AE.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.python.keras import Input\n",
    "from tensorflow.python.keras.models import Model\n",
    "from tensorflow.python.keras.engine.keras_tensor import KerasTensor\n",
    "from tensorflow.python.keras.engine.functional import Functional\n",
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.python.keras.regularizers import l2\n",
    "import os\n",
    "\n",
    "from tensorflow_model_optimization.python.core.sparsity.keras import prune, pruning_callbacks, pruning_schedule\n",
    "from tensorflow_model_optimization.sparsity.keras import strip_pruning, prune_low_magnitude\n",
    "\n",
    "from callbacks import all_callbacks\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import the MNIST data class and call it."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Original dataset and modified dataset with modifiable resolution is shown below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(60000, 64)\n"
     ]
    }
   ],
   "source": [
    "import MNIST_dataset as mnist\n",
    "size_final = 8\n",
    "\n",
    "data_zoom = mnist.MNISTData(size_final=size_final, color_depth=5)\n",
    "test = data_zoom.x_test\n",
    "\n",
    "#plt.imshow(test[0].reshape(size_final,size_final), cmap='gray')\n",
    "print(data_zoom.x_train.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": false,
    "editable": false,
    "nbgrader": {
     "cell_type": "markdown",
     "checksum": "cec58d97b4c05e29e9afdeb102569f20",
     "grade": false,
     "grade_id": "cell-6ee6a27aac2a7e5b",
     "locked": true,
     "schema_version": 3,
     "solution": false,
     "task": false
    }
   },
   "source": [
    "# Autoencoder using Keras\n",
    "The aim of this class is to implement a simple autoencoder for the MNIST data and then build an autoencoder that is\n",
    "able to classify MNIST data in its latent dimension.\n",
    "\n",
    "Code partially adapted from [Keras Documentation](https://blog.keras.io/building-autoencoders-in-keras.html)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "deletable": false,
    "editable": false,
    "nbgrader": {
     "cell_type": "code",
     "checksum": "8f093c3123887a615df510d5cb9d3bee",
     "grade": false,
     "grade_id": "cell-b3452809a3ed4ee3",
     "locked": true,
     "schema_version": 3,
     "solution": false,
     "task": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "class Autoencoder:\n",
    "    \"\"\"Autoencoder Class\"\"\"\n",
    "\n",
    "    def __init__(self, data: mnist.MNISTData, num_classes=10, pruned = False):\n",
    "        self.x_train = data.x_train\n",
    "        self.x_test = data.x_test\n",
    "        self.y_train = data.y_train\n",
    "        self.y_test = data.y_test\n",
    "\n",
    "        self.num_classes = num_classes\n",
    "        self.pruned = pruned\n",
    "        self.input_shape = (self.x_train.shape[-1],)\n",
    "        self.input = Input(shape=self.input_shape, name='encoder_input')\n",
    "        self.latent_dim = 2\n",
    "\n",
    "        self.encoder = None\n",
    "        self.encoder_model = None\n",
    "        self.decoder = None\n",
    "        self.latent_classifier = None\n",
    "        self.autoencoder = None\n",
    "\n",
    "        self.history = None\n",
    "\n",
    "        self.pruning_params = {\"pruning_schedule\" : pruning_schedule.ConstantSparsity(0.75, begin_step=2000, frequency=100)}\n",
    "    \n",
    "    def build_encoder(self):\n",
    "        \"\"\"Build your encoder architecture and store the output in self.encoder. \n",
    "        The final encoding dimension is 2.\"\"\"\n",
    "\n",
    "        self.encoder = Dense(32,  activation='relu')(self.input)\n",
    "        \n",
    "        if self.pruned:\n",
    "            self.encoder = prune_low_magnitude(Dense(16,  activation='relu'), **self.pruning_params)(self.encoder) \n",
    "        \n",
    "        if not self.pruned:\n",
    "            self.encoder = Dense(16,  activation='relu')(self.encoder)   \n",
    "\n",
    "        self.encoder = Dense(self.latent_dim,  activation='relu', name = 'encoder_output')(self.encoder)\n",
    "        #building a model for the encoder in order to be able to predict and plot the latent dimension\n",
    "        self.encoder_model = Model(self.input, self.encoder, name='encoder')\n",
    "\n",
    "    def build_decoder(self):\n",
    "        \"\"\"Build the decoder architecture.\"\"\"\n",
    "        if self.encoder is None:\n",
    "            raise RuntimeError(\"The encoder has to be built before you can build the decoder!\")\n",
    "\n",
    "        if self.pruned:\n",
    "            self.decoder = prune_low_magnitude(Dense(16,  activation='relu'), **self.pruning_params)(self.encoder)\n",
    "\n",
    "        if not self.pruned:\n",
    "            self.decoder = Dense(16, activation='relu')(self.encoder)\n",
    "        self.decoder = Dense(32,activation='relu')(self.decoder)     \n",
    "        \n",
    "        self.decoder = Dense(self.input_shape[0],activation='sigmoid', name='ecoder_output')(self.decoder)\n",
    "\n",
    "    def build_classifier(self):\n",
    "        \"\"\" Building the classifier architecture.\"\"\"\n",
    "        if self.encoder is None:\n",
    "            raise RuntimeError(\"The encoder has to be built before you can build the classifier!\")\n",
    "        \n",
    "        if self.pruned:\n",
    "            self.latent_classifier = prune_low_magnitude(Dense(16,  activation='relu'), **self.pruning_params)(self.encoder)\n",
    "        \n",
    "        if not self.pruned:\n",
    "            self.latent_classifier = Dense(16, activation='relu')(self.encoder)\n",
    "        \n",
    "        self.latent_classifier = Dense(32, activation='relu')(self.latent_classifier)\n",
    "        self.latent_classifier = Dense(self.num_classes, activation='softmax',name='classifier_output')(self.latent_classifier)\n",
    "\n",
    "    def build_autoencoder(self, use_latent_classifier=False):\n",
    "        \"\"\" Build the whole autoencoder Model from self.encoder and self.decoder.\"\"\"\n",
    "        self.dir()\n",
    "        \n",
    "        if self.encoder is None:\n",
    "            raise RuntimeError(\"The encoder has to be built before you can build the autoencoder!\")\n",
    "        if self.decoder is None:\n",
    "            raise RuntimeError(\"The decoder has to be built before you can build the autoencoder!\")\n",
    "        \n",
    "        if use_latent_classifier:\n",
    "            if self.latent_classifier is None:\n",
    "                raise RuntimeError(\"If you want to use the option with the latent classifier, you have to build it \"\n",
    "                                   \"beforehand!\")\n",
    "            self.autoencoder = Model(self.input, outputs=[self.decoder, self.latent_classifier])\n",
    "            self.autoencoder.compile(loss=['mse', 'categorical_crossentropy'], loss_weights=[1, 0.1], optimizer='adam',\n",
    "                                     metrics=\"accuracy\")\n",
    "            if not self.pruned:\n",
    "                self.autoencoder.save('model/AE_model/KERAS_check_model_w_classifier.h5')\n",
    "            if self.pruned:\n",
    "                self.autoencoder.save('model/AE_model/KERAS_check_pruned_model_w_classifier.h5')\n",
    "    \n",
    "        else:\n",
    "            self.autoencoder = Model(self.input, outputs=self.decoder, name = 'autoencoder')\n",
    "            self.autoencoder.compile(loss='mse', optimizer='adam') \n",
    "            if not self.pruned:\n",
    "                self.autoencoder.save('model/AE_model/KERAS_check_model_wo_classifier.h5')\n",
    "            if self.pruned:\n",
    "                self.autoencoder.save('model/AE_model/KERAS_check_pruned_model_wo_classifier.h5')\n",
    "        self.autoencoder.summary()\n",
    "\n",
    "    def fit_data(self, batch_size=256, epochs=20, use_latent_classifier=False):\n",
    "        \"\"\"Write the fit function for the autoencoder. \n",
    "        Storing the fit history in self.history to be able to plot the fitting scores.\"\"\"\n",
    "        \n",
    "        callbacks = all_callbacks(stop_patience = 1000,\n",
    "                              lr_factor = 0.5,\n",
    "                              lr_patience = 10,\n",
    "                              lr_epsilon = 0.000001,\n",
    "                              lr_cooldown = 2,\n",
    "                              lr_minimum = 0.0000001,\n",
    "                              outputDir = 'model/AE_model/model_2')\n",
    "        callbacks.callbacks.append(pruning_callbacks.UpdatePruningStep()) \n",
    "        \n",
    "        if use_latent_classifier:\n",
    "            self.history = self.autoencoder.fit(self.x_train, [self.x_train, self.y_train],\n",
    "                                                validation_data=(self.x_test, [self.x_test, self.y_test]),\n",
    "                                                batch_size=batch_size, epochs=epochs,\n",
    "                                                shuffle=True, callbacks=callbacks.callbacks)\n",
    "            self.autoencoder = strip_pruning(self.autoencoder)\n",
    "            if not self.pruned:\n",
    "                self.autoencoder.save('model/AE_model/KERAS_check_best_model_classifier.h5')\n",
    "            if self.pruned:\n",
    "                self.autoencoder.save('model/AE_model/KERAS_check_pruned_best_model_classifier.h5')\n",
    "        else:                         \n",
    "            self.history = self.autoencoder.fit(self.x_train, self.x_train,\n",
    "                                                validation_data=(self.x_test, self.x_test),\n",
    "                                                batch_size=batch_size, epochs=epochs,\n",
    "                                                shuffle=True, callbacks=callbacks.callbacks)  \n",
    "            self.autoencoder = strip_pruning(self.autoencoder)\n",
