared_error: 0.2004 - val_loss: 0.1079 - val_mean_squared_error: 0.1079\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.2005 - mean_squared_error: 0.20\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 28us/step - loss: 0.2004 - mean_squared_error: 0.2004 - val_loss: 0.1079 - val_mean_squared_error: 0.1079\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1815 - mean_squared_error: 0.1815\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1815 - mean_squared_error: 0.1815\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1827 - mean_squared_error: 0.1827\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1827 - mean_squared_error: 0.1827\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1716 - mean_squared_error: 0.1716\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1716 - mean_squared_error: 0.1716\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 31us/step - loss: 0.1715 - mean_squared_error: 0.1715 - val_loss: 0.2062 - val_mean_squared_error: 0.2062\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 31us/step - loss: 0.1715 - mean_squared_error: 0.1715 - val_loss: 0.2062 - val_mean_squared_error: 0.2062\n",
"== Status ==\n",
"Using FIFO scheduling algorithm.\n",
"Resources requested: 2/4 CPUs, 0/0 GPUs (0/1.0 ps, 0/1.0 trainer)\n",
"Memory usage on this node: 29.5/67.4 GB\n",
"Result logdir: /home/shan/ray_results/automl\n",
"Number of trials: 1 ({'RUNNING': 1})\n",
"RUNNING trials:\n",
" - train_func_0_batch_size=1024,dropout=0.48112,dropout_1=0.26833,dropout_2=0.39814,latent_dim=32,lr=0.0080377,lstm_1_units=64,lstm_2_units=8,past_seq_len=3,selected_features=['HOUR(datetime)' 'IS_BUSY_HOURS(datetime)' 'WEEKDAY(datetime)']:\tRUNNING, [2 CPUs, 0 GPUs], [pid=7207], 5 s, 1 iter\n",
"\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Train on 8253 samples, validate on 1029 samples\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Train on 8253 samples, validate on 1029 samples\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.2466 - mean_squared_error: 0.2466\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.2466 - mean_squared_error: 0.2466\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1960 - mean_squared_error: 0.1960\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1960 - mean_squared_error: 0.1960\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1723 - mean_squared_error: 0.1723\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1723 - mean_squared_error: 0.1723\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1650 - mean_squared_error: 0.16\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1648 - mean_squared_error: 0.1648 - val_loss: 0.0950 - val_mean_squared_error: 0.0950\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1474 - mean_squared_error: 0.1474\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1650 - mean_squared_error: 0.16\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1648 - mean_squared_error: 0.1648 - val_loss: 0.0950 - val_mean_squared_error: 0.0950\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1474 - mean_squared_error: 0.1474\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.2130 - mean_squared_error: 0.2130\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.2130 - mean_squared_error: 0.2130\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1809 - mean_squared_error: 0.1809\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1809 - mean_squared_error: 0.1809\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1680 - mean_squared_error: 0.16\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1678 - mean_squared_error: 0.1678 - val_loss: 0.0839 - val_mean_squared_error: 0.0839\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1680 - mean_squared_error: 0.16\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1678 - mean_squared_error: 0.1678 - val_loss: 0.0839 - val_mean_squared_error: 0.0839\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1549 - mean_squared_error: 0.1549\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1549 - mean_squared_error: 0.1549\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1627 - mean_squared_error: 0.1627\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1627 - mean_squared_error: 0.1627\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1579 - mean_squared_error: 0.1579\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1579 - mean_squared_error: 0.1579\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1508 - mean_squared_error: 0.1508 - val_loss: 0.1198 - val_mean_squared_error: 0.1198\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1508 - mean_squared_error: 0.1508 - val_loss: 0.1198 - val_mean_squared_error: 0.1198\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1437 - mean_squared_error: 0.1437\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1437 - mean_squared_error: 0.1437\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1404 - mean_squared_error: 0.1404\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1404 - mean_squared_error: 0.1404\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1384 - mean_squared_error: 0.1384\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1384 - mean_squared_error: 0.1384\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1468 - mean_squared_error: 0.1468\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1468 - mean_squared_error: 0.14\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 28us/step - loss: 0.1469 - mean_squared_error: 0.1469 - val_loss: 0.0838 - val_mean_squared_error: 0.0838\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 28us/step - loss: 0.1469 - mean_squared_error: 0.1469 - val_loss: 0.0838 - val_mean_squared_error: 0.0838\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1304 - mean_squared_error: 0.1304\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1304 - mean_squared_error: 0.1304\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1260 - mean_squared_error: 0.1260\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1260 - mean_squared_error: 0.1260\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1252 - mean_squared_error: 0.1252\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1252 - mean_squared_error: 0.1252\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1308 - mean_squared_error: 0.1308 - val_loss: 0.0829 - val_mean_squared_error: 0.0829\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1308 - mean_squared_error: 0.1308 - val_loss: 0.0829 - val_mean_squared_error: 0.0829\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Train on 8253 samples, validate on 1029 samples\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Train on 8253 samples, validate on 1029 samples\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1382 - mean_squared_error: 0.1382\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1382 - mean_squared_error: 0.1382\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1300 - mean_squared_error: 0.1300\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1300 - mean_squared_error: 0.1300\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1257 - mean_squared_error: 0.1257\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1257 - mean_squared_error: 0.1257\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1245 - mean_squared_error: 0.12\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1243 - mean_squared_error: 0.1243 - val_loss: 0.0532 - val_mean_squared_error: 0.0532\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1366 - mean_squared_error: 0.1366\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1245 - mean_squared_error: 0.12\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1243 - mean_squared_error: 0.1243 - val_loss: 0.0532 - val_mean_squared_error: 0.0532\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1366 - mean_squared_error: 0.1366\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1184 - mean_squared_error: 0.1184\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1184 - mean_squared_error: 0.1184\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1220 - mean_squared_error: 0.1220\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1220 - mean_squared_error: 0.1220\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 25us/step - loss: 0.1258 - mean_squared_error: 0.1258 - val_loss: 0.0817 - val_mean_squared_error: 0.0817\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 25us/step - loss: 0.1258 - mean_squared_error: 0.1258 - val_loss: 0.0817 - val_mean_squared_error: 0.0817\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1439 - mean_squared_error: 0.1439\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1439 - mean_squared_error: 0.1439\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1270 - mean_squared_error: 0.1270\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1270 - mean_squared_error: 0.1270\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1245 - mean_squared_error: 0.1245\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1245 - mean_squared_error: 0.1245\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1189 - mean_squared_error: 0.1189\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1189 - mean_squared_error: 0.1189\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1191 - mean_squared_error: 0.1191 - val_loss: 0.0416 - val_mean_squared_error: 0.0416\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1295 - mean_squared_error: 0.1295\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1191 - mean_squared_error: 0.1191 - val_loss: 0.0416 - val_mean_squared_error: 0.0416\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1295 - mean_squared_error: 0.1295\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1437 - mean_squared_error: 0.1437\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1437 - mean_squared_error: 0.1437\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1312 - mean_squared_error: 0.1312\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1312 - mean_squared_error: 0.1312\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1302 - mean_squared_error: 0.1302 - val_loss: 0.0622 - val_mean_squared_error: 0.0622\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1185 - mean_squared_error: 0.1185\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1302 - mean_squared_error: 0.1302 - val_loss: 0.0622 - val_mean_squared_error: 0.0622\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1185 - mean_squared_error: 0.1185\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1229 - mean_squared_error: 0.1229\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1229 - mean_squared_error: 0.1229\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 5120/8253 [=================>............] - ETA: 0s - loss: 0.1213 - mean_squared_error: 0.1213\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 5120/8253 [=================>............] - ETA: 0s - loss: 0.1213 - mean_squared_error: 0.1213\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1162 - mean_squared_error: 0.11\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1159 - mean_squared_error: 0.1159 - val_loss: 0.0325 - val_mean_squared_error: 0.0325\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Train on 8253 samples, validate on 1029 samples\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1162 - mean_squared_error: 0.11\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1159 - mean_squared_error: 0.1159 - val_loss: 0.0325 - val_mean_squared_error: 0.0325\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Train on 8253 samples, validate on 1029 samples\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1205 - mean_squared_error: 0.1205\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1205 - mean_squared_error: 0.1205\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1195 - mean_squared_error: 0.1195\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1195 - mean_squared_error: 0.1195\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1177 - mean_squared_error: 0.1177\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1177 - mean_squared_error: 0.1177\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1168 - mean_squared_error: 0.1168 - val_loss: 0.1096 - val_mean_squared_error: 0.1096\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1833 - mean_squared_error: 0.1833\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1168 - mean_squared_error: 0.1168 - val_loss: 0.1096 - val_mean_squared_error: 0.1096\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1833 - mean_squared_error: 0.1833\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1240 - mean_squared_error: 0.1240\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1240 - mean_squared_error: 0.1240\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1136 - mean_squared_error: 0.1136\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1136 - mean_squared_error: 0.1136\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1122 - mean_squared_error: 0.1122 - val_loss: 0.0391 - val_mean_squared_error: 0.0391\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1208 - mean_squared_error: 0.1208\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1122 - mean_squared_error: 0.1122 - val_loss: 0.0391 - val_mean_squared_error: 0.0391\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1208 - mean_squared_error: 0.1208\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1276 - mean_squared_error: 0.1276\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1276 - mean_squared_error: 0.1276\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 5120/8253 [=================>............] - ETA: 0s - loss: 0.1227 - mean_squared_error: 0.1227\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 5120/8253 [=================>............] - ETA: 0s - loss: 0.1227 - mean_squared_error: 0.1227\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1186 - mean_squared_error: 0.1186\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1186 - mean_squared_error: 0.1186\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1166 - mean_squared_error: 0.1166 - val_loss: 0.1924 - val_mean_squared_error: 0.1924\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.2109 - mean_squared_error: 0.2109\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1166 - mean_squared_error: 0.1166 - val_loss: 0.1924 - val_mean_squared_error: 0.1924\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.2109 - mean_squared_error: 0.2109\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1509 - mean_squared_error: 0.1509\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1509 - mean_squared_error: 0.1509\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1291 - mean_squared_error: 0.1291\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1291 - mean_squared_error: 0.1291\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1235 - mean_squared_error: 0.12\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1232 - mean_squared_error: 0.1232 - val_loss: 0.0370 - val_mean_squared_error: 0.0370\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1235 - mean_squared_error: 0.12\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1232 - mean_squared_error: 0.1232 - val_loss: 0.0370 - val_mean_squared_error: 0.0370\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1046 - mean_squared_error: 0.1046\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1046 - mean_squared_error: 0.1046\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1043 - mean_squared_error: 0.1043\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1043 - mean_squared_error: 0.1043\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 5120/8253 [=================>............] - ETA: 0s - loss: 0.1047 - mean_squared_error: 0.1047\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 5120/8253 [=================>............] - ETA: 0s - loss: 0.1047 - mean_squared_error: 0.1047\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1038 - mean_squared_error: 0.10\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1040 - mean_squared_error: 0.1040 - val_loss: 0.0512 - val_mean_squared_error: 0.0512\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Train on 8253 samples, validate on 1029 samples\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1038 - mean_squared_error: 0.10\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1040 - mean_squared_error: 0.1040 - val_loss: 0.0512 - val_mean_squared_error: 0.0512\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Train on 8253 samples, validate on 1029 samples\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1463 - mean_squared_error: 0.1463\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1463 - mean_squared_error: 0.1463\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1244 - mean_squared_error: 0.1244\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1244 - mean_squared_error: 0.1244\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1182 - mean_squared_error: 0.1182\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1182 - mean_squared_error: 0.1182\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1167 - mean_squared_error: 0.1167 - val_loss: 0.0406 - val_mean_squared_error: 0.0406\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1316 - mean_squared_error: 0.1316\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 27us/step - loss: 0.1167 - mean_squared_error: 0.1167 - val_loss: 0.0406 - val_mean_squared_error: 0.0406\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1316 - mean_squared_error: 0.1316\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1280 - mean_squared_error: 0.1280\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1280 - mean_squared_error: 0.1280\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1178 - mean_squared_error: 0.1178\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1178 - mean_squared_error: 0.1178\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1155 - mean_squared_error: 0.1155 - val_loss: 0.0401 - val_mean_squared_error: 0.0401\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1155 - mean_squared_error: 0.1155 - val_loss: 0.0401 - val_mean_squared_error: 0.0401\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1412 - mean_squared_error: 0.1412\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1412 - mean_squared_error: 0.1412\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1237 - mean_squared_error: 0.1237\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 3072/8253 [==========>...................] - ETA: 0s - loss: 0.1237 - mean_squared_error: 0.1237\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 5120/8253 [=================>............] - ETA: 0s - loss: 0.1180 - mean_squared_error: 0.1180\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 5120/8253 [=================>............] - ETA: 0s - loss: 0.1180 - mean_squared_error: 0.1180\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1102 - mean_squared_error: 0.1102\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1102 - mean_squared_error: 0.11\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 29us/step - loss: 0.1109 - mean_squared_error: 0.1109 - val_loss: 0.0486 - val_mean_squared_error: 0.0486\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 29us/step - loss: 0.1109 - mean_squared_error: 0.1109 - val_loss: 0.0486 - val_mean_squared_error: 0.0486\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1275 - mean_squared_error: 0.1275\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1275 - mean_squared_error: 0.1275\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1213 - mean_squared_error: 0.1213\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1213 - mean_squared_error: 0.1213\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1164 - mean_squared_error: 0.1164\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 6144/8253 [=====================>........] - ETA: 0s - loss: 0.1164 - mean_squared_error: 0.1164\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1133 - mean_squared_error: 0.1133\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8192/8253 [============================>.] - ETA: 0s - loss: 0.1133 - mean_squared_error: 0.11\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1131 - mean_squared_error: 0.1131 - val_loss: 0.0905 - val_mean_squared_error: 0.0905\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1248 - mean_squared_error: 0.1248\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 26us/step - loss: 0.1131 - mean_squared_error: 0.1131 - val_loss: 0.0905 - val_mean_squared_error: 0.0905\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 1024/8253 [==>...........................] - ETA: 0s - loss: 0.1248 - mean_squared_error: 0.1248\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1186 - mean_squared_error: 0.1186\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 4096/8253 [=============>................] - ETA: 0s - loss: 0.1186 - mean_squared_error: 0.1186\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1112 - mean_squared_error: 0.1112\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 7168/8253 [=========================>....] - ETA: 0s - loss: 0.1112 - mean_squared_error: 0.11\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 28us/step - loss: 0.1094 - mean_squared_error: 0.1094 - val_loss: 0.0332 - val_mean_squared_error: 0.0332\n",
"== Status ==\n",
"Using FIFO scheduling algorithm.\n",
"Resources requested: 0/4 CPUs, 0/0 GPUs (0/1.0 ps, 0/1.0 trainer)\n",
"Memory usage on this node: 29.4/67.4 GB\n",
"Result logdir: /home/shan/ray_results/automl\n",
"Number of trials: 1 ({'TERMINATED': 1})\n",
"TERMINATED trials:\n",
" - train_func_0_batch_size=1024,dropout=0.48112,dropout_1=0.26833,dropout_2=0.39814,latent_dim=32,lr=0.0080377,lstm_1_units=64,lstm_2_units=8,past_seq_len=3,selected_features=['HOUR(datetime)' 'IS_BUSY_HOURS(datetime)' 'WEEKDAY(datetime)']:\tTERMINATED, [2 CPUs, 0 GPUs], [pid=7207], 10 s, 5 iter\n",
"\n",
"The best configurations are:\n",
"selected_features : ['HOUR(datetime)' 'IS_BUSY_HOURS(datetime)' 'WEEKDAY(datetime)']\n",
"lstm_1_units : 64\n",
"dropout_1 : 0.26833423376715093\n",
"lstm_2_units : 8\n",
"dropout_2 : 0.39814194762629107\n",
"latent_dim : 32\n",
"dropout : 0.48112037684170644\n",
"lr : 0.00803765177979179\n",
"batch_size : 1024\n",
"epochs : 5\n",
"past_seq_len : 3\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7207)\u001b[0m 8253/8253 [==============================] - 0s 28us/step - loss: 0.1094 - mean_squared_error: 0.1094 - val_loss: 0.0332 - val_mean_squared_error: 0.0332\n",
"Training completed.\n",
"CPU times: user 3.62 s, sys: 147 ms, total: 3.76 s\n",
"Wall time: 15.4 s\n"
]
}
],
"source": [
"%%time\n",
"# fit train_df and validate with val_df, return the best trial as pipeline.\n",
"# the default recipe is SmokeRecipe,which runs one epoch and one iteration with only 1 random sample.\n",
"# you can change recipe by define `recipe` in `fit`. The recipes you can choose are SmokeRecipe, RandomRecipe, GridRandomRecipe and BayesRecipe.\n",
"pipeline = tsp.fit(train_df,\n",
" validation_df=val_df,\n",
" metric=\"mse\",\n",
" recipe=RandomRecipe(look_back=(2, 4)))\n",
"print(\"Training completed.\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 3. Test"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"# predict test_df with the best trial\n",
"pred_df = pipeline.predict(test_df)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" datetime | \n",
" value | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 2015-01-10 13:30:00 | \n",
" 22192.162109 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2015-01-10 14:00:00 | \n",
" 21757.605469 | \n",
"
\n",
" \n",
" | 2 | \n",
" 2015-01-10 14:30:00 | \n",
" 20591.683594 | \n",
"
\n",
" \n",
" | 3 | \n",
" 2015-01-10 15:00:00 | \n",
" 21799.269531 | \n",
"
\n",
" \n",
" | 4 | \n",
" 2015-01-10 15:30:00 | \n",
" 22432.207031 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" datetime | \n",
" value | \n",
"
\n",
" \n",
" \n",
" \n",
" | 50 | \n",
" 2015-01-11 13:00:00 | \n",
" 21296 | \n",
"
\n",
" \n",
" | 51 | \n",
" 2015-01-11 13:30:00 | \n",
" 20381 | \n",
"
\n",
" \n",
" | 52 | \n",
" 2015-01-11 14:00:00 | \n",
" 19508 | \n",
"
\n",
" \n",
" | 53 | \n",
" 2015-01-11 14:30:00 | \n",
" 19210 | \n",
"
\n",
" \n",
" | 54 | \n",
" 2015-01-11 15:00:00 | \n",
" 18255 | \n",
"
\n",
" \n",
"
\n",
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"# plot the predicted values and actual values\n",
"plot_result(test_df, pred_df)"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Evaluate: the mean square error is 1921356.3949139877\n",
"Evaluate: the smape value is 6.465159906962593\n"
]
}
],
"source": [
"# evaluate test_df\n",
"mse, smape = pipeline.evaluate(test_df, metrics=[\"mse\", \"smape\"])\n",
"print(\"Evaluate: the mean square error is\", mse)\n",
"print(\"Evaluate: the smape value is\", smape)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 4. save and restore"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We provide save and restore interface to save the pipeline with the best trial for easily rebuilding."
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Pipeline is saved in /tmp/saved_pipeline/my.ppl\n"
]
},
{
"data": {
"text/plain": [
"'/tmp/saved_pipeline/my.ppl'"
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# save the pipeline with best trial\n",
"pipeline.save(\"/tmp/saved_pipeline/my.ppl\")"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Restore pipeline from /tmp/saved_pipeline/my.ppl\n"
]
}
],
"source": [
"from zoo.automl.pipeline.time_sequence import load_ts_pipeline\n",
"new_pipeline = load_ts_pipeline(\"/tmp/saved_pipeline/my.ppl\")"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [],
"source": [
"# you can do predict and evaluate again\n",
"# we use test_df as input in order to compare results before and after restoration \n",
"new_pred = new_pipeline.predict(test_df)"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" datetime | \n",
" value | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 2015-01-10 13:30:00 | \n",
" 22192.162109 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2015-01-10 14:00:00 | \n",
" 21757.605469 | \n",
"
\n",
" \n",
" | 2 | \n",
" 2015-01-10 14:30:00 | \n",
" 20591.683594 | \n",
"
\n",
" \n",
" | 3 | \n",
" 2015-01-10 15:00:00 | \n",
" 21799.269531 | \n",
"
\n",
" \n",
" | 4 | \n",
" 2015-01-10 15:30:00 | \n",
" 22432.207031 | \n",
"
\n",
" \n",
"
\n",
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"# predict results of test_df\n",
"new_pred_df = new_pipeline.predict(test_df)\n",
"plot_result(test_df, new_pred_df)"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Evaluate: the mean square error is 1972492.7909922532\n",
"Evaluate: the smape value is 6.915714915664744\n"
]
}
],
"source": [
"# evaluate test_df\n",
"mse, smape = new_pipeline.evaluate(test_df, metrics=[\"mse\", \"smape\"])\n",
"print(\"Evaluate: the mean square error is\", mse)\n",
"print(\"Evaluate: the smape value is\", smape)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 6. multi step forecasting "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"You can do multi step forecasting by simply changing the `future_seq_len` option while creating a new `TimeSequencePredictor` object."
