t81_558_class_08_2_keras_ensembles.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a href=\"https://colab.research.google.com/github/jeffheaton/t81_558_deep_learning/blob/master/t81_558_class_08_2_keras_ensembles.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# T81-558: Applications of Deep Neural Networks\n",
    "**Module 8: Kaggle Data Sets**\n",
    "* Instructor: [Jeff Heaton](https://sites.wustl.edu/jeffheaton/), McKelvey School of Engineering, [Washington University in St. Louis](https://engineering.wustl.edu/Programs/Pages/default.aspx)\n",
    "* For more information visit the [class website](https://sites.wustl.edu/jeffheaton/t81-558/)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Module 8 Material\n",
    "\n",
    "* Part 8.1: Introduction to Kaggle [[Video]](https://www.youtube.com/watch?v=v4lJBhdCuCU&list=PLjy4p-07OYzulelvJ5KVaT2pDlxivl_BN) [[Notebook]](t81_558_class_08_1_kaggle_intro.ipynb)\n",
    "* **Part 8.2: Building Ensembles with Scikit-Learn and Keras** [[Video]](https://www.youtube.com/watch?v=LQ-9ZRBLasw&list=PLjy4p-07OYzulelvJ5KVaT2pDlxivl_BN) [[Notebook]](t81_558_class_08_2_keras_ensembles.ipynb)\n",
    "* Part 8.3: How Should you Architect Your Keras Neural Network: Hyperparameters [[Video]](https://www.youtube.com/watch?v=1q9klwSoUQw&list=PLjy4p-07OYzulelvJ5KVaT2pDlxivl_BN) [[Notebook]](t81_558_class_08_3_keras_hyperparameters.ipynb)\n",
    "* Part 8.4: Bayesian Hyperparameter Optimization for Keras [[Video]](https://www.youtube.com/watch?v=sXdxyUCCm8s&list=PLjy4p-07OYzulelvJ5KVaT2pDlxivl_BN) [[Notebook]](t81_558_class_08_4_bayesian_hyperparameter_opt.ipynb)\n",
    "* Part 8.5: Current Semester's Kaggle [[Video]](https://www.youtube.com/watch?v=PHQt0aUasRg&list=PLjy4p-07OYzulelvJ5KVaT2pDlxivl_BN) [[Notebook]](t81_558_class_08_5_kaggle_project.ipynb)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Google CoLab Instructions\n",
    "\n",
    "The following code ensures that Google CoLab is running the correct version of TensorFlow.\n",
    "  Running the following code will map your GDrive to ```/content/drive```."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "try:\n",
    "    from google.colab import drive\n",
    "    drive.mount('/content/drive', force_remount=True)\n",
    "    COLAB = True\n",
    "    print(\"Note: using Google CoLab\")\n",
    "    %tensorflow_version 2.x\n",
    "except:\n",
    "    print(\"Note: not using Google CoLab\")\n",
    "    COLAB = False\n",
    "\n",
    "# Nicely formatted time string\n",
    "def hms_string(sec_elapsed):\n",
    "    h = int(sec_elapsed / (60 * 60))\n",
    "    m = int((sec_elapsed % (60 * 60)) / 60)\n",
    "    s = sec_elapsed % 60\n",
    "    return \"{}:{:>02}:{:>05.2f}\".format(h, m, s)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Part 8.2: Building Ensembles with Scikit-Learn and Keras\n",
    "\n",
    "### Evaluating Feature Importance\n",
    "\n",
    "Feature importance tells us how important each of the features (from the feature/import vector is to the prediction of a neural network or another model.  There are many different ways to evaluate the feature importance for neural networks.  The following paper presents an excellent (and readable) overview of the various means of assessing the significance of neural network inputs/features.\n",
    "\n",
    "* An accurate comparison of methods for quantifying variable importance in artificial neural networks using simulated data [[Cite:olden2004accurate]](http://depts.washington.edu/oldenlab/wordpress/wp-content/uploads/2013/03/EcologicalModelling_2004.pdf). *Ecological Modelling*, 178(3), 389-397.\n",
    "\n",
    "In summary, the following methods are available to neural networks:\n",
    "\n",
    "* Connection Weights Algorithm\n",
    "* Partial Derivatives\n",
    "* Input Perturbation\n",
    "* Sensitivity Analysis\n",
    "* Forward Stepwise Addition \n",
    "* Improved Stepwise Selection 1\n",
    "* Backward Stepwise Elimination\n",
    "* Improved Stepwise Selection\n",
    "\n",
    "For this class, we will use the **Input Perturbation** feature ranking algorithm.  This algorithm will work with any regression or classification network. I provide an implementation of the input perturbation algorithm for scikit-learn in the next section. This code implements a function below that will work with any scikit-learn model.\n",
    "\n",
    "[Leo Breiman](https://en.wikipedia.org/wiki/Leo_Breiman) provided this algorithm in his seminal paper on random forests. [[Citebreiman2001random:]](https://www.stat.berkeley.edu/~breiman/randomforest2001.pdf)  Although he presented this algorithm in conjunction with random forests, it is model-independent and appropriate for any supervised learning model.  This algorithm, known as the input perturbation algorithm, works by evaluating a trained model’s accuracy with each of the inputs individually shuffled from a data set.  Shuffling an input causes it to become useless—effectively removing it from the model. More important inputs will produce a less accurate score when they are removed by shuffling them. This process makes sense because important features will contribute to the accuracy of the model.  I first presented the TensorFlow implementation of this algorithm in the following paper.\n",
    "\n",
    "* Early stabilizing feature importance for TensorFlow deep neural networks[[Cite:heaton2017early]](https://www.heatonresearch.com/dload/phd/IJCNN%202017-v2-final.pdf)\n",
    "\n",
    "This algorithm will use log loss to evaluate a classification problem and RMSE for regression."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import metrics\n",
    "import scipy as sp\n",
    "import numpy as np\n",
    "import math\n",
    "from sklearn import metrics\n",
    "\n",
    "def perturbation_rank(model, x, y, names, regression):\n",
    "    errors = []\n",
    "\n",
    "    for i in range(x.shape[1]):\n",
    "        hold = np.array(x[:, i])\n",
    "        np.random.shuffle(x[:, i])\n",
    "        \n",
    "        if regression:\n",
    "            pred = model.predict(x)\n",
    "            error = metrics.mean_squared_error(y, pred)\n",
    "        else:\n",
    "            pred = model.predict_proba(x)\n",
    "            error = metrics.log_loss(y, pred)\n",
    "            \n",
    "        errors.append(error)\n",
    "        x[:, i] = hold\n",
    "        \n",
    "    max_error = np.max(errors)\n",
    "    importance = [e/max_error for e in errors]\n",
    "\n",
    "    data = {'name':names,'error':errors,'importance':importance}\n",
    "    result = pd.DataFrame(data, columns = ['name','error','importance'])\n",
    "    result.sort_values(by=['importance'], ascending=[0], inplace=True)\n",
    "    result.reset_index(inplace=True, drop=True)\n",
    "    return result"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Classification and Input Perturbation Ranking\n",
    "\n",
    "We now look at the code to perform perturbation ranking for a classification neural network.  The implementation technique is slightly different for classification vs regression, so I must provide two different implementations.  The primary difference between classification and regression is how we evaluate the accuracy of the neural network in each of these two network types.  For regression neural networks, we will use the Root Mean Square (RMSE) error calculation; whereas, we will use log loss for classification.\n",
    "\n",
    "The code presented below creates a classification neural network that will predict for the classic iris dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Logging before flag parsing goes to stderr.\n",
      "W0819 15:14:28.751466 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:28.882447 12720 deprecation.py:323] From C:\\Users\\jheaton\\AppData\\Local\\Continuum\\miniconda3\\envs\\tensorflow\\lib\\site-packages\\tensorflow_core\\python\\keras\\optimizer_v2\\optimizer_v2.py:466: BaseResourceVariable.constraint (from tensorflow.python.ops.resource_variable_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Apply a constraint manually following the optimizer update step.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "112/112 - 1s - loss: 1.2174\n",
      "Epoch 2/100\n",
      "112/112 - 0s - loss: 1.0857\n",
      "Epoch 3/100\n",
      "112/112 - 0s - loss: 1.0299\n",
      "Epoch 4/100\n",
      "112/112 - 0s - loss: 0.9896\n",
      "Epoch 5/100\n",
      "112/112 - 0s - loss: 0.9515\n",
      "Epoch 6/100\n",
      "112/112 - 0s - loss: 0.9139\n",
      "Epoch 7/100\n",
      "112/112 - 0s - loss: 0.8793\n",
      "Epoch 8/100\n",
      "112/112 - 0s - loss: 0.8433\n",
      "Epoch 9/100\n",
      "112/112 - 0s - loss: 0.8144\n",
      "Epoch 10/100\n",
      "112/112 - 0s - loss: 0.7815\n",
      "Epoch 11/100\n",
      "112/112 - 0s - loss: 0.7521\n",
      "Epoch 12/100\n",
      "112/112 - 0s - loss: 0.7203\n",
      "Epoch 13/100\n",
      "112/112 - 0s - loss: 0.6933\n",
      "Epoch 14/100\n",
      "112/112 - 0s - loss: 0.6650\n",
      "Epoch 15/100\n",
      "112/112 - 0s - loss: 0.6376\n",
      "Epoch 16/100\n",
      "112/112 - 0s - loss: 0.6142\n",
      "Epoch 17/100\n",
      "112/112 - 0s - loss: 0.5917\n",
      "Epoch 18/100\n",
      "112/112 - 0s - loss: 0.5702\n",
      "Epoch 19/100\n",
      "112/112 - 0s - loss: 0.5507\n",
      "Epoch 20/100\n",
      "112/112 - 0s - loss: 0.5329\n",
      "Epoch 21/100\n",
      "112/112 - 0s - loss: 0.5157\n",
      "Epoch 22/100\n",
      "112/112 - 0s - loss: 0.4992\n",
      "Epoch 23/100\n",
      "112/112 - 0s - loss: 0.4858\n",
      "Epoch 24/100\n",
      "112/112 - 0s - loss: 0.4705\n",
      "Epoch 25/100\n",
      "112/112 - 0s - loss: 0.4557\n",
      "Epoch 26/100\n",
      "112/112 - 0s - loss: 0.4425\n",
      "Epoch 27/100\n",
      "112/112 - 0s - loss: 0.4329\n",
      "Epoch 28/100\n",
      "112/112 - 0s - loss: 0.4167\n",
      "Epoch 29/100\n",
      "112/112 - 0s - loss: 0.4079\n",
      "Epoch 30/100\n",
      "112/112 - 0s - loss: 0.3972\n",
      "Epoch 31/100\n",
      "112/112 - 0s - loss: 0.3866\n",
      "Epoch 32/100\n",
      "112/112 - 0s - loss: 0.3775\n",
      "Epoch 33/100\n",
      "112/112 - 0s - loss: 0.3674\n",
      "Epoch 34/100\n",
      "112/112 - 0s - loss: 0.3574\n",
      "Epoch 35/100\n",
      "112/112 - 0s - loss: 0.3471\n",
      "Epoch 36/100\n",
      "112/112 - 0s - loss: 0.3400\n",
      "Epoch 37/100\n",
      "112/112 - 0s - loss: 0.3290\n",
      "Epoch 38/100\n",
      "112/112 - 0s - loss: 0.3211\n",
      "Epoch 39/100\n",
      "112/112 - 0s - loss: 0.3141\n",
      "Epoch 40/100\n",
      "112/112 - 0s - loss: 0.3075\n",
      "Epoch 41/100\n",
      "112/112 - 0s - loss: 0.2985\n",
      "Epoch 42/100\n",
      "112/112 - 0s - loss: 0.2901\n",
      "Epoch 43/100\n",
      "112/112 - 0s - loss: 0.2852\n",
      "Epoch 44/100\n",
      "112/112 - 0s - loss: 0.2796\n",
      "Epoch 45/100\n",
      "112/112 - 0s - loss: 0.2742\n",
      "Epoch 46/100\n",
      "112/112 - 0s - loss: 0.2632\n",
      "Epoch 47/100\n",
      "112/112 - 0s - loss: 0.2556\n",
      "Epoch 48/100\n",
      "112/112 - 0s - loss: 0.2528\n",
      "Epoch 49/100\n",
      "112/112 - 0s - loss: 0.2471\n",
      "Epoch 50/100\n",
      "112/112 - 0s - loss: 0.2407\n",
      "Epoch 51/100\n",
      "112/112 - 0s - loss: 0.2338\n",
      "Epoch 52/100\n",
      "112/112 - 0s - loss: 0.2346\n",
      "Epoch 53/100\n",
      "112/112 - 0s - loss: 0.2221\n",
      "Epoch 54/100\n",
      "112/112 - 0s - loss: 0.2227\n",
      "Epoch 55/100\n",
      "112/112 - 0s - loss: 0.2098\n",
      "Epoch 56/100\n",
      "112/112 - 0s - loss: 0.2173\n",
      "Epoch 57/100\n",
      "112/112 - 0s - loss: 0.2105\n",
      "Epoch 58/100\n",
      "112/112 - 0s - loss: 0.1997\n",
      "Epoch 59/100\n",
      "112/112 - 0s - loss: 0.1969\n",
      "Epoch 60/100\n",
      "112/112 - 0s - loss: 0.1908\n",
      "Epoch 61/100\n",
      "112/112 - 0s - loss: 0.1885\n",
      "Epoch 62/100\n",
      "112/112 - 0s - loss: 0.1832\n",
      "Epoch 63/100\n",
      "112/112 - 0s - loss: 0.1802\n",
      "Epoch 64/100\n",
      "112/112 - 0s - loss: 0.1771\n",
      "Epoch 65/100\n",
      "112/112 - 0s - loss: 0.1737\n",
      "Epoch 66/100\n",
      "112/112 - 0s - loss: 0.1702\n",
      "Epoch 67/100\n",
      "112/112 - 0s - loss: 0.1670\n",
      "Epoch 68/100\n",
      "112/112 - 0s - loss: 0.1648\n",
      "Epoch 69/100\n",
      "112/112 - 0s - loss: 0.1617\n",
      "Epoch 70/100\n",
      "112/112 - 0s - loss: 0.1586\n",
      "Epoch 71/100\n",
      "112/112 - 0s - loss: 0.1560\n",
      "Epoch 72/100\n",
      "112/112 - 0s - loss: 0.1532\n",
      "Epoch 73/100\n",
      "112/112 - 0s - loss: 0.1493\n",
      "Epoch 74/100\n",
      "112/112 - 0s - loss: 0.1587\n",
      "Epoch 75/100\n",
      "112/112 - 0s - loss: 0.1423\n",
      "Epoch 76/100\n",
      "112/112 - 0s - loss: 0.1529\n",
      "Epoch 77/100\n",
      "112/112 - 0s - loss: 0.1449\n",
      "Epoch 78/100\n",
      "112/112 - 0s - loss: 0.1462\n",
      "Epoch 79/100\n",
      "112/112 - 0s - loss: 0.1439\n",
      "Epoch 80/100\n",
      "112/112 - 0s - loss: 0.1363\n",
      "Epoch 81/100\n",
      "112/112 - 0s - loss: 0.1367\n",
      "Epoch 82/100\n",
      "112/112 - 0s - loss: 0.1318\n",
      "Epoch 83/100\n",
      "112/112 - 0s - loss: 0.1315\n",
      "Epoch 84/100\n",
      "112/112 - 0s - loss: 0.1280\n",
      "Epoch 85/100\n",
      "112/112 - 0s - loss: 0.1287\n",
      "Epoch 86/100\n",
      "112/112 - 0s - loss: 0.1279\n",
      "Epoch 87/100\n",
      "112/112 - 0s - loss: 0.1229\n",
      "Epoch 88/100\n",
      "112/112 - 0s - loss: 0.1232\n",
      "Epoch 89/100\n",
      "112/112 - 0s - loss: 0.1192\n",
      "Epoch 90/100\n",
      "112/112 - 0s - loss: 0.1195\n",
      "Epoch 91/100\n",
      "112/112 - 0s - loss: 0.1203\n",
      "Epoch 92/100\n",
      "112/112 - 0s - loss: 0.1160\n",
      "Epoch 93/100\n",
      "112/112 - 0s - loss: 0.1156\n",
      "Epoch 94/100\n",
      "112/112 - 0s - loss: 0.1154\n",
      "Epoch 95/100\n",
      "112/112 - 0s - loss: 0.1131\n",
      "Epoch 96/100\n",
      "112/112 - 0s - loss: 0.1110\n",
      "Epoch 97/100\n",
      "112/112 - 0s - loss: 0.1144\n",
      "Epoch 98/100\n",
      "112/112 - 0s - loss: 0.1085\n",
      "Epoch 99/100\n",
      "112/112 - 0s - loss: 0.1107\n",
      "Epoch 100/100\n",
      "112/112 - 0s - loss: 0.1092\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x19623322320>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import io\n",
    "import requests\n",
    "import numpy as np\n",
    "from sklearn import metrics\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Activation\n",
    "from tensorflow.keras.callbacks import EarlyStopping\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "df = pd.read_csv(\n",
    "    \"https://data.heatonresearch.com/data/t81-558/iris.csv\", \n",
    "    na_values=['NA', '?'])\n",
    "\n",
    "# Convert to numpy - Classification\n",
    "x = df[['sepal_l', 'sepal_w', 'petal_l', 'petal_w']].values\n",
    "dummies = pd.get_dummies(df['species']) # Classification\n",
    "species = dummies.columns\n",
    "y = dummies.values\n",
    "\n",
    "# Split into train/test\n",
    "x_train, x_test, y_train, y_test = train_test_split(    \n",
    "    x, y, test_size=0.25, random_state=42)\n",
    "\n",
    "# Build neural network\n",
    "model = Sequential()\n",
    "model.add(Dense(50, input_dim=x.shape[1], activation='relu')) # Hidden 1\n",
    "model.add(Dense(25, activation='relu')) # Hidden 2\n",
    "model.add(Dense(y.shape[1],activation='softmax')) # Output\n",
    "model.compile(loss='categorical_crossentropy', optimizer='adam')\n",
    "model.fit(x_train,y_train,verbose=2,epochs=100)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we evaluate the accuracy of the trained model.  Here we see that the neural network is performing great, with an accuracy of 1.0.  For a more complex dataset, we might fear overfitting with such high accuracy.  However, for this example, we are more interested in determining the importance of each column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:30.387507 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 1.0\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "pred = model.predict(x_test)\n",
    "predict_classes = np.argmax(pred,axis=1)\n",
    "expected_classes = np.argmax(y_test,axis=1)\n",
    "correct = accuracy_score(expected_classes,predict_classes)\n",
    "print(f\"Accuracy: {correct}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are now ready to call the input perturbation algorithm.  First, we extract the column names and remove the target column.  The target column does not have importance, as it is the objective, not one of the inputs.  In supervised learning, the target is of the utmost importance.\n",
    "\n",
    "We can see the importance displayed in the following table.  The most important column is always 1.0, and lessor columns will continue in a downward trend.  The least important column will have the lowest rank."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:30.444489 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:30.453488 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:30.461488 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:30.471489 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>error</th>\n",
       "      <th>importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>petal_l</td>\n",
       "      <td>2.846739</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>petal_w</td>\n",
       "      <td>0.671724</td>\n",
       "      <td>0.235963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>sepal_l</td>\n",
       "      <td>0.141148</td>\n",
       "      <td>0.049582</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>sepal_w</td>\n",
       "      <td>0.091809</td>\n",
       "      <td>0.032250</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      name     error  importance\n",
       "0  petal_l  2.846739    1.000000\n",
       "1  petal_w  0.671724    0.235963\n",
       "2  sepal_l  0.141148    0.049582\n",
       "3  sepal_w  0.091809    0.032250"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Rank the features\n",
    "from IPython.display import display, HTML\n",
    "\n",
    "names = list(df.columns) # x+y column names\n",
    "names.remove(\"species\") # remove the target(y)\n",
    "rank = perturbation_rank(model, x_test, y_test, names, False)\n",
    "display(rank)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Regression and Input Perturbation Ranking\n",
    "\n",
    "We now see how to use input perturbation ranking for a regression neural network.  We will use the MPG dataset as a demonstration.  The code below loads the MPG dataset and creates a regression neural network for this dataset.  The code trains the neural network and calculates an RMSE evaluation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:30.628494 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "298/298 - 0s - loss: 1559.2899\n",
      "Epoch 2/100\n",
      "298/298 - 0s - loss: 443.0820\n",
      "Epoch 3/100\n",
      "298/298 - 0s - loss: 300.1852\n",
      "Epoch 4/100\n",
      "298/298 - 0s - loss: 253.6714\n",
      "Epoch 5/100\n",
      "298/298 - 0s - loss: 187.6016\n",
      "Epoch 6/100\n",
      "298/298 - 0s - loss: 183.5883\n",
      "Epoch 7/100\n",
      "298/298 - 0s - loss: 183.9289\n",
      "Epoch 8/100\n",
      "298/298 - 0s - loss: 167.7776\n",
      "Epoch 9/100\n",
      "298/298 - 0s - loss: 160.7908\n",
      "Epoch 10/100\n",
      "298/298 - 0s - loss: 155.8137\n",
      "Epoch 11/100\n",
      "298/298 - 0s - loss: 149.1986\n",
      "Epoch 12/100\n",
      "298/298 - 0s - loss: 144.5110\n",
      "Epoch 13/100\n",
      "298/298 - 0s - loss: 139.0513\n",
      "Epoch 14/100\n",
      "298/298 - 0s - loss: 136.2702\n",
      "Epoch 15/100\n",
      "298/298 - 0s - loss: 136.8744\n",
      "Epoch 16/100\n",
      "298/298 - 0s - loss: 127.1330\n",
      "Epoch 17/100\n",
      "298/298 - 0s - loss: 132.9591\n",
      "Epoch 18/100\n",
      "298/298 - 0s - loss: 128.9401\n",
      "Epoch 19/100\n",
      "298/298 - 0s - loss: 115.0540\n",
      "Epoch 20/100\n",
      "298/298 - 0s - loss: 111.8599\n",
      "Epoch 21/100\n",
      "298/298 - 0s - loss: 115.3871\n",
      "Epoch 22/100\n",
      "298/298 - 0s - loss: 106.7886\n",
      "Epoch 23/100\n",
      "298/298 - 0s - loss: 101.2296\n",
      "Epoch 24/100\n",
      "298/298 - 0s - loss: 100.4774\n",
      "Epoch 25/100\n",
      "298/298 - 0s - loss: 97.5229\n",
      "Epoch 26/100\n",
      "298/298 - 0s - loss: 93.4279\n",
      "Epoch 27/100\n",
      "298/298 - 0s - loss: 93.8727\n",
      "Epoch 28/100\n",
      "298/298 - 0s - loss: 89.8434\n",
      "Epoch 29/100\n",
      "298/298 - 0s - loss: 91.6693\n",
      "Epoch 30/100\n",
      "298/298 - 0s - loss: 87.4931\n",
      "Epoch 31/100\n",
      "298/298 - 0s - loss: 89.6551\n",
      "Epoch 32/100\n",
      "298/298 - 0s - loss: 82.5026\n",
      "Epoch 33/100\n",
      "298/298 - 0s - loss: 88.1567\n",