    "            if not self.pruned:\n",
    "                self.autoencoder.save('model/AE_model/KERAS_check_best_model.h5')\n",
    "            if self.pruned:\n",
    "                self.autoencoder.save('model/AE_model/KERAS_check_pruned_best_model.h5')\n",
    "\n",
    "        self.history = self.history.history\n",
    "\n",
    "    def dir(self):\n",
    "        \"\"\"Creation of the folders path to store the results.\"\"\"\n",
    "        dir = os.path.join(\"images\")\n",
    "        if not os.path.exists(dir):\n",
    "            os.mkdir(dir)\n",
    "        dir2 = os.path.join(\"./images/AE\")\n",
    "        if not os.path.exists(dir2):\n",
    "            os.mkdir(dir2)\n",
    "        dir3 = os.path.join(\"model\")\n",
    "        if not os.path.exists(dir3):\n",
    "            os.mkdir(dir3)\n",
    "        dir4 = os.path.join(\"./model/AE_model\")\n",
    "        if not os.path.exists(dir4):\n",
    "            os.mkdir(dir4)\n",
    "        \n",
    "    \n",
    "    def plot_score(self, use_latent_classifier = False, model_name=None):\n",
    "        \"\"\"Plots the scores achieved during the fitting.\"\"\"\n",
    "        if use_latent_classifier is False:\n",
    "            plt.plot(self.history['loss'])\n",
    "            plt.plot(self.history['val_loss'])\n",
    "            plt.ylabel('Model Accuracy')\n",
    "        else:\n",
    "            plt.plot(self.history['classifier_output_accuracy'])\n",
    "            plt.plot(self.history['val_classifier_output_accuracy'])\n",
    "            plt.ylabel('Classifier Accuracy')\n",
    "        \n",
    "        plt.xlabel('Epoch')        \n",
    "        plt.legend(['Train', 'Test'], loc='best')\n",
    "        plt.title('Accuracy of {model_name}'.format(model_name=model_name))\n",
    "        \n",
    "        plt.savefig('./images/AE/Accuracy of {model_name}.png'.format(model_name=model_name))\n",
    "        \n",
    "        plt.show()\n",
    "\n",
    "    def plot_latent_dimension(self, model_name = None):\n",
    "        \"\"\"Plots the latent dimension of the autoencoder.\"\"\"\n",
    "        prediction = self.encoder_model.predict(self.x_test)\n",
    "        fig = plt.figure(figsize=(10, 10))\n",
    "        fig.patch.set_facecolor(\"white\")\n",
    "        plt.scatter(prediction[:, 0], prediction[:, 1], c=np.argmax(self.y_test, axis=1), cmap=\"Set3\")\n",
    "        plt.colorbar()\n",
    "        plt.title('Latent space {model_name}'.format(model_name=model_name))\n",
    "\n",
    "\n",
    "        plt.savefig('./images/AE/Latent dimension of {model_name}.png'.format(model_name=model_name))\n",
    "\n",
    "        plt.show()\n",
    "\n",
    "    def plot_reco(self, use_latent_classifier = False):\n",
    "        \"\"\"Plotting reconstructed images\"\"\"\n",
    "        \n",
    "        n = 6\n",
    "        plt.figure(figsize=(10, 3))\n",
    "        reco_imgs = self.autoencoder.predict(self.x_test)\n",
    "        img_size = int(np.sqrt(self.input_shape[0]))\n",
    "\n",
    "        \n",
    "        if use_latent_classifier :\n",
    "            raise RuntimeError(\"You are considering the autoencoder with classifier!\")\n",
    "        else:\n",
    "        \n",
    "            for i in range(n):\n",
    "                # Display original MNIST\n",
    "                ax = plt.subplot(2, n, i + 1)\n",
    "                plt.imshow(self.x_test[i].reshape(img_size,img_size), cmap='gray_r')\n",
    "                ax.get_xaxis().set_visible(False)\n",
    "                ax.get_yaxis().set_visible(False)\n",
    "\n",
    "                # Display reconstructed MNIST            \n",
    "                ax = plt.subplot(2, n, i + n + 1)\n",
    "                plt.imshow(reco_imgs[i].reshape(img_size,img_size), cmap='gray_r')\n",
    "                ax.get_xaxis().set_visible(False)\n",
    "                ax.get_yaxis().set_visible(False)\n",
    "            plt.savefig('./images/AE/reconstructed images.png')\n",
    "            plt.show()\n",
    " \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is built the standard autoencoder, *without classifier* here, and the MNIST data class called before is used. \n",
    "The model is fitted and adequate plots are reproduced, \n",
    "as well as the comparison between orginal MNIST dataset and reconstructed images."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"autoencoder\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "encoder_input (InputLayer)   [(None, 64)]              0         \n",
      "_________________________________________________________________\n",
      "dense_20 (Dense)             (None, 32)                2080      \n",
      "_________________________________________________________________\n",
      "dense_21 (Dense)             (None, 16)                528       \n",
      "_________________________________________________________________\n",
      "encoder_output (Dense)       (None, 2)                 34        \n",
      "_________________________________________________________________\n",
      "dense_22 (Dense)             (None, 16)                48        \n",
      "_________________________________________________________________\n",
      "dense_23 (Dense)             (None, 32)                544       \n",
      "_________________________________________________________________\n",
      "ecoder_output (Dense)        (None, 64)                2112      \n",
      "=================================================================\n",
      "Total params: 5,346\n",
      "Trainable params: 5,346\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "WARNING:tensorflow:`epsilon` argument is deprecated and will be removed, use `min_delta` instead.\n",
      "WARNING:tensorflow:`period` argument is deprecated. Please use `save_freq` to specify the frequency in number of batches seen.\n",
      "Epoch 1/20\n",
      "  2/235 [..............................] - ETA: 1:35 - loss: 0.2091WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0000s vs `on_train_batch_end` time: 0.8198s). Check your callbacks.\n",
      "201/235 [========================>.....] - ETA: 0s - loss: 0.1231\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.09575, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.09575, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00001: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00001: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 6ms/step - loss: 0.1193 - val_loss: 0.0958\n",
      "Epoch 2/20\n",
      "218/235 [==========================>...] - ETA: 0s - loss: 0.0935\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.09575 to 0.09166, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.09575 to 0.09166, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00002: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00002: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0933 - val_loss: 0.0917\n",
      "Epoch 3/20\n",
      "202/235 [========================>.....] - ETA: 0s - loss: 0.0896\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.09166 to 0.08306, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.09166 to 0.08306, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00003: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00003: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0888 - val_loss: 0.0831\n",
      "Epoch 4/20\n",
      "198/235 [========================>.....] - ETA: 0s - loss: 0.0810\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00004: val_loss improved from 0.08306 to 0.07908, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00004: val_loss improved from 0.08306 to 0.07908, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00004: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00004: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0807 - val_loss: 0.0791\n",
      "Epoch 5/20\n",
      "196/235 [========================>.....] - ETA: 0s - loss: 0.0787\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00005: val_loss improved from 0.07908 to 0.07754, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00005: val_loss improved from 0.07908 to 0.07754, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00005: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00005: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0786 - val_loss: 0.0775\n",
      "Epoch 6/20\n",
      "200/235 [========================>.....] - ETA: 0s - loss: 0.0766\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00006: val_loss improved from 0.07754 to 0.07482, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00006: val_loss improved from 0.07754 to 0.07482, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00006: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00006: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0766 - val_loss: 0.0748\n",
      "Epoch 7/20\n",
      "213/235 [==========================>...] - ETA: 0s - loss: 0.0745\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.07482 to 0.07332, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.07482 to 0.07332, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00007: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00007: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0745 - val_loss: 0.0733\n",
      "Epoch 8/20\n",
      "214/235 [==========================>...] - ETA: 0s - loss: 0.0733\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00008: val_loss improved from 0.07332 to 0.07230, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00008: val_loss improved from 0.07332 to 0.07230, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00008: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00008: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0732 - val_loss: 0.0723\n",
      "Epoch 9/20\n",
