IndusNet_7 classes_100_100_2nd Experiment.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "resized_images_100_100_train_dir='D:\\\\TensorFlow Programs\\\\fruit_image_classification_Indus Fruit Dataset\\\\Indus Fruit Dataset_Resized Images_100_100\\\\Train'\n",
    "resized_images_100_100_test_dir='D:\\\\TensorFlow Programs\\\\fruit_image_classification_Indus Fruit Dataset\\\\Indus Fruit Dataset_Resized Images_100_100\\\\Test'\n",
    "resized_images_100_100_validation_dir='D:\\\\TensorFlow Programs\\\\fruit_image_classification_Indus Fruit Dataset\\\\Indus Fruit Dataset_Resized Images_100_100\\\\Validation'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\Raheel\\Anaconda3\\envs\\EnvName\\lib\\site-packages\\tensorflow\\python\\ops\\resource_variable_ops.py:435: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n",
      "WARNING:tensorflow:From C:\\Users\\Raheel\\Anaconda3\\envs\\EnvName\\lib\\site-packages\\tensorflow\\python\\keras\\layers\\core.py:143: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 98, 98, 64)        1792      \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 96, 96, 64)        36928     \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 94, 94, 128)       73856     \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 92, 92, 128)       147584    \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 46, 46, 128)       0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 46, 46, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_4 (Conv2D)            (None, 44, 44, 128)       147584    \n",
      "_________________________________________________________________\n",
      "conv2d_5 (Conv2D)            (None, 42, 42, 128)       147584    \n",
      "_________________________________________________________________\n",
      "conv2d_6 (Conv2D)            (None, 40, 40, 256)       295168    \n",
      "_________________________________________________________________\n",
      "conv2d_7 (Conv2D)            (None, 38, 38, 256)       590080    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 19, 19, 256)       0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 19, 19, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 17, 17, 256)       590080    \n",
      "_________________________________________________________________\n",
      "conv2d_9 (Conv2D)            (None, 15, 15, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 25088)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 32)                802848    \n",
      "_________________________________________________________________\n",
      "dropout_3 (Dropout)          (None, 32)                0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 7)                 231       \n",
      "=================================================================\n",
      "Total params: 4,013,895\n",
      "Trainable params: 4,013,895\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import keras\n",
    "from tensorflow.python.keras import models\n",
    "from tensorflow.python.keras import layers\n",
    "from tensorflow.python.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.python.keras import regularizers\n",
    "from tensorflow.python.keras import optimizers\n",
    "from tensorflow.python.keras.callbacks import ModelCheckpoint\n",
    "\n",
    "indusNetModel=models.Sequential()\n",
    "\n",
    "indusNetModel.add(layers.Conv2D(64, (3, 3), activation='relu', input_shape=(100, 100, 3)))\n",
    "indusNetModel.add(layers.Conv2D(64, (3, 3), activation='relu'))\n",
    "indusNetModel.add(layers.Conv2D(128, (3, 3), activation='relu'))\n",
    "indusNetModel.add(layers.Conv2D(128, (3, 3), activation='relu'))\n",
    "indusNetModel.add(layers.MaxPooling2D(pool_size=(2, 2)))\n",
    "indusNetModel.add(layers.Dropout(0.2))\n",
    "\n",
    "indusNetModel.add(layers.Conv2D(128, (3, 3), activation='relu'))\n",
    "indusNetModel.add(layers.Conv2D(128, (3, 3), activation='relu'))\n",
    "indusNetModel.add(layers.Conv2D(256, (3, 3), activation='relu'))\n",
    "indusNetModel.add(layers.Conv2D(256, (3, 3), activation='relu'))\n",
    "indusNetModel.add(layers.MaxPooling2D(pool_size=(2, 2)))\n",
    "indusNetModel.add(layers.Dropout(0.2))\n",
    "\n",
    "indusNetModel.add(layers.Conv2D(256, (3, 3), activation='relu'))\n",
    "indusNetModel.add(layers.Conv2D(512, (3, 3), activation='relu'))\n",
    "indusNetModel.add(layers.MaxPooling2D(pool_size=(2, 2)))\n",
    "indusNetModel.add(layers.Dropout(0.2))\n",
    "\n",
    "indusNetModel.add(layers.Flatten())\n",
    "indusNetModel.add(layers.Dense(32, kernel_regularizer=regularizers.l2(0.001), activation='relu'))\n",
    "indusNetModel.add(layers.Dropout(0.2))\n",
    "indusNetModel.add(layers.Dense(7, activation='softmax'))\n",
    "\n",
    "indusNetModel.compile(loss='categorical_crossentropy',optimizer=optimizers.Adam(lr=1e-4),metrics=['categorical_accuracy'])\n",
    "indusNetModel.summary()\n",
    "\n",
    "filepath=\"IndusNet_7_classes_100_100_2nd Experiment.h5\"\n",
    "checkpoint = ModelCheckpoint(filepath, save_best_only=True)\n",
    "callbacks_list = [checkpoint]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 2800 images belonging to 7 classes.\n",
      "Found 280 images belonging to 7 classes.\n",
      "Found 560 images belonging to 7 classes.\n"
     ]
    }
   ],
   "source": [
    "batch_size=16\n",
    "\n",
    "train_datagen = ImageDataGenerator(\n",
    "    rescale=1./255,\n",
    "    rotation_range=40,\n",
    "    width_shift_range=0.2,\n",
    "    height_shift_range=0.2,\n",
    "    shear_range=0.2,\n",
    "    zoom_range=0.2,\n",
    "    horizontal_flip=True,)\n",
    "\n",
    "# Note that the validation  and test data should not be augmented!\n",
    "validation_datagen=ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "test_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "train_generator = train_datagen.flow_from_directory(\n",
    "        # This is the target directory\n",
    "        resized_images_100_100_train_dir,\n",
    "        target_size=(100, 100),\n",
    "        batch_size=batch_size)\n",
    "\n",
    "validation_generator = validation_datagen.flow_from_directory(\n",
    "        resized_images_100_100_validation_dir,\n",
    "        target_size=(100, 100),\n",
    "        batch_size=batch_size, shuffle=False)\n",
    "\n",
    "test_generator = test_datagen.flow_from_directory(\n",
    "        resized_images_100_100_test_dir,\n",
    "        target_size=(100, 100),\n",
    "        batch_size=batch_size, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "number_of_training_samples=2800\n",
    "number_of_validation_samples=280\n",
    "number_of_test_samples=560"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\Raheel\\Anaconda3\\envs\\EnvName\\lib\\site-packages\\tensorflow\\python\\ops\\math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Epoch 1/400\n",
      "18/18 [==============================] - 3s 156ms/step - loss: 1.9654 - categorical_accuracy: 0.1321\n",
      "175/175 [==============================] - 46s 263ms/step - loss: 1.9643 - categorical_accuracy: 0.1568 - val_loss: 1.9654 - val_categorical_accuracy: 0.1321\n",
      "Epoch 2/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 1.6903 - categorical_accuracy: 0.3714\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 1.8254 - categorical_accuracy: 0.2411 - val_loss: 1.6903 - val_categorical_accuracy: 0.3714\n",
      "Epoch 3/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 1.3723 - categorical_accuracy: 0.4357\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 1.4911 - categorical_accuracy: 0.3911 - val_loss: 1.3723 - val_categorical_accuracy: 0.4357\n",
      "Epoch 4/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 1.3192 - categorical_accuracy: 0.4536\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 1.4158 - categorical_accuracy: 0.4011 - val_loss: 1.3192 - val_categorical_accuracy: 0.4536\n",
