02_2_simple-RNN-diffrent-sequence-length.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<center><div style=\"direction:rtl;font-family:B Lotus, B Nazanin, Tahoma\">به نام خدا</div></center>\n",
    "<img src=\"./logo.png\" alt=\"class.vision\" style=\"width: 200px;\"/>\n",
    "<h1><center><div style=\"direction:rtl;font-family:B Lotus, B Nazanin, Tahoma\">تغییر در طول بازه‌های زمانی ورودی شبکه - روش 2</div></center></h1>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, SimpleRNN\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "%matplotlib inline\n",
    "\n",
    "t = np.arange(0,1500)\n",
    "x = np.sin(0.02*t)+ np.random.rand(1500) * 2\n",
    "\n",
    "train,test = x[0:1000], x[1000:]\n",
    "\n",
    "# convert into dataset data and label\n",
    "def convertToDataset(data, step):\n",
    "    #data = np.append(data,np.repeat(data[-1,],step))\n",
    "    X, Y =[], []\n",
    "    for i in range(len(data)-step):\n",
    "        d=i+step  \n",
    "        X.append(data[i:d,])\n",
    "        Y.append(data[d,])\n",
    "    return np.array(X), np.array(Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_step = 10\n",
    "test_step = 20\n",
    "\n",
    "trainX,trainY =convertToDataset(train,train_step)\n",
    "testX,testY =convertToDataset(test,test_step)\n",
    "\n",
    "trainX = np.reshape(trainX, (trainX.shape[0],  trainX.shape[1], 1))\n",
    "testX = np.reshape(testX, (testX.shape[0],testX.shape[1], 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(990, 10, 1)\n",
      "(480, 20, 1)\n"
     ]
    }
   ],
   "source": [
    "print(trainX.shape)\n",
    "print(testX.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <div style=\"direction:rtl;text-align:right;font-family:B Lotus, B Nazanin, Tahoma\">معماری شبکه و compile آن</div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_model(sequence_length):\n",
    "    model = Sequential()\n",
    "    model.add(SimpleRNN(units=64, input_shape=(sequence_length, 1),  activation=\"tanh\"))\n",
    "    model.add(Dense(1))\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "simple_rnn (SimpleRNN)       (None, 64)                4224      \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1)                 65        \n",
      "=================================================================\n",
      "Total params: 4,289\n",
      "Trainable params: 4,289\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = build_model(10)\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor 'simple_rnn_input:0' shape=(?, 10, 1) dtype=float32>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.input"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <div style=\"direction:rtl;text-align:right;font-family:B Lotus, B Nazanin, Tahoma\">آموزش مدل</div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss='mean_squared_error', optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Directory where the checkpoints will be saved\n",
    "checkpoint_dir = './notebook02_training_checkpoints'\n",
    "# Name of the checkpoint files\n",
    "checkpoint_prefix = os.path.join(checkpoint_dir, \"ckpt_{epoch}\")\n",
    "\n",
    "checkpoint_callback=ModelCheckpoint(\n",
    "    filepath=checkpoint_prefix,\n",
    "    save_weights_only=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 3s - loss: 0.4920\n",
      "Epoch 2/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.4102\n",
      "Epoch 3/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.4077\n",
      "Epoch 4/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3988\n",
      "Epoch 5/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3925\n",
      "Epoch 6/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.4015\n",
      "Epoch 7/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.4000\n",
      "Epoch 8/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3984\n",
      "Epoch 9/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3925\n",
      "Epoch 10/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3851\n",
      "Epoch 11/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3976\n",
      "Epoch 12/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3905\n",
      "Epoch 13/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3893\n",
      "Epoch 14/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3872\n",
      "Epoch 15/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3872\n",
      "Epoch 16/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3846\n",
      "Epoch 17/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3850\n",
      "Epoch 18/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3761\n",
      "Epoch 19/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3787\n",
      "Epoch 20/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " - 1s - loss: 0.3866\n",
      "Epoch 21/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3835\n",
      "Epoch 22/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3744\n",
      "Epoch 23/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3782\n",
      "Epoch 24/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3821\n",
      "Epoch 25/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3776\n",
      "Epoch 26/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3718\n",
      "Epoch 27/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3717\n",
      "Epoch 28/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3737\n",
      "Epoch 29/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3653\n",
      "Epoch 30/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3704\n",
      "Epoch 31/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3659\n",
      "Epoch 32/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3701\n",
      "Epoch 33/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3663\n",
      "Epoch 34/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3637\n",
      "Epoch 35/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3555\n",
      "Epoch 36/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3543\n",
      "Epoch 37/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3566\n",
      "Epoch 38/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3466\n",
      "Epoch 39/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3438\n",