      "206/235 [=========================>....] - ETA: 0s - loss: 0.0724\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00009: val_loss improved from 0.07230 to 0.07165, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00009: val_loss improved from 0.07230 to 0.07165, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00009: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00009: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0723 - val_loss: 0.0717\n",
      "Epoch 10/20\n",
      "224/235 [===========================>..] - ETA: 0s - loss: 0.0716\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00010: val_loss improved from 0.07165 to 0.07105, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00010: val_loss improved from 0.07165 to 0.07105, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_epoch10.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0716 - val_loss: 0.0711\n",
      "Epoch 11/20\n",
      "207/235 [=========================>....] - ETA: 0s - loss: 0.0709\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00011: val_loss improved from 0.07105 to 0.07033, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00011: val_loss improved from 0.07105 to 0.07033, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00011: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00011: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0708 - val_loss: 0.0703\n",
      "Epoch 12/20\n",
      "231/235 [============================>.] - ETA: 0s - loss: 0.0702\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.07033 to 0.06992, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.07033 to 0.06992, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00012: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00012: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0702 - val_loss: 0.0699\n",
      "Epoch 13/20\n",
      "229/235 [============================>.] - ETA: 0s - loss: 0.0696\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00013: val_loss improved from 0.06992 to 0.06934, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00013: val_loss improved from 0.06992 to 0.06934, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00013: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00013: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0695 - val_loss: 0.0693\n",
      "Epoch 14/20\n",
      "222/235 [===========================>..] - ETA: 0s - loss: 0.0690\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.06934 to 0.06889, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.06934 to 0.06889, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00014: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00014: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0690 - val_loss: 0.0689\n",
      "Epoch 15/20\n",
      "209/235 [=========================>....] - ETA: 0s - loss: 0.0685\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00015: val_loss improved from 0.06889 to 0.06834, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00015: val_loss improved from 0.06889 to 0.06834, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00015: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00015: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0684 - val_loss: 0.0683\n",
      "Epoch 16/20\n",
      "234/235 [============================>.] - ETA: 0s - loss: 0.0680\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00016: val_loss improved from 0.06834 to 0.06806, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00016: val_loss improved from 0.06834 to 0.06806, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00016: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00016: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 0s 2ms/step - loss: 0.0680 - val_loss: 0.0681\n",
      "Epoch 17/20\n",
      "229/235 [============================>.] - ETA: 0s - loss: 0.0676\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00017: val_loss improved from 0.06806 to 0.06779, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00017: val_loss improved from 0.06806 to 0.06779, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00017: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00017: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0676 - val_loss: 0.0678\n",
      "Epoch 18/20\n",
      "235/235 [==============================] - ETA: 0s - loss: 0.0673\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.06779 to 0.06754, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.06779 to 0.06754, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00018: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00018: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0673 - val_loss: 0.0675\n",
      "Epoch 19/20\n",
      "223/235 [===========================>..] - ETA: 0s - loss: 0.0671\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00019: val_loss improved from 0.06754 to 0.06729, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00019: val_loss improved from 0.06754 to 0.06729, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00019: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00019: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0671 - val_loss: 0.0673\n",
      "Epoch 20/20\n",
      "218/235 [==========================>...] - ETA: 0s - loss: 0.0668\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00020: val_loss improved from 0.06729 to 0.06698, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00020: val_loss improved from 0.06729 to 0.06698, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_epoch20.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0668 - val_loss: 0.0670\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x216 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "network = Autoencoder(data_zoom, pruned=False)\n",
    "network.build_encoder()\n",
    "network.build_decoder()\n",
    "network.build_autoencoder()\n",
    "network.fit_data(epochs=20)\n",
    "network.plot_latent_dimension(\"Autoencoder without classifier\")\n",
    "network.plot_reco()\n",
    "network.plot_score(use_latent_classifier=False,model_name=\"Autoencoder without classifier\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"autoencoder\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "encoder_input (InputLayer)   [(None, 64)]              0         \n",
      "_________________________________________________________________\n",
      "dense_24 (Dense)             (None, 32)                2080      \n",
      "_________________________________________________________________\n",
      "prune_low_magnitude_dense_25 (None, 16)                1042      \n",
      "_________________________________________________________________\n",
      "encoder_output (Dense)       (None, 2)                 34        \n",
      "_________________________________________________________________\n",
      "prune_low_magnitude_dense_26 (None, 16)                82        \n",
      "_________________________________________________________________\n",
      "dense_27 (Dense)             (None, 32)                544       \n",
      "_________________________________________________________________\n",
      "ecoder_output (Dense)        (None, 64)                2112      \n",
      "=================================================================\n",
      "Total params: 5,894\n",
      "Trainable params: 5,346\n",
      "Non-trainable params: 548\n",
      "_________________________________________________________________\n",
      "WARNING:tensorflow:`epsilon` argument is deprecated and will be removed, use `min_delta` instead.\n",
      "WARNING:tensorflow:`period` argument is deprecated. Please use `save_freq` to specify the frequency in number of batches seen.\n",
      "Epoch 1/20\n",
      "  2/235 [..............................] - ETA: 1:42 - loss: 0.2093WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0000s vs `on_train_batch_end` time: 0.8741s). Check your callbacks.\n",
      "224/235 [===========================>..] - ETA: 0s - loss: 0.1238\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.09539, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.09539, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00001: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00001: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 2s 7ms/step - loss: 0.1226 - val_loss: 0.0954\n",
      "Epoch 2/20\n",
      "213/235 [==========================>...] - ETA: 0s - loss: 0.0932\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.09539 to 0.09166, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.09539 to 0.09166, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00002: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00002: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0931 - val_loss: 0.0917\n",
      "Epoch 3/20\n",
      "208/235 [=========================>....] - ETA: 0s - loss: 0.0906\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.09166 to 0.08973, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.09166 to 0.08973, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00003: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00003: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 0s 2ms/step - loss: 0.0905 - val_loss: 0.0897\n",
      "Epoch 4/20\n",
      "215/235 [==========================>...] - ETA: 0s - loss: 0.0892\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00004: val_loss improved from 0.08973 to 0.08865, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00004: val_loss improved from 0.08973 to 0.08865, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00004: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00004: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 0s 2ms/step - loss: 0.0892 - val_loss: 0.0886\n",
      "Epoch 5/20\n",
      "201/235 [========================>.....] - ETA: 0s - loss: 0.0884\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00005: val_loss improved from 0.08865 to 0.08802, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00005: val_loss improved from 0.08865 to 0.08802, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00005: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00005: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0884 - val_loss: 0.0880\n",