      "Epoch 5/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 1.3555 - categorical_accuracy: 0.4607\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 1.3920 - categorical_accuracy: 0.4304 - val_loss: 1.3555 - val_categorical_accuracy: 0.4607\n",
      "Epoch 6/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 1.1746 - categorical_accuracy: 0.5536\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 1.3292 - categorical_accuracy: 0.4589 - val_loss: 1.1746 - val_categorical_accuracy: 0.5536\n",
      "Epoch 7/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 1.1825 - categorical_accuracy: 0.5607\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 1.2865 - categorical_accuracy: 0.4943 - val_loss: 1.1825 - val_categorical_accuracy: 0.5607\n",
      "Epoch 8/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 1.1340 - categorical_accuracy: 0.5679\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 1.2535 - categorical_accuracy: 0.4896 - val_loss: 1.1340 - val_categorical_accuracy: 0.5679\n",
      "Epoch 9/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 1.3352 - categorical_accuracy: 0.4500\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 1.2457 - categorical_accuracy: 0.5086 - val_loss: 1.3352 - val_categorical_accuracy: 0.4500\n",
      "Epoch 10/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 1.1153 - categorical_accuracy: 0.5571\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 1.2171 - categorical_accuracy: 0.5186 - val_loss: 1.1153 - val_categorical_accuracy: 0.5571\n",
      "Epoch 11/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 1.1631 - categorical_accuracy: 0.5357\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 1.1885 - categorical_accuracy: 0.5293 - val_loss: 1.1631 - val_categorical_accuracy: 0.5357\n",
      "Epoch 12/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 1.0404 - categorical_accuracy: 0.5929\n",
      "175/175 [==============================] - 35s 199ms/step - loss: 1.1718 - categorical_accuracy: 0.5379 - val_loss: 1.0404 - val_categorical_accuracy: 0.5929\n",
      "Epoch 13/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 1.0997 - categorical_accuracy: 0.5714\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 1.1483 - categorical_accuracy: 0.5450 - val_loss: 1.0997 - val_categorical_accuracy: 0.5714\n",
      "Epoch 14/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.9742 - categorical_accuracy: 0.6286\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 1.1068 - categorical_accuracy: 0.5768 - val_loss: 0.9742 - val_categorical_accuracy: 0.6286\n",
      "Epoch 15/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 1.2649 - categorical_accuracy: 0.5786\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 1.0553 - categorical_accuracy: 0.6000 - val_loss: 1.2649 - val_categorical_accuracy: 0.5786\n",
      "Epoch 16/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.9286 - categorical_accuracy: 0.6321\n",
      "175/175 [==============================] - 34s 197ms/step - loss: 1.0449 - categorical_accuracy: 0.5989 - val_loss: 0.9286 - val_categorical_accuracy: 0.6321\n",
      "Epoch 17/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.9801 - categorical_accuracy: 0.6357\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 1.0173 - categorical_accuracy: 0.6032 - val_loss: 0.9801 - val_categorical_accuracy: 0.6357\n",
      "Epoch 18/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 0.8646 - categorical_accuracy: 0.6607\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 1.0041 - categorical_accuracy: 0.6182 - val_loss: 0.8646 - val_categorical_accuracy: 0.6607\n",
      "Epoch 19/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.9001 - categorical_accuracy: 0.6500\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.9801 - categorical_accuracy: 0.6300 - val_loss: 0.9001 - val_categorical_accuracy: 0.6500\n",
      "Epoch 20/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.9308 - categorical_accuracy: 0.6464\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.9651 - categorical_accuracy: 0.6293 - val_loss: 0.9308 - val_categorical_accuracy: 0.6464\n",
      "Epoch 21/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.8440 - categorical_accuracy: 0.6750\n",
      "175/175 [==============================] - 35s 198ms/step - loss: 0.9085 - categorical_accuracy: 0.6554 - val_loss: 0.8440 - val_categorical_accuracy: 0.6750\n",
      "Epoch 22/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 0.7590 - categorical_accuracy: 0.7179\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.8725 - categorical_accuracy: 0.6704 - val_loss: 0.7590 - val_categorical_accuracy: 0.7179\n",
      "Epoch 23/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.8582 - categorical_accuracy: 0.6893\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.9061 - categorical_accuracy: 0.6657 - val_loss: 0.8582 - val_categorical_accuracy: 0.6893\n",
      "Epoch 24/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 0.7980 - categorical_accuracy: 0.6857\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.8615 - categorical_accuracy: 0.6739 - val_loss: 0.7980 - val_categorical_accuracy: 0.6857\n",
      "Epoch 25/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.8530 - categorical_accuracy: 0.7107\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.8566 - categorical_accuracy: 0.6779 - val_loss: 0.8530 - val_categorical_accuracy: 0.7107\n",
      "Epoch 26/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.7445 - categorical_accuracy: 0.7214\n",
      "175/175 [==============================] - 35s 198ms/step - loss: 0.8309 - categorical_accuracy: 0.6900 - val_loss: 0.7445 - val_categorical_accuracy: 0.7214\n",
      "Epoch 27/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.7352 - categorical_accuracy: 0.7107A:  - ETA: 0s - loss: 0.7400 - categorical_accuracy: 0.\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.8275 - categorical_accuracy: 0.6986 - val_loss: 0.7352 - val_categorical_accuracy: 0.7107\n",
      "Epoch 28/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6957 - categorical_accuracy: 0.7286\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.7816 - categorical_accuracy: 0.7246 - val_loss: 0.6957 - val_categorical_accuracy: 0.7286\n",
      "Epoch 29/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6824 - categorical_accuracy: 0.7393\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.7827 - categorical_accuracy: 0.7164 - val_loss: 0.6824 - val_categorical_accuracy: 0.7393\n",
      "Epoch 30/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6959 - categorical_accuracy: 0.7357\n",
      "175/175 [==============================] - 34s 191ms/step - loss: 0.7523 - categorical_accuracy: 0.7268 - val_loss: 0.6959 - val_categorical_accuracy: 0.7357\n",
      "Epoch 31/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.7812 - categorical_accuracy: 0.7500\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.7713 - categorical_accuracy: 0.7329 - val_loss: 0.7812 - val_categorical_accuracy: 0.7500\n",
      "Epoch 32/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6505 - categorical_accuracy: 0.7607A: 1s\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.7413 - categorical_accuracy: 0.7314 - val_loss: 0.6505 - val_categorical_accuracy: 0.7607\n",
      "Epoch 33/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6603 - categorical_accuracy: 0.7393\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.7198 - categorical_accuracy: 0.7368 - val_loss: 0.6603 - val_categorical_accuracy: 0.7393\n",
      "Epoch 34/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.6444 - categorical_accuracy: 0.7607\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.7349 - categorical_accuracy: 0.7357 - val_loss: 0.6444 - val_categorical_accuracy: 0.7607\n",