      "Epoch 34/100\n",
      "298/298 - 0s - loss: 91.1699\n",
      "Epoch 35/100\n",
      "298/298 - 0s - loss: 96.9902\n",
      "Epoch 36/100\n",
      "298/298 - 0s - loss: 96.0484\n",
      "Epoch 37/100\n",
      "298/298 - 0s - loss: 89.2738\n",
      "Epoch 38/100\n",
      "298/298 - 0s - loss: 75.7954\n",
      "Epoch 39/100\n",
      "298/298 - 0s - loss: 73.3345\n",
      "Epoch 40/100\n",
      "298/298 - 0s - loss: 74.2692\n",
      "Epoch 41/100\n",
      "298/298 - 0s - loss: 74.6573\n",
      "Epoch 42/100\n",
      "298/298 - 0s - loss: 70.3825\n",
      "Epoch 43/100\n",
      "298/298 - 0s - loss: 65.6756\n",
      "Epoch 44/100\n",
      "298/298 - 0s - loss: 63.5143\n",
      "Epoch 45/100\n",
      "298/298 - 0s - loss: 63.1461\n",
      "Epoch 46/100\n",
      "298/298 - 0s - loss: 61.5945\n",
      "Epoch 47/100\n",
      "298/298 - 0s - loss: 60.8559\n",
      "Epoch 48/100\n",
      "298/298 - 0s - loss: 60.9447\n",
      "Epoch 49/100\n",
      "298/298 - 0s - loss: 58.8260\n",
      "Epoch 50/100\n",
      "298/298 - 0s - loss: 65.2873\n",
      "Epoch 51/100\n",
      "298/298 - 0s - loss: 61.4850\n",
      "Epoch 52/100\n",
      "298/298 - 0s - loss: 76.4146\n",
      "Epoch 53/100\n",
      "298/298 - 0s - loss: 66.9197\n",
      "Epoch 54/100\n",
      "298/298 - 0s - loss: 54.8163\n",
      "Epoch 55/100\n",
      "298/298 - 0s - loss: 55.1214\n",
      "Epoch 56/100\n",
      "298/298 - 0s - loss: 50.5856\n",
      "Epoch 57/100\n",
      "298/298 - 0s - loss: 50.2830\n",
      "Epoch 58/100\n",
      "298/298 - 0s - loss: 49.6034\n",
      "Epoch 59/100\n",
      "298/298 - 0s - loss: 48.8089\n",
      "Epoch 60/100\n",
      "298/298 - 0s - loss: 47.4670\n",
      "Epoch 61/100\n",
      "298/298 - 0s - loss: 45.9854\n",
      "Epoch 62/100\n",
      "298/298 - 0s - loss: 43.3706\n",
      "Epoch 63/100\n",
      "298/298 - 0s - loss: 42.6222\n",
      "Epoch 64/100\n",
      "298/298 - 0s - loss: 42.4879\n",
      "Epoch 65/100\n",
      "298/298 - 0s - loss: 44.4657\n",
      "Epoch 66/100\n",
      "298/298 - 0s - loss: 42.8027\n",
      "Epoch 67/100\n",
      "298/298 - 0s - loss: 46.7832\n",
      "Epoch 68/100\n",
      "298/298 - 0s - loss: 42.7598\n",
      "Epoch 69/100\n",
      "298/298 - 0s - loss: 39.2589\n",
      "Epoch 70/100\n",
      "298/298 - 0s - loss: 40.6138\n",
      "Epoch 71/100\n",
      "298/298 - 0s - loss: 40.9221\n",
      "Epoch 72/100\n",
      "298/298 - 0s - loss: 35.5641\n",
      "Epoch 73/100\n",
      "298/298 - 0s - loss: 36.4354\n",
      "Epoch 74/100\n",
      "298/298 - 0s - loss: 35.6893\n",
      "Epoch 75/100\n",
      "298/298 - 0s - loss: 34.6530\n",
      "Epoch 76/100\n",
      "298/298 - 0s - loss: 32.8944\n",
      "Epoch 77/100\n",
      "298/298 - 0s - loss: 33.5579\n",
      "Epoch 78/100\n",
      "298/298 - 0s - loss: 31.5365\n",
      "Epoch 79/100\n",
      "298/298 - 0s - loss: 31.0707\n",
      "Epoch 80/100\n",
      "298/298 - 0s - loss: 32.3957\n",
      "Epoch 81/100\n",
      "298/298 - 0s - loss: 31.6181\n",
      "Epoch 82/100\n",
      "298/298 - 0s - loss: 29.7164\n",
      "Epoch 83/100\n",
      "298/298 - 0s - loss: 29.7082\n",
      "Epoch 84/100\n",
      "298/298 - 0s - loss: 28.9681\n",
      "Epoch 85/100\n",
      "298/298 - 0s - loss: 29.1248\n",
      "Epoch 86/100\n",
      "298/298 - 0s - loss: 27.4652\n",
      "Epoch 87/100\n",
      "298/298 - 0s - loss: 27.6053\n",
      "Epoch 88/100\n",
      "298/298 - 0s - loss: 26.4133\n",
      "Epoch 89/100\n",
      "298/298 - 0s - loss: 25.0738\n",
      "Epoch 90/100\n",
      "298/298 - 0s - loss: 25.3491\n",
      "Epoch 91/100\n",
      "298/298 - 0s - loss: 25.3465\n",
      "Epoch 92/100\n",
      "298/298 - 0s - loss: 24.3639\n",
      "Epoch 93/100\n",
      "298/298 - 0s - loss: 23.8985\n",
      "Epoch 94/100\n",
      "298/298 - 0s - loss: 23.1552\n",
      "Epoch 95/100\n",
      "298/298 - 0s - loss: 23.1016\n",
      "Epoch 96/100\n",
      "298/298 - 0s - loss: 23.2722\n",
      "Epoch 97/100\n",
      "298/298 - 0s - loss: 22.0348\n",
      "Epoch 98/100\n",
      "298/298 - 0s - loss: 23.9377\n",
      "Epoch 99/100\n",
      "298/298 - 0s - loss: 22.6452\n",
      "Epoch 100/100\n",
      "298/298 - 0s - loss: 21.8604\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:32.600543 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Activation\n",
    "from sklearn.model_selection import train_test_split\n",
    "import pandas as pd\n",
    "import io\n",
    "import os\n",
    "import requests\n",
    "import numpy as np\n",
    "from sklearn import metrics\n",
    "\n",
    "save_path = \".\"\n",
    "\n",
    "df = pd.read_csv(\n",
    "    \"https://data.heatonresearch.com/data/t81-558/auto-mpg.csv\", \n",
    "    na_values=['NA', '?'])\n",
    "\n",
    "cars = df['name']\n",
    "\n",
    "# Handle missing value\n",
    "df['horsepower'] = df['horsepower'].fillna(df['horsepower'].median())\n",
    "\n",
    "# Pandas to Numpy\n",
    "x = df[['cylinders', 'displacement', 'horsepower', 'weight',\n",
    "       'acceleration', 'year', 'origin']].values\n",
    "y = df['mpg'].values # regression\n",
    "\n",
    "# Split into train/test\n",
    "x_train, x_test, y_train, y_test = train_test_split(    \n",
    "    x, y, test_size=0.25, random_state=42)\n",
    "\n",
    "# Build the neural network\n",
    "model = Sequential()\n",
    "model.add(Dense(25, input_dim=x.shape[1], activation='relu')) # Hidden 1\n",
    "model.add(Dense(10, activation='relu')) # Hidden 2\n",
    "model.add(Dense(1)) # Output\n",
    "model.compile(loss='mean_squared_error', optimizer='adam')\n",
    "model.fit(x_train,y_train,verbose=2,epochs=100)\n",
    "\n",
    "# Predict\n",
    "pred = model.predict(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Just as before, we extract the column names and discard the target.  We can now create a ranking of the importance of each of the input features.  The feature with a ranking of 1.0 is the most important."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:32.645547 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:32.656546 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:32.665548 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:32.674548 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:32.683545 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:32.692548 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:32.701548 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>error</th>\n",
       "      <th>importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>displacement</td>\n",
       "      <td>127.022319</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>weight</td>\n",
       "      <td>33.068019</td>\n",
       "      <td>0.260332</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>year</td>\n",
       "      <td>24.988007</td>\n",
       "      <td>0.196721</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>horsepower</td>\n",
       "      <td>20.419030</td>\n",
       "      <td>0.160752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>acceleration</td>\n",
       "      <td>18.785670</td>\n",
       "      <td>0.147893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>origin</td>\n",
       "      <td>18.560458</td>\n",
       "      <td>0.146120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>cylinders</td>\n",
       "      <td>18.292052</td>\n",
       "      <td>0.144007</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           name       error  importance\n",
       "0  displacement  127.022319    1.000000\n",
       "1        weight   33.068019    0.260332\n",
       "2          year   24.988007    0.196721\n",
       "3    horsepower   20.419030    0.160752\n",
       "4  acceleration   18.785670    0.147893\n",
       "5        origin   18.560458    0.146120\n",
       "6     cylinders   18.292052    0.144007"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Rank the features\n",
    "from IPython.display import display, HTML\n",
    "\n",
    "names = list(df.columns) # x+y column names\n",
    "names.remove(\"name\")\n",
    "names.remove(\"mpg\") # remove the target(y)\n",
    "rank = perturbation_rank(model, x_test, y_test, names, True)\n",
    "display(rank)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Biological Response with Neural Network\n",
    "\n",
    "The following sections will demonstrate how to use feature importance ranking and ensembling with a more complex dataset.  Ensembling is the process where you combine multiple models for greater accuracy. Kaggle competition winners frequently make use of ensembling for high ranking solutions.\n",
    "\n",
    "We will use the biological response dataset, a Kaggle dataset, where there is an unusually high number of columns.  Because of the large number of columns, it is essential to use feature ranking to determine the importance of these columns.  We begin by loading the dataset and preprocessing.  This Kaggle dataset is a binary classification problem. You must predict if certain conditions will cause a biological response.\n",
    "\n",
    "* [Predicting a Biological Response](https://www.kaggle.com/c/bioresponse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "import numpy as np\n",
    "from sklearn import metrics\n",
    "from scipy.stats import zscore\n",
    "from sklearn.model_selection import KFold\n",
    "from IPython.display import HTML, display\n",
    "\n",
    "if COLAB:\n",
    "    path = \"/content/drive/My Drive/data/\"\n",
    "else:\n",
    "    path = \"./data/\"\n",
    "    \n",
    "filename_train = os.path.join(path,\"bio_train.csv\")\n",
    "filename_test = os.path.join(path,\"bio_test.csv\")\n",
    "filename_submit = os.path.join(path,\"bio_submit.csv\")\n",
    "\n",
    "df_train = pd.read_csv(filename_train,na_values=['NA','?'])\n",
    "df_test = pd.read_csv(filename_test,na_values=['NA','?'])\n",
    "\n",
    "activity_classes = df_train['Activity']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A large number of columns is evident when we display the shape of the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3751, 1777)\n"
     ]
    }
   ],
   "source": [
    "print(df_train.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The following code constructs a classification neural network and trains it for the biological response dataset.  Once trained, the accuracy is measured."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:34.957605 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:35.015625 12720 deprecation.py:323] From C:\\Users\\jheaton\\AppData\\Local\\Continuum\\miniconda3\\envs\\tensorflow\\lib\\site-packages\\tensorflow_core\\python\\ops\\nn_impl.py:183: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting/Training...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:37.226079 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.304084 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 00007: early stopping\n",
      "Fitting done...\n",
      "Validation logloss: 0.5580600737567578\n",
      "Validation accuracy score: 0.7633262260127932\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Activation\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tensorflow.keras.callbacks import EarlyStopping\n",
    "import numpy as np\n",
    "import sklearn\n",
    "\n",
    "# Encode feature vector\n",
    "# Convert to numpy - Classification\n",
    "x_columns = df_train.columns.drop('Activity')\n",
    "x = df_train[x_columns].values\n",
    "y = df_train['Activity'].values # Classification\n",
    "x_submit = df_test[x_columns].values.astype(np.float32)\n",
    "\n",
    "\n",
    "# Split into train/test\n",
    "x_train, x_test, y_train, y_test = train_test_split(    \n",
    "    x, y, test_size=0.25, random_state=42) \n",
    "\n",
    "print(\"Fitting/Training...\")\n",
    "model = Sequential()\n",
    "model.add(Dense(25, input_dim=x.shape[1], activation='relu'))\n",
    "model.add(Dense(10))\n",
    "model.add(Dense(1,activation='sigmoid'))\n",
    "model.compile(loss='binary_crossentropy', optimizer='adam')\n",
    "monitor = EarlyStopping(monitor='val_loss', min_delta=1e-3, \n",
    "                        patience=5, verbose=1, mode='auto')\n",
    "model.fit(x_train,y_train,validation_data=(x_test,y_test),\n",
    "          callbacks=[monitor],verbose=0,epochs=1000)\n",
    "print(\"Fitting done...\")\n",
    "\n",
    "# Predict\n",
    "pred = model.predict(x_test).flatten()\n",
    "\n",
    "\n",
    "# Clip so that min is never exactly 0, max never 1\n",
    "pred = np.clip(pred,a_min=1e-6,a_max=(1-1e-6)) \n",
    "print(\"Validation logloss: {}\".format(\n",
    "    sklearn.metrics.log_loss(y_test,pred)))\n",
    "\n",
    "# Evaluate success using accuracy\n",
    "pred = pred>0.5 # If greater than 0.5 probability, then true\n",
    "score = metrics.accuracy_score(y_test, pred)\n",
    "print(\"Validation accuracy score: {}\".format(score))\n",
    "\n",
    "# Build real submit file\n",
    "pred_submit = model.predict(x_submit)\n",
    "\n",
    "# Clip so that min is never exactly 0, max never 1 (would be a NaN score)\n",
    "pred = np.clip(pred,a_min=1e-6,a_max=(1-1e-6)) \n",
    "submit_df = pd.DataFrame({'MoleculeId':[x+1 for x \\\n",
    "        in range(len(pred_submit))],'PredictedProbability':\\\n",
    "                          pred_submit.flatten()})\n",
    "submit_df.to_csv(filename_submit, index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What Features/Columns are Important\n",
    "The following uses perturbation ranking to evaluate the neural network."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:37.532091 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.579099 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.621090 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.664100 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.705101 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.749103 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.792104 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.840097 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.886246 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.930247 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:37.974247 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.017249 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.061250 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.104253 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.146253 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.188254 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.231254 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.274256 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.318258 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.361258 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.403259 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.445249 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.488260 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.532262 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.575256 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.625268 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.666254 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.711256 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.759257 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.801269 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.843271 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.884272 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.927273 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:38.970262 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.013276 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.055276 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.099277 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.147278 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.190279 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.232280 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.275280 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.318281 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.361285 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.405286 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.448287 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.492743 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.535744 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.579745 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:39.622746 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.673736 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.717748 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.761739 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.807470 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.848460 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.892464 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.935474 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:39.978475 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.021475 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.063478 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.107467 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.149478 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.191480 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.234482 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.276482 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.318484 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.360487 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.414477 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.461476 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.503479 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.549480 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.593478 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.639480 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.680481 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.726482 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.770485 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.811489 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.851486 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.892486 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.933499 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:40.975500 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.016500 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.060502 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.103493 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.145492 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.186506 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.228507 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.270509 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.312509 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.354511 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.398505 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.442511 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.490514 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.531513 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.573514 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.615518 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.656517 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:41.697509 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.738520 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.778522 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.820523 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.862524 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.904525 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.947525 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:41.989526 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.031517 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.075531 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.118531 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.162520 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.207523 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.250522 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.294523 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.337536 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.380537 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.421542 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.465926 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.508841 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.550842 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.593843 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.635845 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.677846 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.723846 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.765847 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.805849 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.847849 