      "Epoch 6/20\n",
      "231/235 [============================>.] - ETA: 0s - loss: 0.0879\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00006: val_loss improved from 0.08802 to 0.08766, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00006: val_loss improved from 0.08802 to 0.08766, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00006: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00006: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0879 - val_loss: 0.0877\n",
      "Epoch 7/20\n",
      "219/235 [==========================>...] - ETA: 0s - loss: 0.0876\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.08766 to 0.08735, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.08766 to 0.08735, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00007: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00007: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0876 - val_loss: 0.0874\n",
      "Epoch 8/20\n",
      "232/235 [============================>.] - ETA: 0s - loss: 0.0874\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00008: val_loss improved from 0.08735 to 0.08720, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00008: val_loss improved from 0.08735 to 0.08720, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00008: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00008: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0874 - val_loss: 0.0872\n",
      "Epoch 9/20\n",
      "211/235 [=========================>....] - ETA: 0s - loss: 0.0878\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00009: val_loss did not improve from 0.08720\n",
      "\n",
      "Epoch 00009: val_loss did not improve from 0.08720\n",
      "\n",
      "Epoch 00009: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00009: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0878 - val_loss: 0.0872\n",
      "Epoch 10/20\n",
      "231/235 [============================>.] - ETA: 0s - loss: 0.0866\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00010: val_loss improved from 0.08720 to 0.08568, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00010: val_loss improved from 0.08720 to 0.08568, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_epoch10.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0866 - val_loss: 0.0857\n",
      "Epoch 11/20\n",
      "225/235 [===========================>..] - ETA: 0s - loss: 0.0854\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00011: val_loss improved from 0.08568 to 0.08487, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00011: val_loss improved from 0.08568 to 0.08487, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00011: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00011: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0854 - val_loss: 0.0849\n",
      "Epoch 12/20\n",
      "211/235 [=========================>....] - ETA: 0s - loss: 0.0848\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.08487 to 0.08463, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.08487 to 0.08463, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00012: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00012: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0848 - val_loss: 0.0846\n",
      "Epoch 13/20\n",
      "224/235 [===========================>..] - ETA: 0s - loss: 0.0846\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00013: val_loss improved from 0.08463 to 0.08444, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00013: val_loss improved from 0.08463 to 0.08444, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00013: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00013: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0846 - val_loss: 0.0844\n",
      "Epoch 14/20\n",
      "227/235 [===========================>..] - ETA: 0s - loss: 0.0844\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.08444 to 0.08439, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.08444 to 0.08439, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00014: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00014: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0844 - val_loss: 0.0844\n",
      "Epoch 15/20\n",
      "217/235 [==========================>...] - ETA: 0s - loss: 0.0842\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00015: val_loss improved from 0.08439 to 0.08396, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00015: val_loss improved from 0.08439 to 0.08396, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00015: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00015: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0841 - val_loss: 0.0840\n",
      "Epoch 16/20\n",
      "223/235 [===========================>..] - ETA: 0s - loss: 0.0838\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00016: val_loss improved from 0.08396 to 0.08380, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00016: val_loss improved from 0.08396 to 0.08380, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00016: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00016: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0838 - val_loss: 0.0838\n",
      "Epoch 17/20\n",
      "221/235 [===========================>..] - ETA: 0s - loss: 0.0836\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00017: val_loss improved from 0.08380 to 0.08351, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00017: val_loss improved from 0.08380 to 0.08351, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00017: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00017: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0836 - val_loss: 0.0835\n",
      "Epoch 18/20\n",
      "232/235 [============================>.] - ETA: 0s - loss: 0.0834\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.08351 to 0.08311, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.08351 to 0.08311, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00018: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00018: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0834 - val_loss: 0.0831\n",
      "Epoch 19/20\n",
      "221/235 [===========================>..] - ETA: 0s - loss: 0.0830\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00019: val_loss improved from 0.08311 to 0.08273, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00019: val_loss improved from 0.08311 to 0.08273, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00019: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00019: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.0830 - val_loss: 0.0827\n",
      "Epoch 20/20\n",
      "234/235 [============================>.] - ETA: 0s - loss: 0.0826\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00020: val_loss improved from 0.08273 to 0.08218, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00020: val_loss improved from 0.08273 to 0.08218, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_epoch20.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.0826 - val_loss: 0.0822\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x216 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "network = Autoencoder(data_zoom, pruned=True)\n",
    "network.build_encoder()\n",
    "network.build_decoder()\n",
    "network.build_autoencoder()\n",
    "network.fit_data(epochs=20)\n",
    "network.plot_latent_dimension(\"Compressed AE without classifier\")\n",
    "network.plot_reco()\n",
    "network.plot_score(use_latent_classifier=False,model_name=\"Compressed AE without classifier\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Repeating the same procedure for the autoencoder with classifier."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"functional_5\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "encoder_input (InputLayer)      [(None, 64)]         0                                            \n",
      "__________________________________________________________________________________________________\n",
      "dense_28 (Dense)                (None, 32)           2080        encoder_input[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "dense_29 (Dense)                (None, 16)           528         dense_28[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "encoder_output (Dense)          (None, 2)            34          dense_29[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "dense_30 (Dense)                (None, 16)           48          encoder_output[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "dense_32 (Dense)                (None, 16)           48          encoder_output[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "dense_31 (Dense)                (None, 32)           544         dense_30[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "dense_33 (Dense)                (None, 32)           544         dense_32[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "ecoder_output (Dense)           (None, 64)           2112        dense_31[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "classifier_output (Dense)       (None, 10)           330         dense_33[0][0]                   \n",
      "==================================================================================================\n",
      "Total params: 6,268\n",
      "Trainable params: 6,268\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n",
      "WARNING:tensorflow:`epsilon` argument is deprecated and will be removed, use `min_delta` instead.\n",
      "WARNING:tensorflow:`period` argument is deprecated. Please use `save_freq` to specify the frequency in number of batches seen.\n",
      "Epoch 1/20\n",
      "  2/235 [..............................] - ETA: 1:58 - loss: 0.4389 - ecoder_output_loss: 0.2085 - classifier_output_loss: 2.3037 - ecoder_output_accuracy: 0.0293 - classifier_output_accuracy: 0.0391WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0000s vs `on_train_batch_end` time: 1.0262s). Check your callbacks.\n",