      "Epoch 35/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.8430 - categorical_accuracy: 0.6714\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.7105 - categorical_accuracy: 0.7421 - val_loss: 0.8430 - val_categorical_accuracy: 0.6714\n",
      "Epoch 36/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.6782 - categorical_accuracy: 0.7500\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.7048 - categorical_accuracy: 0.7482 - val_loss: 0.6782 - val_categorical_accuracy: 0.7500\n",
      "Epoch 37/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.6687 - categorical_accuracy: 0.7536\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.6938 - categorical_accuracy: 0.7461 - val_loss: 0.6687 - val_categorical_accuracy: 0.7536\n",
      "Epoch 38/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.6601 - categorical_accuracy: 0.7464A: 1s -\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.6610 - categorical_accuracy: 0.7643 - val_loss: 0.6601 - val_categorical_accuracy: 0.7464\n",
      "Epoch 39/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.6788 - categorical_accuracy: 0.7643\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.6635 - categorical_accuracy: 0.7671 - val_loss: 0.6788 - val_categorical_accuracy: 0.7643\n",
      "Epoch 40/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.8998 - categorical_accuracy: 0.6750\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.6690 - categorical_accuracy: 0.7668 - val_loss: 0.8998 - val_categorical_accuracy: 0.6750\n",
      "Epoch 41/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.6732 - categorical_accuracy: 0.7643\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.6608 - categorical_accuracy: 0.7654 - val_loss: 0.6732 - val_categorical_accuracy: 0.7643\n",
      "Epoch 42/400\n",
      "18/18 [==============================] - 1s 70ms/step - loss: 0.7082 - categorical_accuracy: 0.7429 0s - loss: 1.0639 - catego\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.6196 - categorical_accuracy: 0.7786 - val_loss: 0.7082 - val_categorical_accuracy: 0.7429\n",
      "Epoch 43/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.7756 - categorical_accuracy: 0.7536\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.6676 - categorical_accuracy: 0.7693 - val_loss: 0.7756 - val_categorical_accuracy: 0.7536\n",
      "Epoch 44/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.7074 - categorical_accuracy: 0.7393\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.6084 - categorical_accuracy: 0.7893 - val_loss: 0.7074 - val_categorical_accuracy: 0.7393\n",
      "Epoch 45/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.7431 - categorical_accuracy: 0.7571\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.6008 - categorical_accuracy: 0.7943 - val_loss: 0.7431 - val_categorical_accuracy: 0.7571\n",
      "Epoch 46/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.7176 - categorical_accuracy: 0.7536\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.6231 - categorical_accuracy: 0.7839 - val_loss: 0.7176 - val_categorical_accuracy: 0.7536\n",
      "Epoch 47/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.7273 - categorical_accuracy: 0.7821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.5895 - categorical_accuracy: 0.7946 - val_loss: 0.7273 - val_categorical_accuracy: 0.7821\n",
      "Epoch 48/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.6391 - categorical_accuracy: 0.7714\n",
      "175/175 [==============================] - 35s 199ms/step - loss: 0.6050 - categorical_accuracy: 0.7907 - val_loss: 0.6391 - val_categorical_accuracy: 0.7714\n",
      "Epoch 49/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6625 - categorical_accuracy: 0.7607A:  - ETA: 0s - loss: 0.5\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.5797 - categorical_accuracy: 0.7925 - val_loss: 0.6625 - val_categorical_accuracy: 0.7607\n",
      "Epoch 50/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5750 - categorical_accuracy: 0.7929\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.5755 - categorical_accuracy: 0.7975 - val_loss: 0.5750 - val_categorical_accuracy: 0.7929\n",
      "Epoch 51/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6620 - categorical_accuracy: 0.7750\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.5580 - categorical_accuracy: 0.8025 - val_loss: 0.6620 - val_categorical_accuracy: 0.7750\n",
      "Epoch 52/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.5703 - categorical_accuracy: 0.7929\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.5424 - categorical_accuracy: 0.8111 - val_loss: 0.5703 - val_categorical_accuracy: 0.7929\n",
      "Epoch 53/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5600 - categorical_accuracy: 0.8214\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.5711 - categorical_accuracy: 0.8032 - val_loss: 0.5600 - val_categorical_accuracy: 0.8214\n",
      "Epoch 54/400\n",
      "18/18 [==============================] - 1s 78ms/step - loss: 0.6387 - categorical_accuracy: 0.8036\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.5297 - categorical_accuracy: 0.8143 - val_loss: 0.6387 - val_categorical_accuracy: 0.8036\n",
      "Epoch 55/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.5487 - categorical_accuracy: 0.8107\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.5544 - categorical_accuracy: 0.8025 - val_loss: 0.5487 - val_categorical_accuracy: 0.8107\n",
      "Epoch 56/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.5471 - categorical_accuracy: 0.8036\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.5398 - categorical_accuracy: 0.8200 - val_loss: 0.5471 - val_categorical_accuracy: 0.8036\n",
      "Epoch 57/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.5519 - categorical_accuracy: 0.7929\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.5210 - categorical_accuracy: 0.8204 - val_loss: 0.5519 - val_categorical_accuracy: 0.7929\n",
      "Epoch 58/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.6197 - categorical_accuracy: 0.7821\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.5169 - categorical_accuracy: 0.8261 - val_loss: 0.6197 - val_categorical_accuracy: 0.7821\n",
      "Epoch 59/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5278 - categorical_accuracy: 0.8214\n",
      "175/175 [==============================] - 35s 197ms/step - loss: 0.5029 - categorical_accuracy: 0.8282 - val_loss: 0.5278 - val_categorical_accuracy: 0.8214\n",
      "Epoch 60/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6715 - categorical_accuracy: 0.7821\n",
      "175/175 [==============================] - 34s 191ms/step - loss: 0.5106 - categorical_accuracy: 0.8225 - val_loss: 0.6715 - val_categorical_accuracy: 0.7821\n",
      "Epoch 61/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 0.5583 - categorical_accuracy: 0.8071\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.5020 - categorical_accuracy: 0.8361 - val_loss: 0.5583 - val_categorical_accuracy: 0.8071\n",
      "Epoch 62/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.5530 - categorical_accuracy: 0.8214\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.4670 - categorical_accuracy: 0.8350 - val_loss: 0.5530 - val_categorical_accuracy: 0.8214\n",
      "Epoch 63/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5553 - categorical_accuracy: 0.8286\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.4807 - categorical_accuracy: 0.8375 - val_loss: 0.5553 - val_categorical_accuracy: 0.8286\n",
      "Epoch 64/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5128 - categorical_accuracy: 0.8107\n",
      "175/175 [==============================] - 35s 199ms/step - loss: 0.4618 - categorical_accuracy: 0.8400 - val_loss: 0.5128 - val_categorical_accuracy: 0.8107\n",
      "Epoch 65/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.5306 - categorical_accuracy: 0.8429\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.4676 - categorical_accuracy: 0.8482 - val_loss: 0.5306 - val_categorical_accuracy: 0.8429\n",