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.888851 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.929853 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:42.970853 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.012854 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.053855 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.096857 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.137858 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.179858 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.220859 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.261860 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.302861 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.343862 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.384861 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.425865 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.465865 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.506869 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.546868 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.587869 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.628868 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.668870 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:43.708877 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.750873 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.791874 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.831875 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.872876 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.912878 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.953881 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:43.995868 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.035881 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.077881 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.122883 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.164883 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.206885 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.248887 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.289887 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.329888 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.374892 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.418891 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.459891 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.501892 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.542894 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.582895 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.622896 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.663897 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.703897 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.745898 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.785900 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.826902 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.867903 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.907903 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.948904 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:44.988907 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.029906 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.071908 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.112909 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.153910 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.195910 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.236912 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.277912 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.318914 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.358915 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.400905 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.440917 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.481907 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.522918 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.562923 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.602921 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.643923 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:45.684925 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.726913 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.767925 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.807928 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.848927 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.888930 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.929929 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:45.970930 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.011931 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.053933 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.093936 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.135938 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.176936 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.217938 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.259938 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.299940 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.340940 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.381935 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.422942 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.463944 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.503944 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.546948 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.588946 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.631948 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.673949 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.716950 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.758940 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.800953 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.841943 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.884954 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.926956 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:46.967958 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.009958 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.050958 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.091959 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.133960 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.174645 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.215645 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.257646 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.299636 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.341647 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.381636 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.422638 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.462650 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.503642 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.544653 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.584654 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.625654 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:47.666657 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.706660 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.747658 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.788715 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.829715 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.870301 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.911301 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.952303 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:47.993303 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.037304 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.077305 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.118309 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.158308 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.198309 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.239309 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.280311 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.320312 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.361191 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.401189 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.441190 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.482191 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.522192 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.563194 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.604194 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.644196 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.685197 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.725197 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.766187 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.806197 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.848188 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.895202 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.937204 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:48.985661 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.026674 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.072673 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.115675 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.160676 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.205676 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.247677 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.288679 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.328669 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.369681 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.410682 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.450673 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.489690 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.530685 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.571686 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.612687 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:49.653688 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.696690 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.736680 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.777692 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.818693 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.859695 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.901695 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.942696 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:49.985686 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.026699 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.068700 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.109700 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.149702 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.190703 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.231692 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.273705 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.316852 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.357848 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.398848 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.439850 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.480851 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.521852 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.566853 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.612843 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.652855 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.692856 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.733857 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.773859 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.814860 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.854860 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.895861 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.936861 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:50.977863 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.019864 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.061866 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.104858 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.144867 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.185869 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.225871 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.268872 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.309873 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.350862 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.391874 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.431875 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.472865 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.513876 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.554878 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.597880 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:51.639882 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.681881 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.722882 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.763883 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.805885 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.846886 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.887887 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.928889 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:51.969877 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.010890 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.052891 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.093893 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.134894 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.175894 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.216895 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.259896 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.301886 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.342898 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.384889 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.427902 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.470901 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.512904 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.555904 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.600419 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.641008 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.684008 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.725893 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.768894 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.809895 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.852896 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.893897 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.940901 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:52.981900 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.023900 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.063902 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.103894 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.144904 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.185905 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.226906 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.267907 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.307908 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.349909 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.390912 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.431911 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.473913 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.513913 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.554914 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.595916 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:53.636916 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.677917 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.718919 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.759920 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.799921 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.841921 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.881924 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.923924 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:53.963925 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.004925 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.047219 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.087211 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.128222 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.169211 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.210736 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.252737 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.292737 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.333739 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.374740 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.414741 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.455741 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.495743 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.537745 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.578745 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.621746 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.661747 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.703740 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.745751 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.786751 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.829739 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.871741 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.915744 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:54.958754 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.001744 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.044764 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.088745 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.131747 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.171750 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.212761 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.254764 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.295763 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.342765 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.382755 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.423767 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.463757 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.504770 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.544770 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.590761 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:55.630772 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.671774 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.711775 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.752776 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.798777 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.840779 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.880779 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.926781 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:55.969780 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.011770 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.053783 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.095828 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.136829 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.177830 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.219831 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.262832 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.307834 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.348837 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.390836 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.430837 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.472830 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.512839 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.554839 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.599840 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.640841 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.681843 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.723844 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.765847 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.805833 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.846847 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.888854 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.941850 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:56.982851 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.023851 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.064853 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.107858 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.148843 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.190856 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.231857 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.273858 