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [],
"source": [
"# build time sequence predictor\n",
"from zoo.automl.regression.time_sequence_predictor import *\n",
"\n",
"# change future_seq_len into the step you want to forcast.\n",
"tsp = TimeSequencePredictor(future_seq_len=5,\n",
" dt_col=\"datetime\",\n",
" target_col=\"value\",\n",
" extra_features_col=None)"
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"2019-12-18 15:49:28,951\tINFO tune.py:60 -- Tip: to resume incomplete experiments, pass resume='prompt' or resume=True to run()\n",
"2019-12-18 15:49:28,952\tINFO tune.py:223 -- Starting a new experiment.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model selection: LSTM Seq2Seq model is selected.\n",
"== Status ==\n",
"Using FIFO scheduling algorithm.\n",
"Resources requested: 0/4 CPUs, 0/0 GPUs (0/1.0 ps, 0/1.0 trainer)\n",
"Memory usage on this node: 29.3/67.4 GB\n",
"\n",
"== Status ==\n",
"Using FIFO scheduling algorithm.\n",
"Resources requested: 2/4 CPUs, 0/0 GPUs (0/1.0 ps, 0/1.0 trainer)\n",
"Memory usage on this node: 29.3/67.4 GB\n",
"Result logdir: /home/shan/ray_results/automl\n",
"Number of trials: 1 ({'RUNNING': 1})\n",
"RUNNING trials:\n",
" - train_func_0_batch_size=1024,dropout=0.21806,dropout_1=0.49144,dropout_2=0.45671,latent_dim=32,lr=0.0086638,lstm_1_units=64,lstm_2_units=128,selected_features=['WEEKDAY(datetime)' 'HOUR(datetime)' 'IS_AWAKE(datetime)'\n",
" 'IS_WEEKEND(datetime)']:\tRUNNING\n",
"\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Model selection: LSTM Seq2Seq model is selected.\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 2019-12-18 15:49:29,521\tWARNING worker.py:204 -- Calling ray.get or ray.wait in a separate thread may lead to deadlock if the main thread blocks on this thread and there are not enough resources to execute more tasks\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Model selection: LSTM Seq2Seq model is selected.\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 2019-12-18 15:49:29,521\tWARNING worker.py:204 -- Calling ray.get or ray.wait in a separate thread may lead to deadlock if the main thread blocks on this thread and there are not enough resources to execute more tasks\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 2019-12-18 15:49:33.583746: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 2019-12-18 15:49:33.583746: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 6s - loss: 0.8777 - mean_squared_error: 0.8777\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 6s - loss: 0.8777 - mean_squared_error: 0.8777\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 1s - loss: 0.5894 - mean_squared_error: 0.5894\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 1s - loss: 0.5894 - mean_squared_error: 0.5894\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 1s - loss: 0.5197 - mean_squared_error: 0.5197\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 1s - loss: 0.5197 - mean_squared_error: 0.5197\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.4761 - mean_squared_error: 0.4761\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.4761 - mean_squared_error: 0.4761\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.4626 - mean_squared_error: 0.4626\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.4626 - mean_squared_error: 0.4626\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.4102 - mean_squared_error: 0.4102\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.4102 - mean_squared_error: 0.4102\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 1s 180us/step - loss: 0.4089 - mean_squared_error: 0.4089 - val_loss: 0.1493 - val_mean_squared_error: 0.1493\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 1s 180us/step - loss: 0.4089 - mean_squared_error: 0.4089 - val_loss: 0.1493 - val_mean_squared_error: 0.1493\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.2543 - mean_squared_error: 0.2543\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.2543 - mean_squared_error: 0.2543\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.2184 - mean_squared_error: 0.2184\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.2184 - mean_squared_error: 0.2184\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.2055 - mean_squared_error: 0.2055\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.2055 - mean_squared_error: 0.2055\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 21us/step - loss: 0.2056 - mean_squared_error: 0.2056 - val_loss: 0.2469 - val_mean_squared_error: 0.2469\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.1994 - mean_squared_error: 0.1994\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 21us/step - loss: 0.2056 - mean_squared_error: 0.2056 - val_loss: 0.2469 - val_mean_squared_error: 0.2469\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.1994 - mean_squared_error: 0.1994\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.1793 - mean_squared_error: 0.1793\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.1793 - mean_squared_error: 0.1793\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.1748 - mean_squared_error: 0.1748\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.1748 - mean_squared_error: 0.1748\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.1620 - mean_squared_error: 0.1620\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.1620 - mean_squared_error: 0.1620\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.1618 - mean_squared_error: 0.1618 - val_loss: 0.1451 - val_mean_squared_error: 0.1451\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.1403 - mean_squared_error: 0.1403\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.1618 - mean_squared_error: 0.1618 - val_loss: 0.1451 - val_mean_squared_error: 0.1451\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.1403 - mean_squared_error: 0.1403\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.1295 - mean_squared_error: 0.1295\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.1295 - mean_squared_error: 0.1295\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.1408 - mean_squared_error: 0.1408\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.1408 - mean_squared_error: 0.1408\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.1362 - mean_squared_error: 0.13\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 28us/step - loss: 0.1363 - mean_squared_error: 0.1363 - val_loss: 0.1758 - val_mean_squared_error: 0.1758\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.1362 - mean_squared_error: 0.13\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 28us/step - loss: 0.1363 - mean_squared_error: 0.1363 - val_loss: 0.1758 - val_mean_squared_error: 0.1758\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.1291 - mean_squared_error: 0.1291\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.1291 - mean_squared_error: 0.1291\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.1414 - mean_squared_error: 0.1414\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.1414 - mean_squared_error: 0.1414\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.1385 - mean_squared_error: 0.1385\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.1385 - mean_squared_error: 0.1385\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.1223 - mean_squared_error: 0.12\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.1219 - mean_squared_error: 0.1219 - val_loss: 0.0901 - val_mean_squared_error: 0.0901\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m WARNING:tensorflow:Layer decoder_lstm was passed non-serializable keyword arguments: {'training': None, 'initial_state': [, ]}. They will not be included in the serialized model (and thus will be missing at deserialization time).\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.1223 - mean_squared_error: 0.12\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.1219 - mean_squared_error: 0.1219 - val_loss: 0.0901 - val_mean_squared_error: 0.0901\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m WARNING:tensorflow:Layer decoder_lstm was passed non-serializable keyword arguments: {'training': None, 'initial_state': [, ]}. They will not be included in the serialized model (and thus will be missing at deserialization time).\n",
"== Status ==\n",
"Using FIFO scheduling algorithm.\n",
"Resources requested: 2/4 CPUs, 0/0 GPUs (0/1.0 ps, 0/1.0 trainer)\n",
"Memory usage on this node: 29.5/67.4 GB\n",
"Result logdir: /home/shan/ray_results/automl\n",
"Number of trials: 1 ({'RUNNING': 1})\n",
"RUNNING trials:\n",
" - train_func_0_batch_size=1024,dropout=0.21806,dropout_1=0.49144,dropout_2=0.45671,latent_dim=32,lr=0.0086638,lstm_1_units=64,lstm_2_units=128,selected_features=['WEEKDAY(datetime)' 'HOUR(datetime)' 'IS_AWAKE(datetime)'\n",
" 'IS_WEEKEND(datetime)']:\tRUNNING, [2 CPUs, 0 GPUs], [pid=7230], 6 s, 1 iter\n",
"\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0920 - mean_squared_error: 0.0920\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0920 - mean_squared_error: 0.0920\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.1052 - mean_squared_error: 0.1052\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.1052 - mean_squared_error: 0.1052\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.1129 - mean_squared_error: 0.1129\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.1129 - mean_squared_error: 0.1129\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.1088 - mean_squared_error: 0.1088 - val_loss: 0.0872 - val_mean_squared_error: 0.0872\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0893 - mean_squared_error: 0.0893\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.1088 - mean_squared_error: 0.1088 - val_loss: 0.0872 - val_mean_squared_error: 0.0872\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0893 - mean_squared_error: 0.0893\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0802 - mean_squared_error: 0.0802\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0802 - mean_squared_error: 0.0802\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0819 - mean_squared_error: 0.0819\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0819 - mean_squared_error: 0.0819\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0906 - mean_squared_error: 0.0906 - val_loss: 0.1481 - val_mean_squared_error: 0.1481\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.1385 - mean_squared_error: 0.1385\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0906 - mean_squared_error: 0.0906 - val_loss: 0.1481 - val_mean_squared_error: 0.1481\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.1385 - mean_squared_error: 0.1385\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0907 - mean_squared_error: 0.0907\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0907 - mean_squared_error: 0.0907\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0816 - mean_squared_error: 0.0816\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0816 - mean_squared_error: 0.0816\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 22us/step - loss: 0.0815 - mean_squared_error: 0.0815 - val_loss: 0.1598 - val_mean_squared_error: 0.1598\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 22us/step - loss: 0.0815 - mean_squared_error: 0.0815 - val_loss: 0.1598 - val_mean_squared_error: 0.1598\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.1915 - mean_squared_error: 0.1915\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.1915 - mean_squared_error: 0.1915\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.1312 - mean_squared_error: 0.1312\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.1312 - mean_squared_error: 0.1312\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.1009 - mean_squared_error: 0.1009\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.1009 - mean_squared_error: 0.1009\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 22us/step - loss: 0.0929 - mean_squared_error: 0.0929 - val_loss: 0.0622 - val_mean_squared_error: 0.0622\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0688 - mean_squared_error: 0.0688\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 22us/step - loss: 0.0929 - mean_squared_error: 0.0929 - val_loss: 0.0622 - val_mean_squared_error: 0.0622\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0688 - mean_squared_error: 0.0688\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0643 - mean_squared_error: 0.0643\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0643 - mean_squared_error: 0.0643\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0688 - mean_squared_error: 0.0688\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0688 - mean_squared_error: 0.0688\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0806 - mean_squared_error: 0.0806 - val_loss: 0.0501 - val_mean_squared_error: 0.0501\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m WARNING:tensorflow:Layer decoder_lstm was passed non-serializable keyword arguments: {'training': None, 'initial_state': [, ]}. They will not be included in the serialized model (and thus will be missing at deserialization time).\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0806 - mean_squared_error: 0.0806 - val_loss: 0.0501 - val_mean_squared_error: 0.0501\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m WARNING:tensorflow:Layer decoder_lstm was passed non-serializable keyword arguments: {'training': None, 'initial_state': [, ]}. They will not be included in the serialized model (and thus will be missing at deserialization time).\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0748 - mean_squared_error: 0.0748\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0748 - mean_squared_error: 0.0748\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0679 - mean_squared_error: 0.0679\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0679 - mean_squared_error: 0.0679\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0650 - mean_squared_error: 0.0650\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0650 - mean_squared_error: 0.0650\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0652 - mean_squared_error: 0.0652 - val_loss: 0.1480 - val_mean_squared_error: 0.1480\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0988 - mean_squared_error: 0.0988\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0652 - mean_squared_error: 0.0652 - val_loss: 0.1480 - val_mean_squared_error: 0.1480\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0988 - mean_squared_error: 0.0988\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0799 - mean_squared_error: 0.0799\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0799 - mean_squared_error: 0.0799\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0700 - mean_squared_error: 0.0700\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0700 - mean_squared_error: 0.0700\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.0688 - mean_squared_error: 0.0688 - val_loss: 0.1066 - val_mean_squared_error: 0.1066\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0656 - mean_squared_error: 0.0656\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.0688 - mean_squared_error: 0.0688 - val_loss: 0.1066 - val_mean_squared_error: 0.1066\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0656 - mean_squared_error: 0.0656\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0744 - mean_squared_error: 0.0744\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0744 - mean_squared_error: 0.0744\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0700 - mean_squared_error: 0.0700\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0700 - mean_squared_error: 0.0700\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0681 - mean_squared_error: 0.0681\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0681 - mean_squared_error: 0.0681\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 31us/step - loss: 0.0682 - mean_squared_error: 0.0682 - val_loss: 0.0751 - val_mean_squared_error: 0.0751\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 31us/step - loss: 0.0682 - mean_squared_error: 0.0682 - val_loss: 0.0751 - val_mean_squared_error: 0.0751\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0967 - mean_squared_error: 0.0967\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0967 - mean_squared_error: 0.0967\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0888 - mean_squared_error: 0.0888\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0888 - mean_squared_error: 0.0888\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0762 - mean_squared_error: 0.0762\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0762 - mean_squared_error: 0.0762\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0686 - mean_squared_error: 0.0686\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0686 - mean_squared_error: 0.0686\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 36us/step - loss: 0.0657 - mean_squared_error: 0.0657 - val_loss: 0.0434 - val_mean_squared_error: 0.0434\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0625 - mean_squared_error: 0.0625\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 36us/step - loss: 0.0657 - mean_squared_error: 0.0657 - val_loss: 0.0434 - val_mean_squared_error: 0.0434\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0625 - mean_squared_error: 0.0625\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0640 - mean_squared_error: 0.0640\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0640 - mean_squared_error: 0.0640\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0651 - mean_squared_error: 0.0651\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0651 - mean_squared_error: 0.0651\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0661 - mean_squared_error: 0.0661\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0661 - mean_squared_error: 0.0661\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 36us/step - loss: 0.0644 - mean_squared_error: 0.0644 - val_loss: 0.1157 - val_mean_squared_error: 0.1157\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 36us/step - loss: 0.0644 - mean_squared_error: 0.0644 - val_loss: 0.1157 - val_mean_squared_error: 0.1157\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0674 - mean_squared_error: 0.0674\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0674 - mean_squared_error: 0.0674\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0601 - mean_squared_error: 0.0601\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0601 - mean_squared_error: 0.0601\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0581 - mean_squared_error: 0.0581\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0581 - mean_squared_error: 0.0581\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0588 - mean_squared_error: 0.0588\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0588 - mean_squared_error: 0.0588\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 40us/step - loss: 0.0573 - mean_squared_error: 0.0573 - val_loss: 0.0565 - val_mean_squared_error: 0.0565\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 40us/step - loss: 0.0573 - mean_squared_error: 0.0573 - val_loss: 0.0565 - val_mean_squared_error: 0.0565\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0501 - mean_squared_error: 0.0501\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0501 - mean_squared_error: 0.0501\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0464 - mean_squared_error: 0.0464\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0464 - mean_squared_error: 0.0464\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0500 - mean_squared_error: 0.0500\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0500 - mean_squared_error: 0.0500\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0541 - mean_squared_error: 0.0541\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0541 - mean_squared_error: 0.0541\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 35us/step - loss: 0.0549 - mean_squared_error: 0.0549 - val_loss: 0.0666 - val_mean_squared_error: 0.0666\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 35us/step - loss: 0.0549 - mean_squared_error: 0.0549 - val_loss: 0.0666 - val_mean_squared_error: 0.0666\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0484 - mean_squared_error: 0.0484\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0484 - mean_squared_error: 0.0484\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 2048/8242 [======>.......................] - ETA: 0s - loss: 0.0472 - mean_squared_error: 0.0472\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 2048/8242 [======>.......................] - ETA: 0s - loss: 0.0472 - mean_squared_error: 0.0472\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0512 - mean_squared_error: 0.0512\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0512 - mean_squared_error: 0.0512\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0550 - mean_squared_error: 0.0550\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0550 - mean_squared_error: 0.0550\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0570 - mean_squared_error: 0.05\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 40us/step - loss: 0.0570 - mean_squared_error: 0.0570 - val_loss: 0.0674 - val_mean_squared_error: 0.0674\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0570 - mean_squared_error: 0.05\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 40us/step - loss: 0.0570 - mean_squared_error: 0.0570 - val_loss: 0.0674 - val_mean_squared_error: 0.0674\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0458 - mean_squared_error: 0.0458\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0458 - mean_squared_error: 0.0458\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0450 - mean_squared_error: 0.0450\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0450 - mean_squared_error: 0.0450\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0483 - mean_squared_error: 0.0483\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0483 - mean_squared_error: 0.0483\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0492 - mean_squared_error: 0.0492 - val_loss: 0.1001 - val_mean_squared_error: 0.1001\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0492 - mean_squared_error: 0.0492 - val_loss: 0.1001 - val_mean_squared_error: 0.1001\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0539 - mean_squared_error: 0.0539\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0539 - mean_squared_error: 