      "Epoch 66/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5646 - categorical_accuracy: 0.8179\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.4598 - categorical_accuracy: 0.8514 - val_loss: 0.5646 - val_categorical_accuracy: 0.8179\n",
      "Epoch 67/400\n",
      "18/18 [==============================] - 1s 70ms/step - loss: 0.4973 - categorical_accuracy: 0.8357\n",
      "175/175 [==============================] - 35s 200ms/step - loss: 0.4646 - categorical_accuracy: 0.8414 - val_loss: 0.4973 - val_categorical_accuracy: 0.8357\n",
      "Epoch 68/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.5222 - categorical_accuracy: 0.8429\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.4437 - categorical_accuracy: 0.8450 - val_loss: 0.5222 - val_categorical_accuracy: 0.8429\n",
      "Epoch 69/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.5120 - categorical_accuracy: 0.8393ETA: 13s - loss: 0.3980 - cate - ETA: 4s - loss: 0.4351 - cate\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.4385 - categorical_accuracy: 0.8514 - val_loss: 0.5120 - val_categorical_accuracy: 0.8393\n",
      "Epoch 70/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.5869 - categorical_accuracy: 0.8036\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.4050 - categorical_accuracy: 0.8693 - val_loss: 0.5869 - val_categorical_accuracy: 0.8036\n",
      "Epoch 71/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.5726 - categorical_accuracy: 0.8107A: 0s - los\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.4368 - categorical_accuracy: 0.8479 - val_loss: 0.5726 - val_categorical_accuracy: 0.8107\n",
      "Epoch 72/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5826 - categorical_accuracy: 0.8464\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.4266 - categorical_accuracy: 0.8564 - val_loss: 0.5826 - val_categorical_accuracy: 0.8464\n",
      "Epoch 73/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4938 - categorical_accuracy: 0.8321\n",
      "175/175 [==============================] - 35s 200ms/step - loss: 0.4032 - categorical_accuracy: 0.8689 - val_loss: 0.4938 - val_categorical_accuracy: 0.8321\n",
      "Epoch 74/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5020 - categorical_accuracy: 0.8500\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3970 - categorical_accuracy: 0.8646 - val_loss: 0.5020 - val_categorical_accuracy: 0.8500\n",
      "Epoch 75/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.6098 - categorical_accuracy: 0.8286\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3908 - categorical_accuracy: 0.8718 - val_loss: 0.6098 - val_categorical_accuracy: 0.8286\n",
      "Epoch 76/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5694 - categorical_accuracy: 0.8179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3920 - categorical_accuracy: 0.8661 - val_loss: 0.5694 - val_categorical_accuracy: 0.8179\n",
      "Epoch 77/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5105 - categorical_accuracy: 0.8536\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3904 - categorical_accuracy: 0.8700 - val_loss: 0.5105 - val_categorical_accuracy: 0.8536\n",
      "Epoch 78/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4543 - categorical_accuracy: 0.8643\n",
      "175/175 [==============================] - 35s 199ms/step - loss: 0.3821 - categorical_accuracy: 0.8736 - val_loss: 0.4543 - val_categorical_accuracy: 0.8643\n",
      "Epoch 79/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.5930 - categorical_accuracy: 0.8357\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.3682 - categorical_accuracy: 0.8782 - val_loss: 0.5930 - val_categorical_accuracy: 0.8357\n",
      "Epoch 80/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 0.7356 - categorical_accuracy: 0.7857\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.3803 - categorical_accuracy: 0.8754 - val_loss: 0.7356 - val_categorical_accuracy: 0.7857\n",
      "Epoch 81/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5368 - categorical_accuracy: 0.8464\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.3946 - categorical_accuracy: 0.8779 - val_loss: 0.5368 - val_categorical_accuracy: 0.8464\n",
      "Epoch 82/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4907 - categorical_accuracy: 0.8429\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3619 - categorical_accuracy: 0.8893 - val_loss: 0.4907 - val_categorical_accuracy: 0.8429\n",
      "Epoch 83/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4856 - categorical_accuracy: 0.8643\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3952 - categorical_accuracy: 0.8768 - val_loss: 0.4856 - val_categorical_accuracy: 0.8643\n",
      "Epoch 84/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5271 - categorical_accuracy: 0.8357\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3644 - categorical_accuracy: 0.8818 - val_loss: 0.5271 - val_categorical_accuracy: 0.8357\n",
      "Epoch 85/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4284 - categorical_accuracy: 0.8643\n",
      "175/175 [==============================] - 35s 199ms/step - loss: 0.3865 - categorical_accuracy: 0.8771 - val_loss: 0.4284 - val_categorical_accuracy: 0.8643\n",
      "Epoch 86/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6039 - categorical_accuracy: 0.8429\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.3443 - categorical_accuracy: 0.8893 - val_loss: 0.6039 - val_categorical_accuracy: 0.8429\n",
      "Epoch 87/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.4752 - categorical_accuracy: 0.8607\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.3461 - categorical_accuracy: 0.8861 - val_loss: 0.4752 - val_categorical_accuracy: 0.8607\n",
      "Epoch 88/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4213 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 35s 198ms/step - loss: 0.3834 - categorical_accuracy: 0.8725 - val_loss: 0.4213 - val_categorical_accuracy: 0.8750\n",
      "Epoch 89/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.5410 - categorical_accuracy: 0.8500A: 0s - l\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.3171 - categorical_accuracy: 0.8993 - val_loss: 0.5410 - val_categorical_accuracy: 0.8500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 90/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.4159 - categorical_accuracy: 0.8786ETA: 24s\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3297 - categorical_accuracy: 0.8946 - val_loss: 0.4159 - val_categorical_accuracy: 0.8786\n",
      "Epoch 91/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.4204 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.3500 - categorical_accuracy: 0.8889 - val_loss: 0.4204 - val_categorical_accuracy: 0.8786\n",
      "Epoch 92/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.4625 - categorical_accuracy: 0.8607\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.3358 - categorical_accuracy: 0.8857 - val_loss: 0.4625 - val_categorical_accuracy: 0.8607\n",
      "Epoch 93/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5077 - categorical_accuracy: 0.8357\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.3235 - categorical_accuracy: 0.8950 - val_loss: 0.5077 - val_categorical_accuracy: 0.8357\n",
      "Epoch 94/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4913 - categorical_accuracy: 0.8643\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3383 - categorical_accuracy: 0.8893 - val_loss: 0.4913 - val_categorical_accuracy: 0.8643\n",
      "Epoch 95/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4366 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3169 - categorical_accuracy: 0.8936 - val_loss: 0.4366 - val_categorical_accuracy: 0.8821\n",
      "Epoch 96/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.4597 - categorical_accuracy: 0.8607\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3276 - categorical_accuracy: 0.8957 - val_loss: 0.4597 - val_categorical_accuracy: 0.8607\n",