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.317858 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.358860 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.398861 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.442862 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.483862 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.524865 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.564866 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.608866 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:57.649870 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.691870 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.731870 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.772871 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.816872 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.857873 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.898874 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.941875 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:57.984863 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.027865 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.070866 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.112867 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.154869 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.196870 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.238870 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.279872 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.320873 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.361873 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.403874 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.447364 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.489353 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.532370 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.576359 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.620176 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.663206 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.706210 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.748208 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.791213 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.834211 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.877211 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.918213 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:58.960207 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.002502 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.045492 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.088503 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.132504 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.173495 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.215509 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.256509 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.297509 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.344500 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.384512 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.425513 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.471934 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.513942 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.554943 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.603936 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.670933 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:14:59.723858 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.772858 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.816868 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.855861 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.896877 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.939864 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:14:59.979866 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.020877 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.063878 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.103879 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.144881 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.187881 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.227883 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.268883 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.310885 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.357875 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.397887 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.439889 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.479888 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.520890 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.562891 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.610902 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.653887 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.697893 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.739896 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.781897 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.821899 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.865898 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.908900 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:00.958890 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.001903 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.047902 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.095969 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.140910 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.182897 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.225899 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.267910 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.315212 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.357214 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.398204 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.439216 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.480206 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.521218 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.563219 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.603220 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.644227 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.691223 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.737224 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:01.788226 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.834218 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.879216 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.925229 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:01.970233 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.013235 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.060242 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.111227 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.155223 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.194239 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.235236 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.275238 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.318239 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.358240 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.399241 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.445231 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.500231 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.551252 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.599234 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.647234 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.688238 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.729238 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.771250 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.816261 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.857256 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.897263 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.939243 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:02.980247 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.022246 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.066262 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.109246 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.150247 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.192270 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.233252 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.274264 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.316274 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.357275 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.397273 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.440272 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.479259 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.520274 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.558271 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.601272 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.642262 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.685264 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.732266 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.773265 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.823266 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:03.866268 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.908280 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.949272 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:03.992269 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.037280 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.078272 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.125277 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.173275 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.214280 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.255290 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.295290 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.335279 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.375292 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.416282 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.458963 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.523952 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.562956 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.602955 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.644968 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.687969 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.729970 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.770971 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.810971 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.850972 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.890973 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.933974 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:04.974975 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.016724 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.057734 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.098726 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.140727 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.180728 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.220729 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.261730 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.302731 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.341732 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.381733 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.422735 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.463736 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.508737 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.554738 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.598739 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.642740 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.682832 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.726001 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.769431 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.810432 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.852440 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:05.896424 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.943436 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:05.986437 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.031429 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.074439 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.117440 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.160442 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.202442 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.244444 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.286455 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.329445 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.370447 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.416449 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.458452 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.499450 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.541451 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.582452 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.623455 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.665102 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.705103 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.748104 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.791105 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.833107 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.874107 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.916109 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.957109 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:06.998111 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.039101 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.081112 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.123114 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.164117 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.205115 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.247117 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.288117 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.328118 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.376114 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.422109 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.475110 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.517113 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.562114 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.610115 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.649117 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.691116 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.733117 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.776117 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.819118 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.859123 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.906122 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:07.950135 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:07.993124 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.033125 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.075142 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.116143 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.159140 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.207135 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.253130 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.301135 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.342136 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.383134 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.425136 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.466140 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.507138 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.550139 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.590152 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.635142 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.679146 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.724143 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.770146 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.824150 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.873147 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.918150 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:08.962148 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.009150 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.062151 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.110154 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.152154 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.194156 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.239155 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.284158 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.332158 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.375160 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.416165 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.460164 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.506164 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.550166 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.592167 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.634566 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.675909 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.715911 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.756910 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.797912 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.851901 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.894901 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.935904 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:09.978903 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.026904 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:10.072907 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.113908 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.156910 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.197911 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.239911 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.281911 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.322913 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.362914 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.403914 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.445915 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.486918 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.531919 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.572919 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.613922 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.656939 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.700940 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.743925 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.786925 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.830942 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.871927 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.913933 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.958930 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:10.999931 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.042931 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.084938 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.128937 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.172940 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.214938 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.258936 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.298941 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.339951 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.380940 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.421942 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.474942 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.527943 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.575947 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.614947 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.655948 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.694961 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.737961 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.778963 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.819963 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.861965 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.900966 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.943989 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:11.982969 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.023970 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.064972 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:12.106973 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.146972 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.186974 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.227974 