      "197/235 [========================>.....] - ETA: 0s - loss: 0.3149 - ecoder_output_loss: 0.1183 - classifier_output_loss: 1.9660 - ecoder_output_accuracy: 0.0321 - classifier_output_accuracy: 0.2397\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.23869, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.23869, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00001: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00001: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 2s 7ms/step - loss: 0.3037 - ecoder_output_loss: 0.1140 - classifier_output_loss: 1.8966 - ecoder_output_accuracy: 0.0297 - classifier_output_accuracy: 0.2686 - val_loss: 0.2387 - val_ecoder_output_loss: 0.0909 - val_classifier_output_loss: 1.4782 - val_ecoder_output_accuracy: 0.0176 - val_classifier_output_accuracy: 0.4513\n",
      "Epoch 2/20\n",
      "217/235 [==========================>...] - ETA: 0s - loss: 0.2211 - ecoder_output_loss: 0.0885 - classifier_output_loss: 1.3255 - ecoder_output_accuracy: 0.0198 - classifier_output_accuracy: 0.5137\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.23869 to 0.20403, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.23869 to 0.20403, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00002: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00002: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.2199 - ecoder_output_loss: 0.0884 - classifier_output_loss: 1.3154 - ecoder_output_accuracy: 0.0198 - classifier_output_accuracy: 0.5178 - val_loss: 0.2040 - val_ecoder_output_loss: 0.0858 - val_classifier_output_loss: 1.1819 - val_ecoder_output_accuracy: 0.0185 - val_classifier_output_accuracy: 0.5741\n",
      "Epoch 3/20\n",
      "211/235 [=========================>....] - ETA: 0s - loss: 0.1932 - ecoder_output_loss: 0.0844 - classifier_output_loss: 1.0880 - ecoder_output_accuracy: 0.0267 - classifier_output_accuracy: 0.6150\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.20403 to 0.18255, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.20403 to 0.18255, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00003: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00003: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1921 - ecoder_output_loss: 0.0842 - classifier_output_loss: 1.0795 - ecoder_output_accuracy: 0.0280 - classifier_output_accuracy: 0.6199 - val_loss: 0.1825 - val_ecoder_output_loss: 0.0824 - val_classifier_output_loss: 1.0018 - val_ecoder_output_accuracy: 0.0422 - val_classifier_output_accuracy: 0.6623\n",
      "Epoch 4/20\n",
      "193/235 [=======================>......] - ETA: 0s - loss: 0.1766 - ecoder_output_loss: 0.0817 - classifier_output_loss: 0.9490 - ecoder_output_accuracy: 0.0449 - classifier_output_accuracy: 0.6794\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00004: val_loss improved from 0.18255 to 0.16927, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00004: val_loss improved from 0.18255 to 0.16927, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00004: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00004: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 0s 2ms/step - loss: 0.1758 - ecoder_output_loss: 0.0817 - classifier_output_loss: 0.9411 - ecoder_output_accuracy: 0.0441 - classifier_output_accuracy: 0.6836 - val_loss: 0.1693 - val_ecoder_output_loss: 0.0809 - val_classifier_output_loss: 0.8835 - val_ecoder_output_accuracy: 0.0365 - val_classifier_output_accuracy: 0.7216\n",
      "Epoch 5/20\n",
      "214/235 [==========================>...] - ETA: 0s - loss: 0.1649 - ecoder_output_loss: 0.0807 - classifier_output_loss: 0.8421 - ecoder_output_accuracy: 0.0417 - classifier_output_accuracy: 0.7283\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00005: val_loss improved from 0.16927 to 0.15873, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00005: val_loss improved from 0.16927 to 0.15873, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00005: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00005: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1644 - ecoder_output_loss: 0.0807 - classifier_output_loss: 0.8377 - ecoder_output_accuracy: 0.0415 - classifier_output_accuracy: 0.7309 - val_loss: 0.1587 - val_ecoder_output_loss: 0.0799 - val_classifier_output_loss: 0.7879 - val_ecoder_output_accuracy: 0.0297 - val_classifier_output_accuracy: 0.7628\n",
      "Epoch 6/20\n",
      "229/235 [============================>.] - ETA: 0s - loss: 0.1552 - ecoder_output_loss: 0.0798 - classifier_output_loss: 0.7532 - ecoder_output_accuracy: 0.0404 - classifier_output_accuracy: 0.7681\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00006: val_loss improved from 0.15873 to 0.15171, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00006: val_loss improved from 0.15873 to 0.15171, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00006: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00006: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 0s 2ms/step - loss: 0.1551 - ecoder_output_loss: 0.0798 - classifier_output_loss: 0.7526 - ecoder_output_accuracy: 0.0404 - classifier_output_accuracy: 0.7680 - val_loss: 0.1517 - val_ecoder_output_loss: 0.0792 - val_classifier_output_loss: 0.7253 - val_ecoder_output_accuracy: 0.0341 - val_classifier_output_accuracy: 0.7860\n",
      "Epoch 7/20\n",
      "204/235 [=========================>....] - ETA: 0s - loss: 0.1485 - ecoder_output_loss: 0.0791 - classifier_output_loss: 0.6933 - ecoder_output_accuracy: 0.0430 - classifier_output_accuracy: 0.7895\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.15171 to 0.14519, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.15171 to 0.14519, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00007: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00007: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.1481 - ecoder_output_loss: 0.0791 - classifier_output_loss: 0.6901 - ecoder_output_accuracy: 0.0430 - classifier_output_accuracy: 0.7911 - val_loss: 0.1452 - val_ecoder_output_loss: 0.0781 - val_classifier_output_loss: 0.6707 - val_ecoder_output_accuracy: 0.0361 - val_classifier_output_accuracy: 0.8077\n",
      "Epoch 8/20\n",
      "223/235 [===========================>..] - ETA: 0s - loss: 0.1427 - ecoder_output_loss: 0.0784 - classifier_output_loss: 0.6435 - ecoder_output_accuracy: 0.0444 - classifier_output_accuracy: 0.8082\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00008: val_loss improved from 0.14519 to 0.14041, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00008: val_loss improved from 0.14519 to 0.14041, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00008: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00008: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1427 - ecoder_output_loss: 0.0783 - classifier_output_loss: 0.6437 - ecoder_output_accuracy: 0.0448 - classifier_output_accuracy: 0.8079 - val_loss: 0.1404 - val_ecoder_output_loss: 0.0777 - val_classifier_output_loss: 0.6268 - val_ecoder_output_accuracy: 0.0510 - val_classifier_output_accuracy: 0.8208\n",
      "Epoch 9/20\n",
      "225/235 [===========================>..] - ETA: 0s - loss: 0.1381 - ecoder_output_loss: 0.0777 - classifier_output_loss: 0.6039 - ecoder_output_accuracy: 0.0490 - classifier_output_accuracy: 0.8223\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00009: val_loss improved from 0.14041 to 0.13625, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00009: val_loss improved from 0.14041 to 0.13625, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00009: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00009: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1381 - ecoder_output_loss: 0.0777 - classifier_output_loss: 0.6039 - ecoder_output_accuracy: 0.0489 - classifier_output_accuracy: 0.8226 - val_loss: 0.1363 - val_ecoder_output_loss: 0.0769 - val_classifier_output_loss: 0.5936 - val_ecoder_output_accuracy: 0.0560 - val_classifier_output_accuracy: 0.8333\n",
      "Epoch 10/20\n",
      "213/235 [==========================>...] - ETA: 0s - loss: 0.1344 - ecoder_output_loss: 0.0772 - classifier_output_loss: 0.5718 - ecoder_output_accuracy: 0.0536 - classifier_output_accuracy: 0.8328\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00010: val_loss improved from 0.13625 to 0.13318, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00010: val_loss improved from 0.13625 to 0.13318, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_epoch10.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1345 - ecoder_output_loss: 0.0771 - classifier_output_loss: 0.5733 - ecoder_output_accuracy: 0.0537 - classifier_output_accuracy: 0.8323 - val_loss: 0.1332 - val_ecoder_output_loss: 0.0765 - val_classifier_output_loss: 0.5672 - val_ecoder_output_accuracy: 0.0493 - val_classifier_output_accuracy: 0.8424\n",
      "Epoch 11/20\n",
      "218/235 [==========================>...] - ETA: 0s - loss: 0.1316 - ecoder_output_loss: 0.0768 - classifier_output_loss: 0.5489 - ecoder_output_accuracy: 0.0540 - classifier_output_accuracy: 0.8401\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00011: val_loss improved from 0.13318 to 0.13046, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00011: val_loss improved from 0.13318 to 0.13046, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00011: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00011: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1315 - ecoder_output_loss: 0.0767 - classifier_output_loss: 0.5479 - ecoder_output_accuracy: 0.0540 - classifier_output_accuracy: 0.8404 - val_loss: 0.1305 - val_ecoder_output_loss: 0.0760 - val_classifier_output_loss: 0.5449 - val_ecoder_output_accuracy: 0.0505 - val_classifier_output_accuracy: 0.8494\n",
      "Epoch 12/20\n",