      "Epoch 97/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4998 - categorical_accuracy: 0.8536\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.3159 - categorical_accuracy: 0.9046 - val_loss: 0.4998 - val_categorical_accuracy: 0.8536\n",
      "Epoch 98/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4951 - categorical_accuracy: 0.8607\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3093 - categorical_accuracy: 0.9025 - val_loss: 0.4951 - val_categorical_accuracy: 0.8607\n",
      "Epoch 99/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5261 - categorical_accuracy: 0.8429\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3172 - categorical_accuracy: 0.8957 - val_loss: 0.5261 - val_categorical_accuracy: 0.8429\n",
      "Epoch 100/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4465 - categorical_accuracy: 0.8679\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3354 - categorical_accuracy: 0.8904 - val_loss: 0.4465 - val_categorical_accuracy: 0.8679\n",
      "Epoch 101/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4633 - categorical_accuracy: 0.8607\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2962 - categorical_accuracy: 0.9079 - val_loss: 0.4633 - val_categorical_accuracy: 0.8607\n",
      "Epoch 102/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4468 - categorical_accuracy: 0.8429A: 0s - loss:\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2876 - categorical_accuracy: 0.9071 - val_loss: 0.4468 - val_categorical_accuracy: 0.8429\n",
      "Epoch 103/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4326 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2914 - categorical_accuracy: 0.9100 - val_loss: 0.4326 - val_categorical_accuracy: 0.8821\n",
      "Epoch 104/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4758 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.3163 - categorical_accuracy: 0.8982 - val_loss: 0.4758 - val_categorical_accuracy: 0.8821\n",
      "Epoch 105/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.3886 - categorical_accuracy: 0.8679\n",
      "175/175 [==============================] - 35s 199ms/step - loss: 0.3207 - categorical_accuracy: 0.9021 - val_loss: 0.3886 - val_categorical_accuracy: 0.8679\n",
      "Epoch 106/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5249 - categorical_accuracy: 0.8643\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.2764 - categorical_accuracy: 0.9150 - val_loss: 0.5249 - val_categorical_accuracy: 0.8643\n",
      "Epoch 107/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.7130 - categorical_accuracy: 0.8071\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2940 - categorical_accuracy: 0.9089 - val_loss: 0.7130 - val_categorical_accuracy: 0.8071\n",
      "Epoch 108/400\n",
      "18/18 [==============================] - 1s 70ms/step - loss: 0.5554 - categorical_accuracy: 0.8607\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2918 - categorical_accuracy: 0.9154 - val_loss: 0.5554 - val_categorical_accuracy: 0.8607\n",
      "Epoch 109/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.4912 - categorical_accuracy: 0.8607\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2766 - categorical_accuracy: 0.9136 - val_loss: 0.4912 - val_categorical_accuracy: 0.8607\n",
      "Epoch 110/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4431 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.2803 - categorical_accuracy: 0.9061 - val_loss: 0.4431 - val_categorical_accuracy: 0.8786\n",
      "Epoch 111/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4945 - categorical_accuracy: 0.8679\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2877 - categorical_accuracy: 0.9114 - val_loss: 0.4945 - val_categorical_accuracy: 0.8679\n",
      "Epoch 112/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.4470 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.2692 - categorical_accuracy: 0.9161 - val_loss: 0.4470 - val_categorical_accuracy: 0.8750\n",
      "Epoch 113/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4804 - categorical_accuracy: 0.8643\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2710 - categorical_accuracy: 0.9146 - val_loss: 0.4804 - val_categorical_accuracy: 0.8643\n",
      "Epoch 114/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3945 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2744 - categorical_accuracy: 0.9161 - val_loss: 0.3945 - val_categorical_accuracy: 0.9036\n",
      "Epoch 115/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4262 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2576 - categorical_accuracy: 0.9175 - val_loss: 0.4262 - val_categorical_accuracy: 0.8750\n",
      "Epoch 116/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4697 - categorical_accuracy: 0.8643\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2875 - categorical_accuracy: 0.9096 - val_loss: 0.4697 - val_categorical_accuracy: 0.8643\n",
      "Epoch 117/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5452 - categorical_accuracy: 0.8536\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.2985 - categorical_accuracy: 0.9114 - val_loss: 0.5452 - val_categorical_accuracy: 0.8536\n",
      "Epoch 118/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4147 - categorical_accuracy: 0.8857\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2716 - categorical_accuracy: 0.9150 - val_loss: 0.4147 - val_categorical_accuracy: 0.8857\n",
      "Epoch 119/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4263 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2546 - categorical_accuracy: 0.9225 - val_loss: 0.4263 - val_categorical_accuracy: 0.8929\n",
      "Epoch 120/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4347 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2435 - categorical_accuracy: 0.9268 - val_loss: 0.4347 - val_categorical_accuracy: 0.8786\n",
      "Epoch 121/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5071 - categorical_accuracy: 0.8714\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2453 - categorical_accuracy: 0.9264 - val_loss: 0.5071 - val_categorical_accuracy: 0.8714\n",
      "Epoch 122/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4384 - categorical_accuracy: 0.8750 0s - loss: 0.4884 - categorical_accura\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2860 - categorical_accuracy: 0.9154 - val_loss: 0.4384 - val_categorical_accuracy: 0.8750\n",
      "Epoch 123/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3982 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2628 - categorical_accuracy: 0.9171 - val_loss: 0.3982 - val_categorical_accuracy: 0.9179\n",
      "Epoch 124/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3990 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2292 - categorical_accuracy: 0.9314 - val_loss: 0.3990 - val_categorical_accuracy: 0.8821\n",
      "Epoch 125/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4531 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2463 - categorical_accuracy: 0.9261 - val_loss: 0.4531 - val_categorical_accuracy: 0.8786\n",
      "Epoch 126/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4502 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2451 - categorical_accuracy: 0.9300 - val_loss: 0.4502 - val_categorical_accuracy: 0.8929\n",
      "Epoch 127/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4772 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2518 - categorical_accuracy: 0.9207 - val_loss: 0.4772 - val_categorical_accuracy: 0.9000\n",
      "Epoch 128/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4520 - categorical_accuracy: 0.8679\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2278 - categorical_accuracy: 0.9268 - val_loss: 0.4520 - val_categorical_accuracy: 0.8679\n",
      "Epoch 129/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5147 - categorical_accuracy: 0.8679\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2353 - categorical_accuracy: 0.9207 - val_loss: 0.5147 - val_categorical_accuracy: 0.8679\n",