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.266978 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.307977 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.347977 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.388979 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.428979 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.469980 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.509983 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.550982 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.591536 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.633536 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.674530 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.719531 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.761531 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.801533 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.846530 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.887535 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.928532 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:12.969537 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.010538 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.052539 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.092541 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.134541 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.175542 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.216543 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.257545 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.297543 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.339547 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.380548 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.420549 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.461548 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.502550 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.543550 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.584550 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.625554 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.666555 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.708557 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.749558 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.790559 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.831555 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.871560 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.912561 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.953561 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:13.994563 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.035566 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:14.077566 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.121561 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.162562 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.206570 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.251577 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.294572 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.336584 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.381571 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.423574 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.464576 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.505584 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.546578 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.591011 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.632738 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.673199 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.714205 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.755200 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.795190 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.836191 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.876202 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.919194 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:14.960206 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.002207 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.044197 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.088029 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.130030 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.170031 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.213022 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.253034 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.301036 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.343206 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.385205 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.426208 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.468207 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.509209 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.552209 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.597199 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.638213 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.679935 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.720935 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.761948 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.803420 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.844414 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.894423 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.938423 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:15.981199 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.022200 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.064202 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:16.105203 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.147209 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.188204 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.236195 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.278208 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.321454 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.362454 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.403455 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.443464 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.484458 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.525456 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.567461 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.607459 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.648463 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.688464 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.728465 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.769462 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.809467 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.849468 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.889466 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.930470 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:16.970468 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.011468 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.053474 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.095472 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.136475 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.178473 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.220478 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.260475 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.302480 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.343522 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.384521 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.425521 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.465512 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.507523 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.548525 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.590526 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.632516 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.673527 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.712529 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.754530 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.794531 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.835531 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.876570 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.917581 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:17.958582 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.000572 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.042591 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:18.083581 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.124592 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.165594 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.207596 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.249585 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.290596 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.330598 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.371600 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.412589 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.454159 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.495162 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.537161 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.580167 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.620164 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.662165 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.703166 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.745165 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.787560 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.829562 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.878563 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.920560 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:18.961561 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.002566 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.044564 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.084566 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.125570 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.168571 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.209571 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.253569 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.295574 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.336575 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.379576 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.421573 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.462578 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.503579 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.544581 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.585589 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.626581 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.666581 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.707585 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.748581 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.789587 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.830582 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.872589 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.914116 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.956128 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:19.996895 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.038896 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:20.078892 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.119897 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.160898 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.202900 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.247898 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.289902 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.329903 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.371904 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.412901 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.452902 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.494907 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.537909 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.582921 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.624911 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.666912 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.708914 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.786908 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.824915 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.864916 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.905918 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.945914 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:20.987919 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.029921 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.070922 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.113930 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.155924 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.202932 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.246926 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.286936 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.328929 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.392927 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.438929 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.476929 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.515933 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.555934 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.596935 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.637944 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.679937 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.720939 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.763940 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.804949 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.846942 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.890943 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.932950 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:21.976942 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.020946 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.060948 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.100949 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:22.144957 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.186950 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.233952 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.276961 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.318962 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.361954 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.402956 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.444804 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.486606 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.527597 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.568600 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.610181 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.652179 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.695178 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.735168 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.777181 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.820174 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.862181 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.904183 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.946184 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:22.989178 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.031176 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.072189 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.114633 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.156647 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.198852 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.240852 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.281853 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.323843 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.365857 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.406855 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.446857 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.487848 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.528858 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.568859 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.610862 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.650851 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.692862 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.733865 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.774865 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.818856 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.860868 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.901869 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.941871 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:23.982860 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.023531 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.066532 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.108532 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:24.149542 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.191535 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.235542 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.277537 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.319546 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.362539 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.403540 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.445542 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.485539 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.532543 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.573722 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.615721 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.656712 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.698725 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.739715 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.780719 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.821492 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.862494 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.903497 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.945486 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:24.986496 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.027499 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.068500 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.110500 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.151501 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.192502 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.238356 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.285221 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.326217 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.368217 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.410218 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.451220 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.494220 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.535219 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.577223 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.619080 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.661080 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.702494 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.744024 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.785025 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.826015 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.867028 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.908028 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.949030 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:25.991019 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.031257 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.072257 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.114259 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:26.156258 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.198949 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.240700 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.282708 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.324710 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.371531 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.414533 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.455535 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.498111 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.540122 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.581124 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.622192 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.663192 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.704182 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.747184 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.791185 