      "217/235 [==========================>...] - ETA: 0s - loss: 0.1290 - ecoder_output_loss: 0.0764 - classifier_output_loss: 0.5257 - ecoder_output_accuracy: 0.0526 - classifier_output_accuracy: 0.8486\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.13046 to 0.12774, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.13046 to 0.12774, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00012: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00012: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1290 - ecoder_output_loss: 0.0764 - classifier_output_loss: 0.5258 - ecoder_output_accuracy: 0.0527 - classifier_output_accuracy: 0.8483 - val_loss: 0.1277 - val_ecoder_output_loss: 0.0758 - val_classifier_output_loss: 0.5198 - val_ecoder_output_accuracy: 0.0506 - val_classifier_output_accuracy: 0.8549\n",
      "Epoch 13/20\n",
      "214/235 [==========================>...] - ETA: 0s - loss: 0.1268 - ecoder_output_loss: 0.0760 - classifier_output_loss: 0.5073 - ecoder_output_accuracy: 0.0520 - classifier_output_accuracy: 0.8532\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00013: val_loss improved from 0.12774 to 0.12577, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00013: val_loss improved from 0.12774 to 0.12577, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00013: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00013: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1268 - ecoder_output_loss: 0.0761 - classifier_output_loss: 0.5066 - ecoder_output_accuracy: 0.0523 - classifier_output_accuracy: 0.8536 - val_loss: 0.1258 - val_ecoder_output_loss: 0.0755 - val_classifier_output_loss: 0.5023 - val_ecoder_output_accuracy: 0.0451 - val_classifier_output_accuracy: 0.8596\n",
      "Epoch 14/20\n",
      "215/235 [==========================>...] - ETA: 0s - loss: 0.1250 - ecoder_output_loss: 0.0759 - classifier_output_loss: 0.4903 - ecoder_output_accuracy: 0.0524 - classifier_output_accuracy: 0.8590\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.12577 to 0.12464, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.12577 to 0.12464, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00014: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00014: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 4ms/step - loss: 0.1250 - ecoder_output_loss: 0.0759 - classifier_output_loss: 0.4913 - ecoder_output_accuracy: 0.0523 - classifier_output_accuracy: 0.8587 - val_loss: 0.1246 - val_ecoder_output_loss: 0.0754 - val_classifier_output_loss: 0.4926 - val_ecoder_output_accuracy: 0.0460 - val_classifier_output_accuracy: 0.8612\n",
      "Epoch 15/20\n",
      "230/235 [============================>.] - ETA: 0s - loss: 0.1238 - ecoder_output_loss: 0.0758 - classifier_output_loss: 0.4800 - ecoder_output_accuracy: 0.0519 - classifier_output_accuracy: 0.8615\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00015: val_loss improved from 0.12464 to 0.12341, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00015: val_loss improved from 0.12464 to 0.12341, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00015: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00015: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 4ms/step - loss: 0.1237 - ecoder_output_loss: 0.0757 - classifier_output_loss: 0.4793 - ecoder_output_accuracy: 0.0518 - classifier_output_accuracy: 0.8617 - val_loss: 0.1234 - val_ecoder_output_loss: 0.0753 - val_classifier_output_loss: 0.4811 - val_ecoder_output_accuracy: 0.0480 - val_classifier_output_accuracy: 0.8618\n",
      "Epoch 16/20\n",
      "216/235 [==========================>...] - ETA: 0s - loss: 0.1223 - ecoder_output_loss: 0.0756 - classifier_output_loss: 0.4669 - ecoder_output_accuracy: 0.0518 - classifier_output_accuracy: 0.8658\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00016: val_loss improved from 0.12341 to 0.12283, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00016: val_loss improved from 0.12341 to 0.12283, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00016: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00016: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1223 - ecoder_output_loss: 0.0756 - classifier_output_loss: 0.4673 - ecoder_output_accuracy: 0.0518 - classifier_output_accuracy: 0.8652 - val_loss: 0.1228 - val_ecoder_output_loss: 0.0751 - val_classifier_output_loss: 0.4772 - val_ecoder_output_accuracy: 0.0512 - val_classifier_output_accuracy: 0.8664\n",
      "Epoch 17/20\n",
      "228/235 [============================>.] - ETA: 0s - loss: 0.1213 - ecoder_output_loss: 0.0755 - classifier_output_loss: 0.4581 - ecoder_output_accuracy: 0.0509 - classifier_output_accuracy: 0.8677\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00017: val_loss improved from 0.12283 to 0.12105, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00017: val_loss improved from 0.12283 to 0.12105, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00017: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00017: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 4ms/step - loss: 0.1212 - ecoder_output_loss: 0.0754 - classifier_output_loss: 0.4576 - ecoder_output_accuracy: 0.0513 - classifier_output_accuracy: 0.8679 - val_loss: 0.1210 - val_ecoder_output_loss: 0.0749 - val_classifier_output_loss: 0.4613 - val_ecoder_output_accuracy: 0.0499 - val_classifier_output_accuracy: 0.8691\n",
      "Epoch 18/20\n",
      "227/235 [===========================>..] - ETA: 0s - loss: 0.1202 - ecoder_output_loss: 0.0753 - classifier_output_loss: 0.4486 - ecoder_output_accuracy: 0.0527 - classifier_output_accuracy: 0.8709\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.12105 to 0.11997, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.12105 to 0.11997, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00018: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00018: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1202 - ecoder_output_loss: 0.0753 - classifier_output_loss: 0.4487 - ecoder_output_accuracy: 0.0524 - classifier_output_accuracy: 0.8709 - val_loss: 0.1200 - val_ecoder_output_loss: 0.0747 - val_classifier_output_loss: 0.4524 - val_ecoder_output_accuracy: 0.0468 - val_classifier_output_accuracy: 0.8723\n",
      "Epoch 19/20\n",
      "226/235 [===========================>..] - ETA: 0s - loss: 0.1192 - ecoder_output_loss: 0.0752 - classifier_output_loss: 0.4400 - ecoder_output_accuracy: 0.0520 - classifier_output_accuracy: 0.8737\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00019: val_loss improved from 0.11997 to 0.11944, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00019: val_loss improved from 0.11997 to 0.11944, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00019: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00019: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1192 - ecoder_output_loss: 0.0752 - classifier_output_loss: 0.4401 - ecoder_output_accuracy: 0.0517 - classifier_output_accuracy: 0.8737 - val_loss: 0.1194 - val_ecoder_output_loss: 0.0747 - val_classifier_output_loss: 0.4470 - val_ecoder_output_accuracy: 0.0460 - val_classifier_output_accuracy: 0.8734\n",
      "Epoch 20/20\n",
      "234/235 [============================>.] - ETA: 0s - loss: 0.1182 - ecoder_output_loss: 0.0751 - classifier_output_loss: 0.4315 - ecoder_output_accuracy: 0.0516 - classifier_output_accuracy: 0.8761\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00020: val_loss improved from 0.11944 to 0.11923, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00020: val_loss improved from 0.11944 to 0.11923, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_epoch20.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 4ms/step - loss: 0.1182 - ecoder_output_loss: 0.0751 - classifier_output_loss: 0.4316 - ecoder_output_accuracy: 0.0516 - classifier_output_accuracy: 0.8761 - val_loss: 0.1192 - val_ecoder_output_loss: 0.0746 - val_classifier_output_loss: 0.4461 - val_ecoder_output_accuracy: 0.0459 - val_classifier_output_accuracy: 0.8733\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "network_with_classifier = Autoencoder(data_zoom, pruned=False)\n",
    "network_with_classifier.build_encoder()\n",
    "network_with_classifier.build_decoder()\n",
    "network_with_classifier.build_classifier()\n",
    "network_with_classifier.build_autoencoder(use_latent_classifier=True)\n",
    "network_with_classifier.fit_data( batch_size=256,epochs=20, use_latent_classifier=True)\n",
    "network_with_classifier.plot_latent_dimension(\"Autoencoder with classifier\")\n",
    "network_with_classifier.plot_score(use_latent_classifier=True,model_name=\"Autoencoder with classifier\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"functional_9\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "encoder_input (InputLayer)      [(None, 64)]         0                                            \n",
      "__________________________________________________________________________________________________\n",
      "dense_40 (Dense)                (None, 32)           2080        encoder_input[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "prune_low_magnitude_dense_41 (P (None, 16)           1042        dense_40[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "encoder_output (Dense)          (None, 2)            34          prune_low_magnitude_dense_41[0][0\n",
      "__________________________________________________________________________________________________\n",
      "prune_low_magnitude_dense_42 (P (None, 16)           82          encoder_output[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "prune_low_magnitude_dense_44 (P (None, 16)           82          encoder_output[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "dense_43 (Dense)                (None, 32)           544         prune_low_magnitude_dense_42[0][0\n",
      "__________________________________________________________________________________________________\n",
      "dense_45 (Dense)                (None, 32)           544         prune_low_magnitude_dense_44[0][0\n",
      "__________________________________________________________________________________________________\n",
      "ecoder_output (Dense)           (None, 64)           2112        dense_43[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "classifier_output (Dense)       (None, 10)           330         dense_45[0][0]                   \n",