      "Epoch 130/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4778 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2516 - categorical_accuracy: 0.9246 - val_loss: 0.4778 - val_categorical_accuracy: 0.8893\n",
      "Epoch 131/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4604 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.2434 - categorical_accuracy: 0.9218 - val_loss: 0.4604 - val_categorical_accuracy: 0.8893\n",
      "Epoch 132/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.4238 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2322 - categorical_accuracy: 0.9329 - val_loss: 0.4238 - val_categorical_accuracy: 0.9000\n",
      "Epoch 133/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4634 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2230 - categorical_accuracy: 0.9346 - val_loss: 0.4634 - val_categorical_accuracy: 0.8893\n",
      "Epoch 134/400\n",
      "18/18 [==============================] - 1s 69ms/step - loss: 0.5027 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2148 - categorical_accuracy: 0.9346 - val_loss: 0.5027 - val_categorical_accuracy: 0.8893\n",
      "Epoch 135/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4289 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2432 - categorical_accuracy: 0.9268 - val_loss: 0.4289 - val_categorical_accuracy: 0.8786\n",
      "Epoch 136/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4359 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2103 - categorical_accuracy: 0.9357 - val_loss: 0.4359 - val_categorical_accuracy: 0.8893\n",
      "Epoch 137/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4606 - categorical_accuracy: 0.9071A: 0s - loss:\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2087 - categorical_accuracy: 0.9364 - val_loss: 0.4606 - val_categorical_accuracy: 0.9071\n",
      "Epoch 138/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4516 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2195 - categorical_accuracy: 0.9364 - val_loss: 0.4516 - val_categorical_accuracy: 0.8893\n",
      "Epoch 139/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4870 - categorical_accuracy: 0.8607\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2797 - categorical_accuracy: 0.9236 - val_loss: 0.4870 - val_categorical_accuracy: 0.8607\n",
      "Epoch 140/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.3544 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 35s 200ms/step - loss: 0.2223 - categorical_accuracy: 0.9346 - val_loss: 0.3544 - val_categorical_accuracy: 0.9000\n",
      "Epoch 141/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.4115 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.2099 - categorical_accuracy: 0.9382 - val_loss: 0.4115 - val_categorical_accuracy: 0.8893\n",
      "Epoch 142/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.3824 - categorical_accuracy: 0.8893A: 2s - loss: 0.2108 - categorical_accuracy - E\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.2156 - categorical_accuracy: 0.9346 - val_loss: 0.3824 - val_categorical_accuracy: 0.8893\n",
      "Epoch 143/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.5694 - categorical_accuracy: 0.8571\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1902 - categorical_accuracy: 0.9443 - val_loss: 0.5694 - val_categorical_accuracy: 0.8571\n",
      "Epoch 144/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 0.4715 - categorical_accuracy: 0.8679\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.2126 - categorical_accuracy: 0.9386 - val_loss: 0.4715 - val_categorical_accuracy: 0.8679\n",
      "Epoch 145/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.4178 - categorical_accuracy: 0.8750A: 2s - loss: 0.2312 - categorica - ETA: 0s - los\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.2282 - categorical_accuracy: 0.9400 - val_loss: 0.4178 - val_categorical_accuracy: 0.8750\n",
      "Epoch 146/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.5753 - categorical_accuracy: 0.8464\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.2265 - categorical_accuracy: 0.9289 - val_loss: 0.5753 - val_categorical_accuracy: 0.8464\n",
      "Epoch 147/400\n",
      "18/18 [==============================] - 1s 72ms/step - loss: 0.4000 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1994 - categorical_accuracy: 0.9418 - val_loss: 0.4000 - val_categorical_accuracy: 0.8786\n",
      "Epoch 148/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.5190 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.2027 - categorical_accuracy: 0.9361 - val_loss: 0.5190 - val_categorical_accuracy: 0.8750\n",
      "Epoch 149/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.4076 - categorical_accuracy: 0.8857\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.2051 - categorical_accuracy: 0.9336 - val_loss: 0.4076 - val_categorical_accuracy: 0.8857\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 150/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.3767 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.1998 - categorical_accuracy: 0.9425 - val_loss: 0.3767 - val_categorical_accuracy: 0.8964\n",
      "Epoch 151/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.4366 - categorical_accuracy: 0.8643\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.2105 - categorical_accuracy: 0.9386 - val_loss: 0.4366 - val_categorical_accuracy: 0.8643\n",
      "Epoch 152/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.4353 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.1944 - categorical_accuracy: 0.9443 - val_loss: 0.4353 - val_categorical_accuracy: 0.8750\n",
      "Epoch 153/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.3769 - categorical_accuracy: 0.8964A: 3s - loss: 0.1931 - categorical_accura\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.1973 - categorical_accuracy: 0.9425 - val_loss: 0.3769 - val_categorical_accuracy: 0.8964\n",
      "Epoch 154/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.4113 - categorical_accuracy: 0.8857ETA: 16s - loss: 0.1905 - categoric - ETA: 12s - loss: 0.1916 - categorical_accur - ETA:  - ETA: 5s - loss:\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.1908 - categorical_accuracy: 0.9454 - val_loss: 0.4113 - val_categorical_accuracy: 0.8857\n",
      "Epoch 155/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.3934 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.1873 - categorical_accuracy: 0.9471 - val_loss: 0.3934 - val_categorical_accuracy: 0.9071\n",
      "Epoch 156/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 0.4614 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1817 - categorical_accuracy: 0.9475 - val_loss: 0.4614 - val_categorical_accuracy: 0.8750\n",
      "Epoch 157/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.4678 - categorical_accuracy: 0.8679\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.2076 - categorical_accuracy: 0.9400 - val_loss: 0.4678 - val_categorical_accuracy: 0.8679\n",
      "Epoch 158/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4349 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1878 - categorical_accuracy: 0.9439 - val_loss: 0.4349 - val_categorical_accuracy: 0.9036\n",
      "Epoch 159/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4129 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2038 - categorical_accuracy: 0.9443 - val_loss: 0.4129 - val_categorical_accuracy: 0.8750\n",
      "Epoch 160/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4547 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1811 - categorical_accuracy: 0.9443 - val_loss: 0.4547 - val_categorical_accuracy: 0.8964\n",
      "Epoch 161/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4099 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.1700 - categorical_accuracy: 0.9521 - val_loss: 0.4099 - val_categorical_accuracy: 0.8929\n",