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.835187 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.876186 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.918188 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:26.960190 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.001189 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.042192 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.083192 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.124708 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.166721 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.209709 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.251559 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.292557 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.334559 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.374561 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.417562 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.458563 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.505565 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.546567 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.586567 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.627567 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.668568 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.709558 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.750570 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.790572 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.831573 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.872573 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.912575 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.953577 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:27.994575 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.035577 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.077578 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.118580 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:28.159581 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.200582 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.241392 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.282404 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.323394 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.365405 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.406408 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.448408 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.488398 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.529411 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.570411 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.611401 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.652402 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.693175 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.733165 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.773166 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.815178 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.856169 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.897181 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.938181 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:28.979183 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.019183 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.060188 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.102175 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.143187 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.184189 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.225189 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.266188 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.305192 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.347193 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.388194 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.429183 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.471195 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.513198 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.554198 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.595199 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.637200 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.678745 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.719735 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.760747 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.800748 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.840750 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.881751 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.921752 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:29.964752 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.006777 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.046755 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.085756 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:30.128756 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.168760 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.209759 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.250759 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.290761 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.330762 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.371763 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.411765 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.453279 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.495269 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.536282 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.577282 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.619277 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.662284 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.703283 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.745285 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.789328 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.830319 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.872319 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.914330 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:30.955333 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.002334 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.043325 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.088336 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.130614 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.172609 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.214618 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.258611 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.303619 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.344613 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.387614 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.430615 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.473622 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.515622 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.556625 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.598619 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.639628 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.680621 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.722622 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.769631 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.813620 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.856635 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.897626 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.939628 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:31.982625 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.022630 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.064632 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.110632 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:32.152633 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.192634 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.234632 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.274636 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.317634 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.359639 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.401636 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.443653 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.490648 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.532644 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.574645 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.615645 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.657642 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.698647 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.738649 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.780650 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.823647 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.864650 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.904653 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.946655 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:32.987655 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.028653 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.071656 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.112658 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.158655 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.200199 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.243190 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.284202 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.325191 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.365206 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.406205 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.451207 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.491196 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.532208 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.573198 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.613210 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.653200 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.694212 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.734214 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.775216 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.818202 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.859216 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.900217 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.941218 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:33.982218 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.023220 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.065221 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.106220 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:34.150224 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.192002 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.233542 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.274543 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.319545 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.360547 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.402546 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.444548 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.484549 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.526550 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.566540 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.613882 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.655893 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.697230 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.738231 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.781135 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.822130 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.864131 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.905133 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.946134 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:34.988135 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.029136 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.070139 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.111145 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.157144 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.197146 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.239147 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.280148 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.320150 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.361150 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.401152 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.441151 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.484157 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.525153 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.566156 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.607156 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.647156 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.688160 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.730160 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.772154 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.816162 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.859163 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.900164 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.942166 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:35.983166 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.029168 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.071170 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.112074 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:36.156072 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.197075 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.238076 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.280078 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.322078 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.363067 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.404069 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.445060 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.486060 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.528060 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.570052 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.611063 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.652064 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.693065 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.733068 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.775068 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.819069 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.861069 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.901064 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.942072 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:36.984072 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.026073 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.067076 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.114077 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.155077 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.196081 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.237080 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.278081 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.319082 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.360084 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.402073 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.442085 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.483085 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.525088 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.566088 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.608089 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.649081 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.691091 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.732090 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.773086 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.814094 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.856095 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.896097 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.937089 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:37.978101 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.022089 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.068101 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.109102 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:38.150108 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.192093 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.233111 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.274100 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.319107 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.361108 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.401109 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.442104 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.483111 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.524114 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.565114 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.606115 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.645116 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.687117 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.727118 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.768120 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.810120 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.851121 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.893122 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.936123 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:38.977118 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.018127 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.061127 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.104128 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.148122 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.190131 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.231125 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.273072 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.314062 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.356062 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.398064 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.442783 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.483790 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.526787 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.567784 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.610787 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.652795 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.693789 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.736790 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.778792 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.820799 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.862794 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.903795 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.944797 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:39.985804 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.027798 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.069799 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.111807 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:40.154809 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.196803 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.238804 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.279805 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.321813 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.363807 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.404808 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.445809 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.489819 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.532810 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.573812 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.614814 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.655814 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.696816 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.737817 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.781817 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.835814 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.879814 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.918821 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:40.963818 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.006818 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.051821 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.093819 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.142824 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.184825 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.226826 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.266830 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.307316 