      "==================================================================================================\n",
      "Total params: 6,850\n",
      "Trainable params: 6,268\n",
      "Non-trainable params: 582\n",
      "__________________________________________________________________________________________________\n",
      "WARNING:tensorflow:`epsilon` argument is deprecated and will be removed, use `min_delta` instead.\n",
      "WARNING:tensorflow:`period` argument is deprecated. Please use `save_freq` to specify the frequency in number of batches seen.\n",
      "Epoch 1/20\n",
      "  2/235 [..............................] - ETA: 2:29 - loss: 0.4390 - ecoder_output_loss: 0.2087 - classifier_output_loss: 2.3032 - ecoder_output_accuracy: 0.0371 - classifier_output_accuracy: 0.0996WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0000s vs `on_train_batch_end` time: 1.2791s). Check your callbacks.\n",
      "223/235 [===========================>..] - ETA: 0s - loss: 0.3184 - ecoder_output_loss: 0.1216 - classifier_output_loss: 1.9688 - ecoder_output_accuracy: 0.0485 - classifier_output_accuracy: 0.2639\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.24613, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00001: val_loss improved from inf to 0.24613, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00001: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00001: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 2s 10ms/step - loss: 0.3151 - ecoder_output_loss: 0.1201 - classifier_output_loss: 1.9499 - ecoder_output_accuracy: 0.0488 - classifier_output_accuracy: 0.2700 - val_loss: 0.2461 - val_ecoder_output_loss: 0.0926 - val_classifier_output_loss: 1.5357 - val_ecoder_output_accuracy: 0.0318 - val_classifier_output_accuracy: 0.4146\n",
      "Epoch 2/20\n",
      "219/235 [==========================>...] - ETA: 0s - loss: 0.2283 - ecoder_output_loss: 0.0905 - classifier_output_loss: 1.3789 - ecoder_output_accuracy: 0.0320 - classifier_output_accuracy: 0.4735\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.24613 to 0.20561, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00002: val_loss improved from 0.24613 to 0.20561, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00002: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00002: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.2270 - ecoder_output_loss: 0.0903 - classifier_output_loss: 1.3674 - ecoder_output_accuracy: 0.0316 - classifier_output_accuracy: 0.4785 - val_loss: 0.2056 - val_ecoder_output_loss: 0.0879 - val_classifier_output_loss: 1.1775 - val_ecoder_output_accuracy: 0.0257 - val_classifier_output_accuracy: 0.5385\n",
      "Epoch 3/20\n",
      "213/235 [==========================>...] - ETA: 0s - loss: 0.1936 - ecoder_output_loss: 0.0854 - classifier_output_loss: 1.0826 - ecoder_output_accuracy: 0.0296 - classifier_output_accuracy: 0.6301\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.20561 to 0.17459, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00003: val_loss improved from 0.20561 to 0.17459, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00003: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00003: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1925 - ecoder_output_loss: 0.0852 - classifier_output_loss: 1.0733 - ecoder_output_accuracy: 0.0310 - classifier_output_accuracy: 0.6348 - val_loss: 0.1746 - val_ecoder_output_loss: 0.0819 - val_classifier_output_loss: 0.9273 - val_ecoder_output_accuracy: 0.0349 - val_classifier_output_accuracy: 0.7102\n",
      "Epoch 4/20\n",
      "226/235 [===========================>..] - ETA: 0s - loss: 0.1683 - ecoder_output_loss: 0.0814 - classifier_output_loss: 0.8698 - ecoder_output_accuracy: 0.0504 - classifier_output_accuracy: 0.7262\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00004: val_loss improved from 0.17459 to 0.15750, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00004: val_loss improved from 0.17459 to 0.15750, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00004: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00004: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 4ms/step - loss: 0.1679 - ecoder_output_loss: 0.0813 - classifier_output_loss: 0.8661 - ecoder_output_accuracy: 0.0505 - classifier_output_accuracy: 0.7274 - val_loss: 0.1575 - val_ecoder_output_loss: 0.0805 - val_classifier_output_loss: 0.7699 - val_ecoder_output_accuracy: 0.0482 - val_classifier_output_accuracy: 0.7616\n",
      "Epoch 5/20\n",
      "231/235 [============================>.] - ETA: 0s - loss: 0.1546 - ecoder_output_loss: 0.0807 - classifier_output_loss: 0.7383 - ecoder_output_accuracy: 0.0544 - classifier_output_accuracy: 0.7749\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00005: val_loss improved from 0.15750 to 0.14805, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00005: val_loss improved from 0.15750 to 0.14805, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00005: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00005: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 4ms/step - loss: 0.1545 - ecoder_output_loss: 0.0807 - classifier_output_loss: 0.7380 - ecoder_output_accuracy: 0.0545 - classifier_output_accuracy: 0.7749 - val_loss: 0.1480 - val_ecoder_output_loss: 0.0802 - val_classifier_output_loss: 0.6784 - val_ecoder_output_accuracy: 0.0502 - val_classifier_output_accuracy: 0.7996\n",
      "Epoch 6/20\n",
      "213/235 [==========================>...] - ETA: 0s - loss: 0.1467 - ecoder_output_loss: 0.0804 - classifier_output_loss: 0.6629 - ecoder_output_accuracy: 0.0555 - classifier_output_accuracy: 0.8010\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00006: val_loss improved from 0.14805 to 0.14167, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00006: val_loss improved from 0.14805 to 0.14167, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00006: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00006: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 4ms/step - loss: 0.1464 - ecoder_output_loss: 0.0804 - classifier_output_loss: 0.6599 - ecoder_output_accuracy: 0.0560 - classifier_output_accuracy: 0.8016 - val_loss: 0.1417 - val_ecoder_output_loss: 0.0798 - val_classifier_output_loss: 0.6188 - val_ecoder_output_accuracy: 0.0541 - val_classifier_output_accuracy: 0.8126\n",
      "Epoch 7/20\n",
      "216/235 [==========================>...] - ETA: 0s - loss: 0.1405 - ecoder_output_loss: 0.0799 - classifier_output_loss: 0.6053 - ecoder_output_accuracy: 0.0586 - classifier_output_accuracy: 0.8203\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.14167 to 0.13600, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00007: val_loss improved from 0.14167 to 0.13600, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00007: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00007: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 4ms/step - loss: 0.1402 - ecoder_output_loss: 0.0799 - classifier_output_loss: 0.6034 - ecoder_output_accuracy: 0.0587 - classifier_output_accuracy: 0.8214 - val_loss: 0.1360 - val_ecoder_output_loss: 0.0791 - val_classifier_output_loss: 0.5687 - val_ecoder_output_accuracy: 0.0504 - val_classifier_output_accuracy: 0.8338\n",
      "Epoch 8/20\n",
      "214/235 [==========================>...] - ETA: 0s - loss: 0.1357 - ecoder_output_loss: 0.0792 - classifier_output_loss: 0.5650 - ecoder_output_accuracy: 0.0571 - classifier_output_accuracy: 0.8335\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00008: val_loss improved from 0.13600 to 0.13156, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00008: val_loss improved from 0.13600 to 0.13156, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00008: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00008: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1357 - ecoder_output_loss: 0.0792 - classifier_output_loss: 0.5653 - ecoder_output_accuracy: 0.0568 - classifier_output_accuracy: 0.8334 - val_loss: 0.1316 - val_ecoder_output_loss: 0.0782 - val_classifier_output_loss: 0.5334 - val_ecoder_output_accuracy: 0.0476 - val_classifier_output_accuracy: 0.8468\n",
      "Epoch 9/20\n",
      "199/235 [========================>.....] - ETA: 0s - loss: 0.1655 - ecoder_output_loss: 0.0824 - classifier_output_loss: 0.8311 - ecoder_output_accuracy: 0.0515 - classifier_output_accuracy: 0.7355\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00009: val_loss did not improve from 0.13156\n",
      "\n",
      "Epoch 00009: val_loss did not improve from 0.13156\n",
      "\n",
      "Epoch 00009: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00009: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.1653 - ecoder_output_loss: 0.0821 - classifier_output_loss: 0.8322 - ecoder_output_accuracy: 0.0493 - classifier_output_accuracy: 0.7354 - val_loss: 0.1575 - val_ecoder_output_loss: 0.0800 - val_classifier_output_loss: 0.7753 - val_ecoder_output_accuracy: 0.0351 - val_classifier_output_accuracy: 0.7707\n",
      "Epoch 10/20\n",
      "214/235 [==========================>...] - ETA: 0s - loss: 0.1495 - ecoder_output_loss: 0.0791 - classifier_output_loss: 0.7043 - ecoder_output_accuracy: 0.0479 - classifier_output_accuracy: 0.7897\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00010: val_loss did not improve from 0.13156\n",
      "\n",
      "Epoch 00010: val_loss did not improve from 0.13156\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "Epoch 00010: saving model to model/AE_model/model_2\\KERAS_check_model_epoch10.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.1488 - ecoder_output_loss: 0.0789 - classifier_output_loss: 0.6985 - ecoder_output_accuracy: 0.0487 - classifier_output_accuracy: 0.7916 - val_loss: 0.1399 - val_ecoder_output_loss: 0.0776 - val_classifier_output_loss: 0.6238 - val_ecoder_output_accuracy: 0.0492 - val_classifier_output_accuracy: 0.8302\n",
      "Epoch 11/20\n",