0.0539\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0507 - mean_squared_error: 0.0507\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0507 - mean_squared_error: 0.0507\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0498 - mean_squared_error: 0.0498\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0498 - mean_squared_error: 0.0498\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0487 - mean_squared_error: 0.0487\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0487 - mean_squared_error: 0.0487\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 39us/step - loss: 0.0498 - mean_squared_error: 0.0498 - val_loss: 0.0607 - val_mean_squared_error: 0.0607\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 39us/step - loss: 0.0498 - mean_squared_error: 0.0498 - val_loss: 0.0607 - val_mean_squared_error: 0.0607\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0411 - mean_squared_error: 0.0411\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0411 - mean_squared_error: 0.0411\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0413 - mean_squared_error: 0.0413\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0413 - mean_squared_error: 0.0413\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0431 - mean_squared_error: 0.0431\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0431 - mean_squared_error: 0.0431\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0467 - mean_squared_error: 0.0467\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0467 - mean_squared_error: 0.0467\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0483 - mean_squared_error: 0.0483\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 37us/step - loss: 0.0483 - mean_squared_error: 0.0483 - val_loss: 0.0675 - val_mean_squared_error: 0.0675\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0483 - mean_squared_error: 0.04\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 37us/step - loss: 0.0483 - mean_squared_error: 0.0483 - val_loss: 0.0675 - val_mean_squared_error: 0.0675\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0550 - mean_squared_error: 0.0550\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0550 - mean_squared_error: 0.0550\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0446 - mean_squared_error: 0.0446\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0446 - mean_squared_error: 0.0446\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0434 - mean_squared_error: 0.0434\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0434 - mean_squared_error: 0.0434\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0417 - mean_squared_error: 0.0417\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0417 - mean_squared_error: 0.0417\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 34us/step - loss: 0.0426 - mean_squared_error: 0.0426 - val_loss: 0.0930 - val_mean_squared_error: 0.0930\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0484 - mean_squared_error: 0.0484\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 34us/step - loss: 0.0426 - mean_squared_error: 0.0426 - val_loss: 0.0930 - val_mean_squared_error: 0.0930\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0484 - mean_squared_error: 0.0484\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0440 - mean_squared_error: 0.0440\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0440 - mean_squared_error: 0.0440\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0439 - mean_squared_error: 0.0439\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0439 - mean_squared_error: 0.0439\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0443 - mean_squared_error: 0.0443\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0443 - mean_squared_error: 0.0443\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0461 - mean_squared_error: 0.0461\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0461 - mean_squared_error: 0.0461\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 37us/step - loss: 0.0461 - mean_squared_error: 0.0461 - val_loss: 0.0844 - val_mean_squared_error: 0.0844\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 37us/step - loss: 0.0461 - mean_squared_error: 0.0461 - val_loss: 0.0844 - val_mean_squared_error: 0.0844\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0531 - mean_squared_error: 0.0531\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0531 - mean_squared_error: 0.0531\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0479 - mean_squared_error: 0.0479\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0479 - mean_squared_error: 0.0479\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0436 - mean_squared_error: 0.0436\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0436 - mean_squared_error: 0.0436\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0416 - mean_squared_error: 0.0416\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0416 - mean_squared_error: 0.0416\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 37us/step - loss: 0.0411 - mean_squared_error: 0.0411 - val_loss: 0.0816 - val_mean_squared_error: 0.0816\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0463 - mean_squared_error: 0.0463\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 37us/step - loss: 0.0411 - mean_squared_error: 0.0411 - val_loss: 0.0816 - val_mean_squared_error: 0.0816\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0463 - mean_squared_error: 0.0463\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0446 - mean_squared_error: 0.0446\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0446 - mean_squared_error: 0.0446\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0450 - mean_squared_error: 0.0450\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0450 - mean_squared_error: 0.0450\n",
"== Status ==\n",
"Using FIFO scheduling algorithm.\n",
"Resources requested: 2/4 CPUs, 0/0 GPUs (0/1.0 ps, 0/1.0 trainer)\n",
"Memory usage on this node: 29.5/67.4 GB\n",
"Result logdir: /home/shan/ray_results/automl\n",
"Number of trials: 1 ({'RUNNING': 1})\n",
"RUNNING trials:\n",
" - train_func_0_batch_size=1024,dropout=0.21806,dropout_1=0.49144,dropout_2=0.45671,latent_dim=32,lr=0.0086638,lstm_1_units=64,lstm_2_units=128,selected_features=['WEEKDAY(datetime)' 'HOUR(datetime)' 'IS_AWAKE(datetime)'\n",
" 'IS_WEEKEND(datetime)']:\tRUNNING, [2 CPUs, 0 GPUs], [pid=7230], 11 s, 5 iter\n",
"\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0438 - mean_squared_error: 0.04\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 26us/step - loss: 0.0437 - mean_squared_error: 0.0437 - val_loss: 0.0625 - val_mean_squared_error: 0.0625\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0438 - mean_squared_error: 0.04\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 26us/step - loss: 0.0437 - mean_squared_error: 0.0437 - val_loss: 0.0625 - val_mean_squared_error: 0.0625\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0384 - mean_squared_error: 0.0384\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0384 - mean_squared_error: 0.0384\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0349 - mean_squared_error: 0.0349\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0349 - mean_squared_error: 0.0349\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0374 - mean_squared_error: 0.0374\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0374 - mean_squared_error: 0.0374\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0403 - mean_squared_error: 0.0403 - val_loss: 0.0803 - val_mean_squared_error: 0.0803\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0529 - mean_squared_error: 0.0529\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0403 - mean_squared_error: 0.0403 - val_loss: 0.0803 - val_mean_squared_error: 0.0803\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0529 - mean_squared_error: 0.0529\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0399 - mean_squared_error: 0.0399\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0399 - mean_squared_error: 0.0399\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0372 - mean_squared_error: 0.0372\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0372 - mean_squared_error: 0.03\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0368 - mean_squared_error: 0.0368 - val_loss: 0.0616 - val_mean_squared_error: 0.0616\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0368 - mean_squared_error: 0.0368 - val_loss: 0.0616 - val_mean_squared_error: 0.0616\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0387 - mean_squared_error: 0.0387\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0387 - mean_squared_error: 0.0387\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0369 - mean_squared_error: 0.0369\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0369 - mean_squared_error: 0.0369\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0387 - mean_squared_error: 0.0387\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0387 - mean_squared_error: 0.0387\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.0400 - mean_squared_error: 0.0400 - val_loss: 0.0527 - val_mean_squared_error: 0.0527\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.0400 - mean_squared_error: 0.0400 - val_loss: 0.0527 - val_mean_squared_error: 0.0527\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0363 - mean_squared_error: 0.0363\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0363 - mean_squared_error: 0.0363\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0388 - mean_squared_error: 0.0388\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0388 - mean_squared_error: 0.0388\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0387 - mean_squared_error: 0.0387\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0387 - mean_squared_error: 0.0387\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0376 - mean_squared_error: 0.0376 - val_loss: 0.0794 - val_mean_squared_error: 0.0794\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0376 - mean_squared_error: 0.0376 - val_loss: 0.0794 - val_mean_squared_error: 0.0794\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0381 - mean_squared_error: 0.0381\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0381 - mean_squared_error: 0.0381\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0371 - mean_squared_error: 0.0371\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0371 - mean_squared_error: 0.0371\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0365 - mean_squared_error: 0.0365\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0365 - mean_squared_error: 0.03\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0379 - mean_squared_error: 0.0379 - val_loss: 0.0698 - val_mean_squared_error: 0.0698\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0379 - mean_squared_error: 0.0379 - val_loss: 0.0698 - val_mean_squared_error: 0.0698\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0351 - mean_squared_error: 0.0351\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0351 - mean_squared_error: 0.0351\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0347 - mean_squared_error: 0.0347\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0347 - mean_squared_error: 0.0347\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0342 - mean_squared_error: 0.0342\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0342 - mean_squared_error: 0.0342\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0348 - mean_squared_error: 0.0348 - val_loss: 0.0547 - val_mean_squared_error: 0.0547\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0348 - mean_squared_error: 0.0348 - val_loss: 0.0547 - val_mean_squared_error: 0.0547\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0582 - mean_squared_error: 0.0582\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0582 - mean_squared_error: 0.0582\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0427 - mean_squared_error: 0.0427\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0427 - mean_squared_error: 0.0427\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0378 - mean_squared_error: 0.0378\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0378 - mean_squared_error: 0.0378\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0361 - mean_squared_error: 0.0361 - val_loss: 0.0722 - val_mean_squared_error: 0.0722\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0397 - mean_squared_error: 0.0397\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0361 - mean_squared_error: 0.0361 - val_loss: 0.0722 - val_mean_squared_error: 0.0722\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0397 - mean_squared_error: 0.0397\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0370 - mean_squared_error: 0.0370\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0370 - mean_squared_error: 0.0370\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0330 - mean_squared_error: 0.0330\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0330 - mean_squared_error: 0.0330\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0316 - mean_squared_error: 0.03\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 27us/step - loss: 0.0316 - mean_squared_error: 0.0316 - val_loss: 0.0833 - val_mean_squared_error: 0.0833\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0316 - mean_squared_error: 0.03\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 27us/step - loss: 0.0316 - mean_squared_error: 0.0316 - val_loss: 0.0833 - val_mean_squared_error: 0.0833\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0484 - mean_squared_error: 0.0484\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0484 - mean_squared_error: 0.0484\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0407 - mean_squared_error: 0.0407\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0407 - mean_squared_error: 0.0407\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0342 - mean_squared_error: 0.0342\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0342 - mean_squared_error: 0.0342\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0325 - mean_squared_error: 0.0325 - val_loss: 0.0541 - val_mean_squared_error: 0.0541\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0386 - mean_squared_error: 0.0386\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0325 - mean_squared_error: 0.0325 - val_loss: 0.0541 - val_mean_squared_error: 0.0541\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0386 - mean_squared_error: 0.0386\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0442 - mean_squared_error: 0.0442\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0442 - mean_squared_error: 0.0442\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0391 - mean_squared_error: 0.0391\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0391 - mean_squared_error: 0.03\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 22us/step - loss: 0.0385 - mean_squared_error: 0.0385 - val_loss: 0.0612 - val_mean_squared_error: 0.0612\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 22us/step - loss: 0.0385 - mean_squared_error: 0.0385 - val_loss: 0.0612 - val_mean_squared_error: 0.0612\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0562 - mean_squared_error: 0.0562\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0562 - mean_squared_error: 0.0562\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0452 - mean_squared_error: 0.0452\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0452 - mean_squared_error: 0.0452\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0389 - mean_squared_error: 0.0389\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0389 - mean_squared_error: 0.0389\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0353 - mean_squared_error: 0.03\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.0352 - mean_squared_error: 0.0352 - val_loss: 0.0642 - val_mean_squared_error: 0.0642\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0353 - mean_squared_error: 0.03\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.0352 - mean_squared_error: 0.0352 - val_loss: 0.0642 - val_mean_squared_error: 0.0642\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0368 - mean_squared_error: 0.0368\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0368 - mean_squared_error: 0.0368\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0326 - mean_squared_error: 0.0326\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0326 - mean_squared_error: 0.0326\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0321 - mean_squared_error: 0.0321\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0321 - mean_squared_error: 0.0321\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0308 - mean_squared_error: 0.0308\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0308 - mean_squared_error: 0.03\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 31us/step - loss: 0.0308 - mean_squared_error: 0.0308 - val_loss: 0.0530 - val_mean_squared_error: 0.0530\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 31us/step - loss: 0.0308 - mean_squared_error: 0.0308 - val_loss: 0.0530 - val_mean_squared_error: 0.0530\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0422 - mean_squared_error: 0.0422\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0422 - mean_squared_error: 0.0422\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0356 - mean_squared_error: 0.0356\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0356 - mean_squared_error: 0.0356\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0336 - mean_squared_error: 0.0336\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0336 - mean_squared_error: 0.0336\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0342 - mean_squared_error: 0.0342\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0342 - mean_squared_error: 0.0342\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 38us/step - loss: 0.0340 - mean_squared_error: 0.0340 - val_loss: 0.0711 - val_mean_squared_error: 0.0711\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 38us/step - loss: 0.0340 - mean_squared_error: 0.0340 - val_loss: 0.0711 - val_mean_squared_error: 0.0711\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0329 - mean_squared_error: 0.0329\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0329 - mean_squared_error: 0.0329\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0299 - mean_squared_error: 0.0299\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0299 - mean_squared_error: 0.0299\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0290 - mean_squared_error: 0.0290\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0290 - mean_squared_error: 0.0290\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0282 - mean_squared_error: 0.02\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 29us/step - loss: 0.0283 - mean_squared_error: 0.0283 - val_loss: 0.0815 - val_mean_squared_error: 0.0815\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0282 - mean_squared_error: 0.02\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 29us/step - loss: 0.0283 - mean_squared_error: 0.0283 - val_loss: 0.0815 - val_mean_squared_error: 0.0815\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0338 - mean_squared_error: 0.0338\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0338 - mean_squared_error: 0.0338\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0344 - mean_squared_error: 0.0344\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0344 - mean_squared_error: 0.0344\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0349 - mean_squared_error: 0.0349\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0349 - mean_squared_error: 0.0349\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0343 - mean_squared_error: 0.0343\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0343 - mean_squared_error: 0.0343\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 33us/step - loss: 0.0335 - mean_squared_error: 0.0335 - val_loss: 0.0729 - val_mean_squared_error: 0.0729\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 33us/step - loss: 0.0335 - mean_squared_error: 0.0335 - val_loss: 0.0729 - val_mean_squared_error: 0.0729\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0323 - mean_squared_error: 0.0323\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0323 - mean_squared_error: 0.0323\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0286 - mean_squared_error: 0.0286\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0286 - mean_squared_error: 0.0286\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0269 - mean_squared_error: 0.0269\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0269 - mean_squared_error: 0.0269\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0282 - mean_squared_error: 0.0282\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0282 - mean_squared_error: 0.0282\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 35us/step - loss: 0.0276 - mean_squared_error: 0.0276 - val_loss: 0.0430 - val_mean_squared_error: 0.0430\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 35us/step - loss: 0.0276 - mean_squared_error: 0.0276 - val_loss: 0.0430 - val_mean_squared_error: 0.0430\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0461 - mean_squared_error: 0.0461\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0461 - mean_squared_error: 0.0461\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 2048/8242 [======>.......................] - ETA: 0s - loss: 0.0457 - mean_squared_error: 0.0457\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 2048/8242 [======>.......................] - ETA: 0s - loss: 0.0457 - mean_squared_error: 0.0457\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0383 - mean_squared_error: 0.0383\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0383 - mean_squared_error: 0.0383\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0351 - mean_squared_error: 0.0351\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0351 - mean_squared_error: 0.0351\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0338 - mean_squared_error: 0.0338\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0338 - mean_squared_error: 0.03\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 36us/step - loss: 0.0338 - mean_squared_error: 0.0338 - val_loss: 0.0566 - val_mean_squared_error: 0.0566\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 36us/step - loss: 0.0338 - mean_squared_error: 0.0338 - val_loss: 0.0566 - val_mean_squared_error: 