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.347306 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.388472 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.431473 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.472475 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.514483 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.556483 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.598479 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.640479 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.681488 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.724480 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.770482 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.811483 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.852481 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.894484 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.934486 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:41.975484 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.016485 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.058489 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.099491 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.140491 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:42.181488 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.222494 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.263494 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.304495 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.345493 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.386525 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.427537 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.468537 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.509539 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.551539 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.593541 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.635915 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.678914 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.720046 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.763035 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.805046 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.849406 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.890407 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.931401 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:42.973404 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.015003 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.056420 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.098421 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.139422 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.181430 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.223423 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.264425 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.306423 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.347425 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.388428 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.430430 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.472431 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.514438 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.557441 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.599434 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.641435 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.682442 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.724437 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.766439 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.809439 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.852447 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.893436 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.934443 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:43.977444 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.018455 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.060445 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.101447 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.142445 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:44.183445 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.224451 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.265451 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.306452 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.348448 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.389453 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.430456 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.471457 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.512453 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.553457 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.594471 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.636258 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.678260 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.719260 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.761261 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.803262 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.844263 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.885266 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.926265 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:44.967254 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.008268 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.049269 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.089269 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.131270 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.172272 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.213273 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.253273 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.295179 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.335184 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.377182 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.417183 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.457185 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.505174 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.545188 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.587137 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.629138 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.670928 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.710917 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.752932 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.793932 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.833931 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.874934 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.915935 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.956937 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:45.998937 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.039927 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.080942 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.121942 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:46.162941 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.202942 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.242937 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.283946 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.323945 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.365946 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.407032 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.450032 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.491604 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.532598 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.573596 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.614597 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.655598 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.696599 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.736603 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.778604 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.819613 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.861606 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.904608 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.945608 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:46.987616 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.030611 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.078612 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.120613 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.162622 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.204615 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.245616 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.288617 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.330619 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.371619 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.413617 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.455622 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.497630 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.539315 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.581313 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.622313 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.664303 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.707313 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.756316 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.802312 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.845319 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.887320 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.929313 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:47.970316 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.012325 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.054328 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.096315 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.139328 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:48.181329 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.222329 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.263329 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.305332 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.345326 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.387334 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.428335 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.471328 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.512336 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.553338 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.594339 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.635328 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.676334 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.718341 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.760343 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.802345 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.844348 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.885346 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.928384 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:48.969384 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.009387 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.051069 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.092069 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.134071 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.176072 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.218072 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.260073 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.302074 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.343068 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.385076 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.429067 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.471077 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.513073 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.554081 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.597083 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.638083 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.680083 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.723084 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.771413 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.814413 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.855412 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.895416 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.936417 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:49.979417 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.020429 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.063072 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.109071 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.150061 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:50.191672 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.233672 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.274961 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.316958 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.356960 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.397964 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.438966 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.480967 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.522974 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.564968 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.605970 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.647971 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.688979 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.730972 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.773975 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.818975 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.859973 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.901303 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.944091 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:50.986093 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.026094 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.068094 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.115095 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.157096 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.198931 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.240923 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.282926 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.324926 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.366934 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.408929 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.456926 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.498930 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.538929 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.580933 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.621934 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.662935 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.705943 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.747938 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.790945 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.836939 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.877948 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.920948 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:51.962950 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:52.003951 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:52.045952 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:52.087946 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:52.129946 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:52.172948 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>error</th>\n",
       "      <th>importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>D27</td>\n",
       "      <td>0.620942</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>D1188</td>\n",
       "      <td>0.567100</td>\n",
       "      <td>0.913289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>D998</td>\n",
       "      <td>0.564381</td>\n",
       "      <td>0.908910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>D1059</td>\n",
       "      <td>0.564303</td>\n",
       "      <td>0.908784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>D1049</td>\n",
       "      <td>0.563695</td>\n",
       "      <td>0.907807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>D51</td>\n",
       "      <td>0.563569</td>\n",
       "      <td>0.907604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>D1402</td>\n",
       "      <td>0.563399</td>\n",
       "      <td>0.907328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>D1159</td>\n",
       "      <td>0.562970</td>\n",
       "      <td>0.906639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>D1167</td>\n",
       "      <td>0.562898</td>\n",
       "      <td>0.906522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>D1271</td>\n",
       "      <td>0.562790</td>\n",
       "      <td>0.906348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>D1069</td>\n",
       "      <td>0.562584</td>\n",
       "      <td>0.906016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>D1241</td>\n",
       "      <td>0.562503</td>\n",
       "      <td>0.905886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>D1270</td>\n",
       "      <td>0.562468</td>\n",
       "      <td>0.905829</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>D1009</td>\n",
       "      <td>0.561970</td>\n",
       "      <td>0.905028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>D1067</td>\n",
       "      <td>0.561968</td>\n",
       "      <td>0.905024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>D1093</td>\n",
       "      <td>0.561948</td>\n",
       "      <td>0.904993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>D1352</td>\n",
       "      <td>0.561881</td>\n",
       "      <td>0.904884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>D199</td>\n",
       "      <td>0.561806</td>\n",
       "      <td>0.904763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>D1111</td>\n",
       "      <td>0.561793</td>\n",
       "      <td>0.904743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>D1010</td>\n",
       "      <td>0.561660</td>\n",
       "      <td>0.904529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>D1025</td>\n",
       "      <td>0.561574</td>\n",
       "      <td>0.904390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>D1012</td>\n",
       "      <td>0.561568</td>\n",
       "      <td>0.904381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>D1334</td>\n",
       "      <td>0.561510</td>\n",
       "      <td>0.904287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>D1359</td>\n",
       "      <td>0.561415</td>\n",
       "      <td>0.904135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>D1191</td>\n",
       "      <td>0.561371</td>\n",
       "      <td>0.904064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>D958</td>\n",
       "      <td>0.561314</td>\n",
       "      <td>0.903972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>D1118</td>\n",
       "      <td>0.561263</td>\n",
       "      <td>0.903890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>D1405</td>\n",
       "      <td>0.561137</td>\n",
       "      <td>0.903686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>D194</td>\n",
       "      <td>0.561109</td>\n",
       "      <td>0.903641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>D1100</td>\n",
       "      <td>0.561059</td>\n",
       "      <td>0.903561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1746</th>\n",
       "      <td>D1625</td>\n",
       "      <td>0.555985</td>\n",
       "      <td>0.895390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1747</th>\n",
       "      <td>D1075</td>\n",
       "      <td>0.555980</td>\n",
       "      <td>0.895382</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1748</th>\n",
       "      <td>D1684</td>\n",
       "      <td>0.555964</td>\n",
       "      <td>0.895355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1749</th>\n",
       "      <td>D1523</td>\n",
       "      <td>0.555881</td>\n",
       "      <td>0.895222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1750</th>\n",
       "      <td>D1693</td>\n",
       "      <td>0.555783</td>\n",
       "      <td>0.895064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1751</th>\n",
       "      <td>D953</td>\n",
       "      <td>0.555651</td>\n",
       "      <td>0.894852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1752</th>\n",
       "      <td>D1234</td>\n",
       "      <td>0.555554</td>\n",
       "      <td>0.894695</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1753</th>\n",
       "      <td>D61</td>\n",
       "      <td>0.555499</td>\n",
       "      <td>0.894607</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1754</th>\n",