      "Epoch 162/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5127 - categorical_accuracy: 0.8714\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1791 - categorical_accuracy: 0.9464 - val_loss: 0.5127 - val_categorical_accuracy: 0.8714\n",
      "Epoch 163/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4281 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.2002 - categorical_accuracy: 0.9386 - val_loss: 0.4281 - val_categorical_accuracy: 0.8893\n",
      "Epoch 164/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4348 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1795 - categorical_accuracy: 0.9471 - val_loss: 0.4348 - val_categorical_accuracy: 0.9071\n",
      "Epoch 165/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4404 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1844 - categorical_accuracy: 0.9482 - val_loss: 0.4404 - val_categorical_accuracy: 0.8821\n",
      "Epoch 166/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4767 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1626 - categorical_accuracy: 0.9539 - val_loss: 0.4767 - val_categorical_accuracy: 0.8929\n",
      "Epoch 167/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4827 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1702 - categorical_accuracy: 0.9543 - val_loss: 0.4827 - val_categorical_accuracy: 0.8929\n",
      "Epoch 168/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.5736 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1637 - categorical_accuracy: 0.9525 - val_loss: 0.5736 - val_categorical_accuracy: 0.8750\n",
      "Epoch 169/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5537 - categorical_accuracy: 0.8500\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1949 - categorical_accuracy: 0.9457 - val_loss: 0.5537 - val_categorical_accuracy: 0.8500\n",
      "Epoch 170/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4852 - categorical_accuracy: 0.8857\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1982 - categorical_accuracy: 0.9411 - val_loss: 0.4852 - val_categorical_accuracy: 0.8857\n",
      "Epoch 171/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.6205 - categorical_accuracy: 0.8536\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1756 - categorical_accuracy: 0.9493 - val_loss: 0.6205 - val_categorical_accuracy: 0.8536\n",
      "Epoch 172/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4706 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1904 - categorical_accuracy: 0.9471 - val_loss: 0.4706 - val_categorical_accuracy: 0.9107\n",
      "Epoch 173/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4743 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1632 - categorical_accuracy: 0.9561 - val_loss: 0.4743 - val_categorical_accuracy: 0.9179\n",
      "Epoch 174/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5349 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1753 - categorical_accuracy: 0.9489 - val_loss: 0.5349 - val_categorical_accuracy: 0.8821\n",
      "Epoch 175/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4949 - categorical_accuracy: 0.8714\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1701 - categorical_accuracy: 0.9514 - val_loss: 0.4949 - val_categorical_accuracy: 0.8714\n",
      "Epoch 176/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4305 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1680 - categorical_accuracy: 0.9475 - val_loss: 0.4305 - val_categorical_accuracy: 0.8821\n",
      "Epoch 177/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.4414 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1784 - categorical_accuracy: 0.9489 - val_loss: 0.4414 - val_categorical_accuracy: 0.8929\n",
      "Epoch 178/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 0.4939 - categorical_accuracy: 0.8714\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1743 - categorical_accuracy: 0.9496 - val_loss: 0.4939 - val_categorical_accuracy: 0.8714\n",
      "Epoch 179/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18/18 [==============================] - 1s 65ms/step - loss: 0.5218 - categorical_accuracy: 0.8821 0s - loss: 0.5554 - categorical_accuracy: \n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.1386 - categorical_accuracy: 0.9579 - val_loss: 0.5218 - val_categorical_accuracy: 0.8821\n",
      "Epoch 180/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.5051 - categorical_accuracy: 0.8714\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.1752 - categorical_accuracy: 0.9539 - val_loss: 0.5051 - val_categorical_accuracy: 0.8714\n",
      "Epoch 181/400\n",
      "18/18 [==============================] - 1s 69ms/step - loss: 0.5896 - categorical_accuracy: 0.8714\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1622 - categorical_accuracy: 0.9539 - val_loss: 0.5896 - val_categorical_accuracy: 0.8714\n",
      "Epoch 182/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4851 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1952 - categorical_accuracy: 0.9461 - val_loss: 0.4851 - val_categorical_accuracy: 0.9179\n",
      "Epoch 183/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4390 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.1581 - categorical_accuracy: 0.9582 - val_loss: 0.4390 - val_categorical_accuracy: 0.8786\n",
      "Epoch 184/400\n",
      "18/18 [==============================] - 1s 69ms/step - loss: 0.6511 - categorical_accuracy: 0.8714\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.1710 - categorical_accuracy: 0.9507 - val_loss: 0.6511 - val_categorical_accuracy: 0.8714\n",
      "Epoch 185/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.6981 - categorical_accuracy: 0.8321\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1551 - categorical_accuracy: 0.9589 - val_loss: 0.6981 - val_categorical_accuracy: 0.8321\n",
      "Epoch 186/400\n",
      "18/18 [==============================] - 1s 69ms/step - loss: 0.5117 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1719 - categorical_accuracy: 0.9514 - val_loss: 0.5117 - val_categorical_accuracy: 0.9107\n",
      "Epoch 187/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4697 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1530 - categorical_accuracy: 0.9536 - val_loss: 0.4697 - val_categorical_accuracy: 0.8893\n",
      "Epoch 188/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3951 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1736 - categorical_accuracy: 0.9550 - val_loss: 0.3951 - val_categorical_accuracy: 0.9036\n",
      "Epoch 189/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.7648 - categorical_accuracy: 0.8321\n",
      "175/175 [==============================] - 35s 199ms/step - loss: 0.1517 - categorical_accuracy: 0.9568 - val_loss: 0.7648 - val_categorical_accuracy: 0.8321\n",
      "Epoch 190/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.5322 - categorical_accuracy: 0.8857\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1598 - categorical_accuracy: 0.9511 - val_loss: 0.5322 - val_categorical_accuracy: 0.8857\n",
      "Epoch 191/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.5020 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1531 - categorical_accuracy: 0.9586 - val_loss: 0.5020 - val_categorical_accuracy: 0.9000\n",
      "Epoch 192/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4476 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1621 - categorical_accuracy: 0.9561 - val_loss: 0.4476 - val_categorical_accuracy: 0.8893\n",
      "Epoch 193/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.6548 - categorical_accuracy: 0.8571\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1378 - categorical_accuracy: 0.9614 - val_loss: 0.6548 - val_categorical_accuracy: 0.8571\n",
      "Epoch 194/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4928 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1841 - categorical_accuracy: 0.9507 - val_loss: 0.4928 - val_categorical_accuracy: 0.8821\n",