      "Epoch 40/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " - 1s - loss: 0.3463\n",
      "Epoch 41/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3561\n",
      "Epoch 42/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3459\n",
      "Epoch 43/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3381\n",
      "Epoch 44/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3365\n",
      "Epoch 45/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3412\n",
      "Epoch 46/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3426\n",
      "Epoch 47/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3295\n",
      "Epoch 48/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3220\n",
      "Epoch 49/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3177\n",
      "Epoch 50/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3178\n",
      "Epoch 51/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3078\n",
      "Epoch 52/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3301\n",
      "Epoch 53/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3137\n",
      "Epoch 54/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.3016\n",
      "Epoch 55/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2949\n",
      "Epoch 56/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2980\n",
      "Epoch 57/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2944\n",
      "Epoch 58/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2972\n",
      "Epoch 59/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2902\n",
      "Epoch 60/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " - 1s - loss: 0.2859\n",
      "Epoch 61/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2773\n",
      "Epoch 62/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2800\n",
      "Epoch 63/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2620\n",
      "Epoch 64/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2646\n",
      "Epoch 65/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2577\n",
      "Epoch 66/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2543\n",
      "Epoch 67/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2483\n",
      "Epoch 68/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2405\n",
      "Epoch 69/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2337\n",
      "Epoch 70/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2342\n",
      "Epoch 71/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2173\n",
      "Epoch 72/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2247\n",
      "Epoch 73/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2170\n",
      "Epoch 74/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2205\n",
      "Epoch 75/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2102\n",
      "Epoch 76/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2108\n",
      "Epoch 77/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1987\n",
      "Epoch 78/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1991\n",
      "Epoch 79/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.2037\n",
      "Epoch 80/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " - 1s - loss: 0.1884\n",
      "Epoch 81/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1865\n",
      "Epoch 82/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1899\n",
      "Epoch 83/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1786\n",
      "Epoch 84/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1767\n",
      "Epoch 85/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1744\n",
      "Epoch 86/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1722\n",
      "Epoch 87/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1584\n",
      "Epoch 88/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1574\n",
      "Epoch 89/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1525\n",
      "Epoch 90/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1518\n",
      "Epoch 91/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1574\n",
      "Epoch 92/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1444\n",
      "Epoch 93/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1486\n",
      "Epoch 94/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1365\n",
      "Epoch 95/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1379\n",
      "Epoch 96/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1279\n",
      "Epoch 97/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1311\n",
      "Epoch 98/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1216\n",
      "Epoch 99/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      " - 1s - loss: 0.1223\n",
      "Epoch 100/100\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.Adam object at 0x000002AD84DA83C8>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " - 1s - loss: 0.1285\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(trainX,trainY, epochs=100, batch_size=16,\n",
    "                    verbose=2, callbacks=[checkpoint_callback])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <div style=\"direction:rtl;text-align:right;font-family:B Lotus, B Nazanin, Tahoma\"> ارزیابی مدل</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'./notebook02_training_checkpoints\\\\ckpt_100'"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.train.latest_checkpoint(checkpoint_dir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = build_model(20)\n",
    "\n",
    "model.load_weights(tf.train.latest_checkpoint(checkpoint_dir))\n",
    "\n",
    "model.compile(loss='mean_squared_error', optimizer='adam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.841938066482544\n"
     ]
    }
   ],
   "source": [
    "trainScore = model.evaluate(testX, testY, verbose=0)\n",
    "print(trainScore)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "<div style=\"direction:rtl;text-align:right;font-family:B Lotus, B Nazanin, Tahoma\"> دوره پیشرفته یادگیری عمیق<br>علیرضا اخوان پور<br>  آبان و آذر 1399<br>\n",
    "</div>\n",
    "<a href=\"http://class.vision\">Class.Vision</a> - <a href=\"http://AkhavanPour.ir\">AkhavanPour.ir</a> - <a href=\"https://github.com/Alireza-Akhavan/\">GitHub</a>\n",
    "\n",
    "</div>"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}