       "      <td>D1427</td>\n",
       "      <td>0.555428</td>\n",
       "      <td>0.894493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1755</th>\n",
       "      <td>D1179</td>\n",
       "      <td>0.555384</td>\n",
       "      <td>0.894421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1756</th>\n",
       "      <td>D1170</td>\n",
       "      <td>0.555237</td>\n",
       "      <td>0.894185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1757</th>\n",
       "      <td>D1543</td>\n",
       "      <td>0.555220</td>\n",
       "      <td>0.894158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1758</th>\n",
       "      <td>D1195</td>\n",
       "      <td>0.555140</td>\n",
       "      <td>0.894028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1759</th>\n",
       "      <td>D1294</td>\n",
       "      <td>0.555076</td>\n",
       "      <td>0.893926</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1760</th>\n",
       "      <td>D1454</td>\n",
       "      <td>0.555045</td>\n",
       "      <td>0.893875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1761</th>\n",
       "      <td>D1146</td>\n",
       "      <td>0.555012</td>\n",
       "      <td>0.893822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1762</th>\n",
       "      <td>D1426</td>\n",
       "      <td>0.555009</td>\n",
       "      <td>0.893817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1763</th>\n",
       "      <td>D1337</td>\n",
       "      <td>0.554909</td>\n",
       "      <td>0.893657</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1764</th>\n",
       "      <td>D1492</td>\n",
       "      <td>0.554867</td>\n",
       "      <td>0.893590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1765</th>\n",
       "      <td>D1043</td>\n",
       "      <td>0.554814</td>\n",
       "      <td>0.893504</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1766</th>\n",
       "      <td>D1060</td>\n",
       "      <td>0.554770</td>\n",
       "      <td>0.893433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1767</th>\n",
       "      <td>D1290</td>\n",
       "      <td>0.554696</td>\n",
       "      <td>0.893313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1768</th>\n",
       "      <td>D1423</td>\n",
       "      <td>0.554650</td>\n",
       "      <td>0.893239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1769</th>\n",
       "      <td>D1338</td>\n",
       "      <td>0.554276</td>\n",
       "      <td>0.892636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1770</th>\n",
       "      <td>D1421</td>\n",
       "      <td>0.554225</td>\n",
       "      <td>0.892555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1771</th>\n",
       "      <td>D1442</td>\n",
       "      <td>0.554007</td>\n",
       "      <td>0.892204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1772</th>\n",
       "      <td>D1407</td>\n",
       "      <td>0.553966</td>\n",
       "      <td>0.892138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1773</th>\n",
       "      <td>D1681</td>\n",
       "      <td>0.553890</td>\n",
       "      <td>0.892016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1774</th>\n",
       "      <td>D1415</td>\n",
       "      <td>0.552120</td>\n",
       "      <td>0.889166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1775</th>\n",
       "      <td>D1286</td>\n",
       "      <td>0.551627</td>\n",
       "      <td>0.888371</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1776 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       name     error  importance\n",
       "0       D27  0.620942    1.000000\n",
       "1     D1188  0.567100    0.913289\n",
       "2      D998  0.564381    0.908910\n",
       "3     D1059  0.564303    0.908784\n",
       "4     D1049  0.563695    0.907807\n",
       "...     ...       ...         ...\n",
       "1771  D1442  0.554007    0.892204\n",
       "1772  D1407  0.553966    0.892138\n",
       "1773  D1681  0.553890    0.892016\n",
       "1774  D1415  0.552120    0.889166\n",
       "1775  D1286  0.551627    0.888371\n",
       "\n",
       "[1776 rows x 3 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Rank the features\n",
    "from IPython.display import display, HTML\n",
    "\n",
    "names = list(df_train.columns) # x+y column names\n",
    "names.remove(\"Activity\") # remove the target(y)\n",
    "rank = perturbation_rank(model, x_test, y_test, names, False)\n",
    "display(rank)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Neural Network Ensemble\n",
    "\n",
    "A neural network ensemble combines neural network predictions with other models. The program determines the exact blend of all of these models by logistic regression. The following code performs this blend for a classification.  If you present the final predictions from the ensemble to Kaggle, you will see that the result is very accurate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading data...\n",
      "Model: 0 : <tensorflow.python.keras.wrappers.scikit_learn.KerasClassifier object at 0x000001962E56C9B0>\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:55.228026 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3375/3375 [==============================] - 1s 177us/sample - loss: 0.6070\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:55.831674 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:55.875673 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold #0: loss=0.5540271252836152\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:56.294684 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3375/3375 [==============================] - 0s 147us/sample - loss: 0.6303\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:56.795923 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:56.836932 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold #1: loss=0.5445477180683725\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:57.260947 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3375/3375 [==============================] - 0s 147us/sample - loss: 0.6088\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:57.760948 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:57.803957 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold #2: loss=0.5401721048981553\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:58.222973 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3375/3375 [==============================] - 1s 181us/sample - loss: 0.5985\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:58.837977 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:58.880986 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold #3: loss=0.4995116027234963\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:59.298736 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3376/3376 [==============================] - 1s 149us/sample - loss: 0.6074\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:15:59.808040 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:15:59.852048 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold #4: loss=0.552265225014246\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:16:00.273070 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3376/3376 [==============================] - 1s 180us/sample - loss: 0.5967\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:16:00.884081 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:16:00.929075 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold #5: loss=0.5920530577133667\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:16:01.352097 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3376/3376 [==============================] - 0s 147us/sample - loss: 0.5912\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:16:01.852018 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:16:01.896020 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold #6: loss=0.5066588401181843\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:16:02.313030 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3377/3377 [==============================] - 0s 144us/sample - loss: 0.6281\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:16:02.804981 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:16:02.971660 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold #7: loss=0.5611699477808485\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:16:03.392671 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3377/3377 [==============================] - 0s 146us/sample - loss: 0.6233\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:16:03.889980 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:16:03.932979 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold #8: loss=0.5665587789108542\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:16:04.351011 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3377/3377 [==============================] - 0s 146us/sample - loss: 0.6127\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0819 15:16:04.848786 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n",
      "W0819 15:16:04.890800 12720 training_utils.py:1210] When passing input data as arrays, do not specify `steps_per_epoch`/`steps` argument. Please use `batch_size` instead.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold #9: loss=0.5394746160266767\n",
      "KerasClassifier: Mean loss=0.5456439016537816\n",
      "Model: 1 : KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
      "                     metric_params=None, n_jobs=None, n_neighbors=3, p=2,\n",
      "                     weights='uniform')\n",
      "Fold #0: loss=3.606678388314123\n",
      "Fold #1: loss=2.2197228940978317\n",
      "Fold #2: loss=3.6717523663107237\n",
      "Fold #3: loss=2.5045156203944594\n",
      "Fold #4: loss=4.443553550438037\n",
      "Fold #5: loss=4.410524301688227\n",
      "Fold #6: loss=3.400455469543658\n",
      "Fold #7: loss=3.0885474338547683\n",
      "Fold #8: loss=2.1219335323249253\n",
      "Fold #9: loss=3.0613772690497245\n",
      "KNeighborsClassifier: Mean loss=3.2529060826016476\n",
      "Model: 2 : RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
      "                       max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
      "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                       min_samples_leaf=1, min_samples_split=2,\n",
      "                       min_weight_fraction_leaf=0.0, n_estimators=100,\n",
      "                       n_jobs=-1, oob_score=False, random_state=None, verbose=0,\n",
      "                       warm_start=False)\n",
      "Fold #0: loss=0.4657177982691548\n",
      "Fold #1: loss=0.426532747936268\n",
      "Fold #2: loss=0.5555646241353958\n",
      "Fold #3: loss=0.4158839384284288\n",
      "Fold #4: loss=0.47983725577294734\n",
      "Fold #5: loss=0.4816998209698582\n",
      "Fold #6: loss=0.40791842169050496\n",
      "Fold #7: loss=0.4769556823742715\n",
      "Fold #8: loss=0.45170716330819044\n",
      "Fold #9: loss=0.4679788790578769\n",
      "RandomForestClassifier: Mean loss=0.4629796331942897\n",
      "Model: 3 : RandomForestClassifier(bootstrap=True, class_weight=None, criterion='entropy',\n",
      "                       max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
      "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                       min_samples_leaf=1, min_samples_split=2,\n",
      "                       min_weight_fraction_leaf=0.0, n_estimators=100,\n",
      "                       n_jobs=-1, oob_score=False, random_state=None, verbose=0,\n",
      "                       warm_start=False)\n",
      "Fold #0: loss=0.4511847247326708\n",
      "Fold #1: loss=0.42856095931249005\n",
      "Fold #2: loss=0.48374399227671866\n",
      "Fold #3: loss=0.41861499005015385\n",
      "Fold #4: loss=0.4815877201194617\n",
      "Fold #5: loss=0.4978144986688418\n",
      "Fold #6: loss=0.40096963197346613\n",
      "Fold #7: loss=0.4546784305314136\n",
      "Fold #8: loss=0.4465855989880112\n",
      "Fold #9: loss=0.4695067213178049\n",
      "RandomForestClassifier: Mean loss=0.45332472679710323\n",
      "Model: 4 : ExtraTreesClassifier(bootstrap=False, class_weight=None, criterion='gini',\n",
      "                     max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
      "                     min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                     min_samples_leaf=1, min_samples_split=2,\n",
      "                     min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=-1,\n",
      "                     oob_score=False, random_state=None, verbose=0,\n",
      "                     warm_start=False)\n",
      "Fold #0: loss=0.45496751079363495\n",
      "Fold #1: loss=0.4221988146584036\n",
      "Fold #2: loss=0.5864779075767735\n",
      "Fold #3: loss=0.4135387147555288\n",
      "Fold #4: loss=0.49165543640777337\n",
      "Fold #5: loss=0.5826853127129209\n",
      "Fold #6: loss=0.4992293347545809\n",
      "Fold #7: loss=0.5696776278324083\n",
      "Fold #8: loss=0.5377009614825886\n",
      "Fold #9: loss=0.6233110429105153\n",
      "ExtraTreesClassifier: Mean loss=0.5181442663885129\n",
      "Model: 5 : ExtraTreesClassifier(bootstrap=False, class_weight=None, criterion='entropy',\n",
      "                     max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
      "                     min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                     min_samples_leaf=1, min_samples_split=2,\n",
      "                     min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=-1,\n",
      "                     oob_score=False, random_state=None, verbose=0,\n",
      "                     warm_start=False)\n",
      "Fold #0: loss=0.44825346440152214\n",
      "Fold #1: loss=0.40736044850901787\n",
      "Fold #2: loss=0.5775460796610716\n",
      "Fold #3: loss=0.42126274280011555\n",
      "Fold #4: loss=0.4949968282719872\n",
      "Fold #5: loss=0.5074636989761913\n",
      "Fold #6: loss=0.4176899941799664\n",
      "Fold #7: loss=0.6431685288404098\n",
      "Fold #8: loss=0.5444145998368657\n",
      "Fold #9: loss=0.46770812815232515\n",
      "ExtraTreesClassifier: Mean loss=0.4929864513629473\n",
      "Model: 6 : GradientBoostingClassifier(criterion='friedman_mse', init=None,\n",
      "                           learning_rate=0.05, loss='deviance', max_depth=6,\n",
      "                           max_features=None, max_leaf_nodes=None,\n",
      "                           min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "                           min_samples_leaf=1, min_samples_split=2,\n",
      "                           min_weight_fraction_leaf=0.0, n_estimators=50,\n",
      "                           n_iter_no_change=None, presort='auto',\n",
      "                           random_state=None, subsample=0.5, tol=0.0001,\n",
      "                           validation_fraction=0.1, verbose=0,\n",
      "                           warm_start=False)\n",
      "Fold #0: loss=0.4804937760760543\n",
      "Fold #1: loss=0.4581518207794376\n",
      "Fold #2: loss=0.48285850183766754\n",
      "Fold #3: loss=0.4399317867686944\n",
      "Fold #4: loss=0.498996013541783\n",
      "Fold #5: loss=0.5045013322264401\n",
      "Fold #6: loss=0.45389969767473726\n",
      "Fold #7: loss=0.4623607887672741\n",
      "Fold #8: loss=0.4551702632706958\n",
      "Fold #9: loss=0.4709134176721333\n",
      "GradientBoostingClassifier: Mean loss=0.47072773986149186\n",
      "\n",
      "Blending models.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import os\n",
    "import pandas as pd\n",
    "import math\n",
    "from tensorflow.keras.wrappers.scikit_learn import KerasClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.ensemble import RandomForestClassifier, \n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "PATH = \"./data/\"\n",
    "SHUFFLE = False\n",
    "FOLDS = 10\n",
    "\n",
    "def build_ann(input_size,classes,neurons):\n",
    "    model = Sequential()\n",
    "    model.add(Dense(neurons, input_dim=input_size, activation='relu'))\n",
    "    model.add(Dense(1))\n",
    "    model.add(Dense(classes,activation='softmax'))\n",
    "    model.compile(loss='categorical_crossentropy', optimizer='adam')\n",
    "    return model\n",
    "\n",
    "def mlogloss(y_test, preds):\n",
    "    epsilon = 1e-15\n",
    "    sum = 0\n",
    "    for row in zip(preds,y_test):\n",
    "        x = row[0][row[1]]\n",
    "        x = max(epsilon,x)\n",
    "        x = min(1-epsilon,x)\n",
    "        sum+=math.log(x)\n",
    "    return( (-1/len(preds))*sum)\n",
    "\n",
    "def stretch(y):\n",
    "    return (y - y.min()) / (y.max() - y.min())\n",
    "\n",
    "\n",
    "def blend_ensemble(x, y, x_submit):\n",
    "    kf = StratifiedKFold(FOLDS)\n",
    "    folds = list(kf.split(x,y))\n",
    "\n",
    "    models = [\n",
    "        KerasClassifier(build_fn=build_ann,neurons=20,\n",
    "                    input_size=x.shape[1],classes=2),\n",
    "        KNeighborsClassifier(n_neighbors=3),\n",
    "        RandomForestClassifier(n_estimators=100, n_jobs=-1, \n",
    "                               criterion='gini'),\n",
    "        RandomForestClassifier(n_estimators=100, n_jobs=-1, \n",
    "                               criterion='entropy'),\n",
    "        ExtraTreesClassifier(n_estimators=100, n_jobs=-1, \n",
    "                             criterion='gini'),\n",
    "        ExtraTreesClassifier(n_estimators=100, n_jobs=-1, \n",
    "                             criterion='entropy'),\n",
    "        GradientBoostingClassifier(learning_rate=0.05, \n",
    "                subsample=0.5, max_depth=6, n_estimators=50)]\n",
    "\n",
    "    dataset_blend_train = np.zeros((x.shape[0], len(models)))\n",
    "    dataset_blend_test = np.zeros((x_submit.shape[0], len(models)))\n",
    "\n",
    "    for j, model in enumerate(models):\n",
    "        print(\"Model: {} : {}\".format(j, model) )\n",
    "        fold_sums = np.zeros((x_submit.shape[0], len(folds)))\n",
    "        total_loss = 0\n",
    "        for i, (train, test) in enumerate(folds):\n",
    "            x_train = x[train]\n",
    "            y_train = y[train]\n",
    "            x_test = x[test]\n",
    "            y_test = y[test]\n",
    "            model.fit(x_train, y_train)\n",
    "            pred = np.array(model.predict_proba(x_test))\n",
    "            # pred = model.predict_proba(x_test)\n",
    "            dataset_blend_train[test, j] = pred[:, 1]\n",
    "            pred2 = np.array(model.predict_proba(x_submit))\n",
    "            #fold_sums[:, i] = model.predict_proba(x_submit)[:, 1]\n",
    "            fold_sums[:, i] = pred2[:, 1]\n",
    "            loss = mlogloss(y_test, pred)\n",
    "            total_loss+=loss\n",
    "            print(\"Fold #{}: loss={}\".format(i,loss))\n",
    "        print(\"{}: Mean loss={}\".format(model.__class__.__name__,\n",
    "                                        total_loss/len(folds)))\n",
    "        dataset_blend_test[:, j] = fold_sums.mean(1)\n",
    "\n",
    "    print()\n",
    "    print(\"Blending models.\")\n",
    "    blend = LogisticRegression(solver='lbfgs')\n",
    "    blend.fit(dataset_blend_train, y)\n",
    "    return blend.predict_proba(dataset_blend_test)\n",
    "\n",
    "if __name__ == '__main__':\n",
    "\n",
    "    np.random.seed(42)  # seed to shuffle the train set\n",
    "\n",
    "    print(\"Loading data...\")\n",
    "    filename_train = os.path.join(PATH, \"bio_train.csv\")\n",
    "    df_train = pd.read_csv(filename_train, na_values=['NA', '?'])\n",
    "\n",
    "    filename_submit = os.path.join(PATH, \"bio_test.csv\")\n",
    "    df_submit = pd.read_csv(filename_submit, na_values=['NA', '?'])\n",
    "\n",
    "    predictors = list(df_train.columns.values)\n",
    "    predictors.remove('Activity')\n",
    "    x = df_train[predictors].values\n",
    "    y = df_train['Activity']\n",
    "    x_submit = df_submit.values\n",
    "\n",
    "    if SHUFFLE:\n",
    "        idx = np.random.permutation(y.size)\n",
    "        x = x[idx]\n",
    "        y = y[idx]\n",
    "\n",
    "    submit_data = blend_ensemble(x, y, x_submit)\n",
    "    submit_data = stretch(submit_data)\n",
    "\n",
    "    ####################\n",
    "    # Build submit file\n",
    "    ####################\n",
    "    ids = [id+1 for id in range(submit_data.shape[0])]\n",
    "    submit_filename = os.path.join(PATH, \"bio_submit.csv\")\n",
    "    submit_df = pd.DataFrame({'MoleculeId': ids, \n",
    "                              'PredictedProbability': \n",
    "                              submit_data[:, 1]},\n",
    "                             columns=['MoleculeId',\n",
    "                            'PredictedProbability'])\n",
    "    submit_df.to_csv(submit_filename, index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3.7 (tensorflow)",
   "language": "python",
   "name": "tensorflow"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}