0.0566\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0334 - mean_squared_error: 0.0334\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0334 - mean_squared_error: 0.0334\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0324 - mean_squared_error: 0.0324\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0324 - mean_squared_error: 0.0324\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0311 - mean_squared_error: 0.0311\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0311 - mean_squared_error: 0.0311\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0303 - mean_squared_error: 0.0303 - val_loss: 0.0789 - val_mean_squared_error: 0.0789\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0303 - mean_squared_error: 0.0303 - val_loss: 0.0789 - val_mean_squared_error: 0.0789\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0303 - mean_squared_error: 0.0303\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0303 - mean_squared_error: 0.0303\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0267 - mean_squared_error: 0.0267\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0267 - mean_squared_error: 0.0267\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0265 - mean_squared_error: 0.0265\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0265 - mean_squared_error: 0.0265\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0269 - mean_squared_error: 0.0269\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0269 - mean_squared_error: 0.0269\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0294 - mean_squared_error: 0.0294\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0294 - mean_squared_error: 0.02\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 40us/step - loss: 0.0293 - mean_squared_error: 0.0293 - val_loss: 0.0650 - val_mean_squared_error: 0.0650\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 40us/step - loss: 0.0293 - mean_squared_error: 0.0293 - val_loss: 0.0650 - val_mean_squared_error: 0.0650\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0307 - mean_squared_error: 0.0307\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0307 - mean_squared_error: 0.0307\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0292 - mean_squared_error: 0.0292\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0292 - mean_squared_error: 0.0292\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0264 - mean_squared_error: 0.0264\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0264 - mean_squared_error: 0.0264\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.0277 - mean_squared_error: 0.0277 - val_loss: 0.0844 - val_mean_squared_error: 0.0844\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 25us/step - loss: 0.0277 - mean_squared_error: 0.0277 - val_loss: 0.0844 - val_mean_squared_error: 0.0844\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0509 - mean_squared_error: 0.0509\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Train on 8242 samples, validate on 1018 samples\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 1/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0509 - mean_squared_error: 0.0509\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0397 - mean_squared_error: 0.0397\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0397 - mean_squared_error: 0.0397\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0320 - mean_squared_error: 0.0320\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 6144/8242 [=====================>........] - ETA: 0s - loss: 0.0320 - mean_squared_error: 0.0320\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0301 - mean_squared_error: 0.0301 - val_loss: 0.0567 - val_mean_squared_error: 0.0567\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0301 - mean_squared_error: 0.0301 - val_loss: 0.0567 - val_mean_squared_error: 0.0567\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 2/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0363 - mean_squared_error: 0.0363\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0363 - mean_squared_error: 0.0363\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0301 - mean_squared_error: 0.0301\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0301 - mean_squared_error: 0.0301\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0286 - mean_squared_error: 0.02\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0281 - mean_squared_error: 0.0281 - val_loss: 0.0732 - val_mean_squared_error: 0.0732\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0286 - mean_squared_error: 0.02\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0281 - mean_squared_error: 0.0281 - val_loss: 0.0732 - val_mean_squared_error: 0.0732\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 3/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0293 - mean_squared_error: 0.0293\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0293 - mean_squared_error: 0.0293\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0262 - mean_squared_error: 0.0262\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0262 - mean_squared_error: 0.0262\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0264 - mean_squared_error: 0.0264\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0264 - mean_squared_error: 0.0264\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0277 - mean_squared_error: 0.0277 - val_loss: 0.0605 - val_mean_squared_error: 0.0605\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 23us/step - loss: 0.0277 - mean_squared_error: 0.0277 - val_loss: 0.0605 - val_mean_squared_error: 0.0605\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 4/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0470 - mean_squared_error: 0.0470\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0470 - mean_squared_error: 0.0470\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0376 - mean_squared_error: 0.0376\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 3072/8242 [==========>...................] - ETA: 0s - loss: 0.0376 - mean_squared_error: 0.0376\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0320 - mean_squared_error: 0.0320\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 5120/8242 [=================>............] - ETA: 0s - loss: 0.0320 - mean_squared_error: 0.0320\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0286 - mean_squared_error: 0.02\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0289 - mean_squared_error: 0.0289 - val_loss: 0.0752 - val_mean_squared_error: 0.0752\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8192/8242 [============================>.] - ETA: 0s - loss: 0.0286 - mean_squared_error: 0.02\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0289 - mean_squared_error: 0.0289 - val_loss: 0.0752 - val_mean_squared_error: 0.0752\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m Epoch 5/5\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0469 - mean_squared_error: 0.0469\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m \n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 1024/8242 [==>...........................] - ETA: 0s - loss: 0.0469 - mean_squared_error: 0.0469\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0335 - mean_squared_error: 0.0335\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 4096/8242 [=============>................] - ETA: 0s - loss: 0.0335 - mean_squared_error: 0.0335\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0290 - mean_squared_error: 0.0290\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 7168/8242 [=========================>....] - ETA: 0s - loss: 0.0290 - mean_squared_error: 0.0290\n",
"== Status ==\n",
"Using FIFO scheduling algorithm.\n",
"Resources requested: 0/4 CPUs, 0/0 GPUs (0/1.0 ps, 0/1.0 trainer)\n",
"Memory usage on this node: 29.4/67.4 GB\n",
"Result logdir: /home/shan/ray_results/automl\n",
"Number of trials: 1 ({'TERMINATED': 1})\n",
"TERMINATED trials:\n",
" - train_func_0_batch_size=1024,dropout=0.21806,dropout_1=0.49144,dropout_2=0.45671,latent_dim=32,lr=0.0086638,lstm_1_units=64,lstm_2_units=128,selected_features=['WEEKDAY(datetime)' 'HOUR(datetime)' 'IS_AWAKE(datetime)'\n",
" 'IS_WEEKEND(datetime)']:\tTERMINATED, [2 CPUs, 0 GPUs], [pid=7230], 17 s, 10 iter\n",
"\n",
"== Status ==\n",
"Using FIFO scheduling algorithm.\n",
"Resources requested: 0/4 CPUs, 0/0 GPUs (0/1.0 ps, 0/1.0 trainer)\n",
"Memory usage on this node: 29.4/67.4 GB\n",
"Result logdir: /home/shan/ray_results/automl\n",
"Number of trials: 1 ({'TERMINATED': 1})\n",
"TERMINATED trials:\n",
" - train_func_0_batch_size=1024,dropout=0.21806,dropout_1=0.49144,dropout_2=0.45671,latent_dim=32,lr=0.0086638,lstm_1_units=64,lstm_2_units=128,selected_features=['WEEKDAY(datetime)' 'HOUR(datetime)' 'IS_AWAKE(datetime)'\n",
" 'IS_WEEKEND(datetime)']:\tTERMINATED, [2 CPUs, 0 GPUs], [pid=7230], 17 s, 10 iter\n",
"\n",
"The best configurations are:\n",
"selected_features : ['WEEKDAY(datetime)' 'HOUR(datetime)' 'IS_AWAKE(datetime)'\n",
" 'IS_WEEKEND(datetime)']\n",
"lstm_1_units : 64\n",
"dropout_1 : 0.4914392380008718\n",
"lstm_2_units : 128\n",
"dropout_2 : 0.4567086325546281\n",
"latent_dim : 32\n",
"dropout : 0.21805827608891876\n",
"lr : 0.008663837715873179\n",
"batch_size : 1024\n",
"epochs : 5\n",
"past_seq_len : 10\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0281 - mean_squared_error: 0.0281 - val_loss: 0.0622 - val_mean_squared_error: 0.0622\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0\n",
"\u001b[2m\u001b[36m(pid=7230)\u001b[0m 8242/8242 [==============================] - 0s 24us/step - loss: 0.0281 - mean_squared_error: 0.0281 - val_loss: 0.0622 - val_mean_squared_error: 0.0622\n",
"Training completed.\n",
"CPU times: user 4.67 s, sys: 174 ms, total: 4.84 s\n",
"Wall time: 21.7 s\n"
]
}
],
"source": [
"%%time\n",
"# you can specify the look back sequence length with a single number or a range of (min_len, max_len) in RandomRecipe.\n",
"pipeline = tsp.fit(train_df,\n",
" validation_df=val_df,\n",
" metric=\"mse\",\n",
" recipe=RandomRecipe(look_back=10))\n",
"print(\"Training completed.\")"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [],
"source": [
"# test\n",
"# predict test_df with the best trial\n",
"pred_df = pipeline.predict(test_df)"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" datetime | \n",
" value_0 | \n",
" value_1 | \n",
" value_2 | \n",
" value_3 | \n",
" value_4 | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 2015-01-10 17:00:00 | \n",
" 20544.215887 | \n",
" 20488.975324 | \n",
" 20003.972263 | \n",
" 19217.881677 | \n",
" 18247.222892 | \n",
"
\n",
" \n",
" | 1 | \n",
" 2015-01-10 17:30:00 | \n",
" 21114.137920 | \n",
" 20863.955988 | \n",
" 20218.575409 | \n",
" 19284.430270 | \n",
" 18182.702733 | \n",
"
\n",
" \n",
" | 2 | \n",
" 2015-01-10 18:00:00 | \n",
" 23327.051782 | \n",
" 22636.173674 | \n",
" 21635.037599 | \n",
" 20404.316844 | \n",
" 19071.609196 | \n",
"
\n",
" \n",
" | 3 | \n",
" 2015-01-10 18:30:00 | \n",
" 24770.840995 | \n",
" 23813.619430 | \n",
" 22561.436395 | \n",
" 21156.500540 | \n",
" 19719.400959 | \n",
"
\n",
" \n",
" | 4 | \n",
" 2015-01-10 19:00:00 | \n",
" 25104.992497 | \n",
" 24141.002251 | \n",
" 22834.490373 | \n",
" 21386.294104 | \n",
" 19924.729628 | \n",
"
\n",
" \n",
"
\n",
"