      "226/235 [===========================>..] - ETA: 0s - loss: 0.1374 - ecoder_output_loss: 0.0774 - classifier_output_loss: 0.5995 - ecoder_output_accuracy: 0.0575 - classifier_output_accuracy: 0.8258\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00011: val_loss did not improve from 0.13156\n",
      "\n",
      "Epoch 00011: val_loss did not improve from 0.13156\n",
      "\n",
      "Epoch 00011: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00011: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 0s 2ms/step - loss: 0.1374 - ecoder_output_loss: 0.0774 - classifier_output_loss: 0.5999 - ecoder_output_accuracy: 0.0578 - classifier_output_accuracy: 0.8261 - val_loss: 0.1330 - val_ecoder_output_loss: 0.0766 - val_classifier_output_loss: 0.5643 - val_ecoder_output_accuracy: 0.0591 - val_classifier_output_accuracy: 0.8467\n",
      "Epoch 12/20\n",
      "227/235 [===========================>..] - ETA: 0s - loss: 0.1319 - ecoder_output_loss: 0.0768 - classifier_output_loss: 0.5509 - ecoder_output_accuracy: 0.0622 - classifier_output_accuracy: 0.8407\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.13156 to 0.12870, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00012: val_loss improved from 0.13156 to 0.12870, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00012: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00012: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1317 - ecoder_output_loss: 0.0768 - classifier_output_loss: 0.5494 - ecoder_output_accuracy: 0.0621 - classifier_output_accuracy: 0.8411 - val_loss: 0.1287 - val_ecoder_output_loss: 0.0761 - val_classifier_output_loss: 0.5261 - val_ecoder_output_accuracy: 0.0555 - val_classifier_output_accuracy: 0.8581\n",
      "Epoch 13/20\n",
      "233/235 [============================>.] - ETA: 0s - loss: 0.1281 - ecoder_output_loss: 0.0763 - classifier_output_loss: 0.5173 - ecoder_output_accuracy: 0.0610 - classifier_output_accuracy: 0.8500\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00013: val_loss improved from 0.12870 to 0.12631, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00013: val_loss improved from 0.12870 to 0.12631, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00013: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00013: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1281 - ecoder_output_loss: 0.0763 - classifier_output_loss: 0.5181 - ecoder_output_accuracy: 0.0611 - classifier_output_accuracy: 0.8497 - val_loss: 0.1263 - val_ecoder_output_loss: 0.0756 - val_classifier_output_loss: 0.5072 - val_ecoder_output_accuracy: 0.0699 - val_classifier_output_accuracy: 0.8604\n",
      "Epoch 14/20\n",
      "222/235 [===========================>..] - ETA: 0s - loss: 0.1253 - ecoder_output_loss: 0.0757 - classifier_output_loss: 0.4961 - ecoder_output_accuracy: 0.0620 - classifier_output_accuracy: 0.8569\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.12631 to 0.12459, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00014: val_loss improved from 0.12631 to 0.12459, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00014: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00014: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.1253 - ecoder_output_loss: 0.0757 - classifier_output_loss: 0.4954 - ecoder_output_accuracy: 0.0625 - classifier_output_accuracy: 0.8569 - val_loss: 0.1246 - val_ecoder_output_loss: 0.0752 - val_classifier_output_loss: 0.4944 - val_ecoder_output_accuracy: 0.0531 - val_classifier_output_accuracy: 0.8629\n",
      "Epoch 15/20\n",
      "198/235 [========================>.....] - ETA: 0s - loss: 0.1234 - ecoder_output_loss: 0.0752 - classifier_output_loss: 0.4821 - ecoder_output_accuracy: 0.0597 - classifier_output_accuracy: 0.8596\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00015: val_loss improved from 0.12459 to 0.12191, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00015: val_loss improved from 0.12459 to 0.12191, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00015: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00015: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.1232 - ecoder_output_loss: 0.0752 - classifier_output_loss: 0.4792 - ecoder_output_accuracy: 0.0595 - classifier_output_accuracy: 0.8603 - val_loss: 0.1219 - val_ecoder_output_loss: 0.0746 - val_classifier_output_loss: 0.4735 - val_ecoder_output_accuracy: 0.0655 - val_classifier_output_accuracy: 0.8698\n",
      "Epoch 16/20\n",
      "210/235 [=========================>....] - ETA: 0s - loss: 0.1213 - ecoder_output_loss: 0.0749 - classifier_output_loss: 0.4643 - ecoder_output_accuracy: 0.0604 - classifier_output_accuracy: 0.8658\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00016: val_loss improved from 0.12191 to 0.12096, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00016: val_loss improved from 0.12191 to 0.12096, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00016: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00016: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 2ms/step - loss: 0.1213 - ecoder_output_loss: 0.0749 - classifier_output_loss: 0.4645 - ecoder_output_accuracy: 0.0603 - classifier_output_accuracy: 0.8651 - val_loss: 0.1210 - val_ecoder_output_loss: 0.0742 - val_classifier_output_loss: 0.4673 - val_ecoder_output_accuracy: 0.0588 - val_classifier_output_accuracy: 0.8736\n",
      "Epoch 17/20\n",
      "227/235 [===========================>..] - ETA: 0s - loss: 0.1199 - ecoder_output_loss: 0.0746 - classifier_output_loss: 0.4530 - ecoder_output_accuracy: 0.0616 - classifier_output_accuracy: 0.8695\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00017: val_loss improved from 0.12096 to 0.12011, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00017: val_loss improved from 0.12096 to 0.12011, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00017: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00017: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1198 - ecoder_output_loss: 0.0746 - classifier_output_loss: 0.4527 - ecoder_output_accuracy: 0.0616 - classifier_output_accuracy: 0.8694 - val_loss: 0.1201 - val_ecoder_output_loss: 0.0739 - val_classifier_output_loss: 0.4620 - val_ecoder_output_accuracy: 0.0593 - val_classifier_output_accuracy: 0.8722\n",
      "Epoch 18/20\n",
      "218/235 [==========================>...] - ETA: 0s - loss: 0.1186 - ecoder_output_loss: 0.0744 - classifier_output_loss: 0.4426 - ecoder_output_accuracy: 0.0666 - classifier_output_accuracy: 0.8718\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.12011 to 0.11878, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00018: val_loss improved from 0.12011 to 0.11878, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00018: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00018: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 4ms/step - loss: 0.1186 - ecoder_output_loss: 0.0743 - classifier_output_loss: 0.4430 - ecoder_output_accuracy: 0.0668 - classifier_output_accuracy: 0.8716 - val_loss: 0.1188 - val_ecoder_output_loss: 0.0738 - val_classifier_output_loss: 0.4501 - val_ecoder_output_accuracy: 0.0650 - val_classifier_output_accuracy: 0.8778\n",
      "Epoch 19/20\n",
      "214/235 [==========================>...] - ETA: 0s - loss: 0.1174 - ecoder_output_loss: 0.0741 - classifier_output_loss: 0.4325 - ecoder_output_accuracy: 0.0675 - classifier_output_accuracy: 0.8752\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00019: val_loss did not improve from 0.11878\n",
      "\n",
      "Epoch 00019: val_loss did not improve from 0.11878\n",
      "\n",
      "Epoch 00019: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00019: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1175 - ecoder_output_loss: 0.0741 - classifier_output_loss: 0.4343 - ecoder_output_accuracy: 0.0674 - classifier_output_accuracy: 0.8748 - val_loss: 0.1189 - val_ecoder_output_loss: 0.0735 - val_classifier_output_loss: 0.4541 - val_ecoder_output_accuracy: 0.0682 - val_classifier_output_accuracy: 0.8786\n",
      "Epoch 20/20\n",
      "230/235 [============================>.] - ETA: 0s - loss: 0.1166 - ecoder_output_loss: 0.0739 - classifier_output_loss: 0.4270 - ecoder_output_accuracy: 0.0687 - classifier_output_accuracy: 0.8765\n",
      "***callbacks***\n",
      "saving losses to model/AE_model/model_2\\losses.log\n",
      "\n",
      "Epoch 00020: val_loss improved from 0.11878 to 0.11747, saving model to model/AE_model/model_2\\KERAS_check_best_model.h5\n",
      "\n",
      "Epoch 00020: val_loss improved from 0.11878 to 0.11747, saving model to model/AE_model/model_2\\KERAS_check_best_model_weights.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_last.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_last_weights.h5\n",
      "\n",
      "Epoch 00020: saving model to model/AE_model/model_2\\KERAS_check_model_epoch20.h5\n",
      "\n",
      "***callbacks end***\n",
      "\n",
      "235/235 [==============================] - 1s 3ms/step - loss: 0.1165 - ecoder_output_loss: 0.0739 - classifier_output_loss: 0.4266 - ecoder_output_accuracy: 0.0687 - classifier_output_accuracy: 0.8765 - val_loss: 0.1175 - val_ecoder_output_loss: 0.0733 - val_classifier_output_loss: 0.4416 - val_ecoder_output_accuracy: 0.0703 - val_classifier_output_accuracy: 0.8821\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "network_with_classifier = Autoencoder(data_zoom, pruned=True)\n",
    "network_with_classifier.build_encoder()\n",
    "network_with_classifier.build_decoder()\n",
    "network_with_classifier.build_classifier()\n",
    "network_with_classifier.build_autoencoder(use_latent_classifier=True)\n",
    "network_with_classifier.fit_data( batch_size=256,epochs=20, use_latent_classifier=True)\n",
    "network_with_classifier.plot_latent_dimension(\"Autoencoder with classifier\")\n",
    "network_with_classifier.plot_score(use_latent_classifier=True,model_name=\"Autoencoder with classifier\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "789093bb736187c46b4a361b2c83e244a48ff9bb7ad40e66e5ffb40455c8edff"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}