      "Epoch 195/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4929 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1499 - categorical_accuracy: 0.9575 - val_loss: 0.4929 - val_categorical_accuracy: 0.8964\n",
      "Epoch 196/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4094 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1250 - categorical_accuracy: 0.9657 - val_loss: 0.4094 - val_categorical_accuracy: 0.9000\n",
      "Epoch 197/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.5316 - categorical_accuracy: 0.8571\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1649 - categorical_accuracy: 0.9589 - val_loss: 0.5316 - val_categorical_accuracy: 0.8571\n",
      "Epoch 198/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4720 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1630 - categorical_accuracy: 0.9550 - val_loss: 0.4720 - val_categorical_accuracy: 0.9071\n",
      "Epoch 199/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.3660 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1457 - categorical_accuracy: 0.9579 - val_loss: 0.3660 - val_categorical_accuracy: 0.9036\n",
      "Epoch 200/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4718 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1473 - categorical_accuracy: 0.9550 - val_loss: 0.4718 - val_categorical_accuracy: 0.9000\n",
      "Epoch 201/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4862 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1198 - categorical_accuracy: 0.9686 - val_loss: 0.4862 - val_categorical_accuracy: 0.8964\n",
      "Epoch 202/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4233 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1369 - categorical_accuracy: 0.9607 - val_loss: 0.4233 - val_categorical_accuracy: 0.9071\n",
      "Epoch 203/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4450 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1446 - categorical_accuracy: 0.9632 - val_loss: 0.4450 - val_categorical_accuracy: 0.8893\n",
      "Epoch 204/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4039 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1919 - categorical_accuracy: 0.9521 - val_loss: 0.4039 - val_categorical_accuracy: 0.9179\n",
      "Epoch 205/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4548 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1637 - categorical_accuracy: 0.9564 - val_loss: 0.4548 - val_categorical_accuracy: 0.9000\n",
      "Epoch 206/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.3805 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1455 - categorical_accuracy: 0.9632 - val_loss: 0.3805 - val_categorical_accuracy: 0.9143\n",
      "Epoch 207/400\n",
      "18/18 [==============================] - 1s 58ms/step - loss: 0.4207 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1429 - categorical_accuracy: 0.9543 - val_loss: 0.4207 - val_categorical_accuracy: 0.9107\n",
      "Epoch 208/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.5706 - categorical_accuracy: 0.8571\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1451 - categorical_accuracy: 0.9596 - val_loss: 0.5706 - val_categorical_accuracy: 0.8571\n",
      "Epoch 209/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18/18 [==============================] - 1s 58ms/step - loss: 0.4019 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1643 - categorical_accuracy: 0.9525 - val_loss: 0.4019 - val_categorical_accuracy: 0.9143\n",
      "Epoch 210/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.3719 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1351 - categorical_accuracy: 0.9618 - val_loss: 0.3719 - val_categorical_accuracy: 0.9107\n",
      "Epoch 211/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.4297 - categorical_accuracy: 0.8929 0s - loss: 0.4760 - catego\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1274 - categorical_accuracy: 0.9657 - val_loss: 0.4297 - val_categorical_accuracy: 0.8929\n",
      "Epoch 212/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.3706 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1583 - categorical_accuracy: 0.9593 - val_loss: 0.3706 - val_categorical_accuracy: 0.9143\n",
      "Epoch 213/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.5588 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1064 - categorical_accuracy: 0.9718 - val_loss: 0.5588 - val_categorical_accuracy: 0.8893\n",
      "Epoch 214/400\n",
      "18/18 [==============================] - 1s 59ms/step - loss: 0.5086 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1481 - categorical_accuracy: 0.9639 - val_loss: 0.5086 - val_categorical_accuracy: 0.8893\n",
      "Epoch 215/400\n",
      "18/18 [==============================] - 1s 58ms/step - loss: 0.4469 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1347 - categorical_accuracy: 0.9632 - val_loss: 0.4469 - val_categorical_accuracy: 0.8929\n",
      "Epoch 216/400\n",
      "18/18 [==============================] - 1s 58ms/step - loss: 0.4526 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 32s 182ms/step - loss: 0.1416 - categorical_accuracy: 0.9614 - val_loss: 0.4526 - val_categorical_accuracy: 0.8786\n",
      "Epoch 217/400\n",
      "18/18 [==============================] - 1s 71ms/step - loss: 0.4081 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 33s 186ms/step - loss: 0.1356 - categorical_accuracy: 0.9636 - val_loss: 0.4081 - val_categorical_accuracy: 0.9071\n",
      "Epoch 218/400\n",
      "18/18 [==============================] - 1s 71ms/step - loss: 0.5153 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 39s 225ms/step - loss: 0.1394 - categorical_accuracy: 0.9636 - val_loss: 0.5153 - val_categorical_accuracy: 0.8893\n",
      "Epoch 219/400\n",
      "18/18 [==============================] - 1s 71ms/step - loss: 0.4396 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 39s 225ms/step - loss: 0.1267 - categorical_accuracy: 0.9654 - val_loss: 0.4396 - val_categorical_accuracy: 0.9071\n",
      "Epoch 220/400\n",
      "18/18 [==============================] - 1s 71ms/step - loss: 0.4403 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 39s 225ms/step - loss: 0.1455 - categorical_accuracy: 0.9639 - val_loss: 0.4403 - val_categorical_accuracy: 0.8821\n",
      "Epoch 221/400\n",
      "18/18 [==============================] - 1s 71ms/step - loss: 0.4555 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 39s 225ms/step - loss: 0.1136 - categorical_accuracy: 0.9739 - val_loss: 0.4555 - val_categorical_accuracy: 0.8964\n",
      "Epoch 222/400\n",
      "18/18 [==============================] - 1s 75ms/step - loss: 0.4772 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 40s 226ms/step - loss: 0.1310 - categorical_accuracy: 0.9679 - val_loss: 0.4772 - val_categorical_accuracy: 0.9036\n",
      "Epoch 223/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.3871 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.1662 - categorical_accuracy: 0.9564 - val_loss: 0.3871 - val_categorical_accuracy: 0.9071\n",
      "Epoch 224/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.3947 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1180 - categorical_accuracy: 0.9679 - val_loss: 0.3947 - val_categorical_accuracy: 0.9036\n",
      "Epoch 225/400\n",
      "18/18 [==============================] - 1s 78ms/step - loss: 0.5991 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1245 - categorical_accuracy: 0.9689 - val_loss: 0.5991 - val_categorical_accuracy: 0.8821\n",
      "Epoch 226/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.5105 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1213 - categorical_accuracy: 0.9661 - val_loss: 0.5105 - val_categorical_accuracy: 0.8929\n",
      "Epoch 227/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.4848 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1314 - categorical_accuracy: 0.9671 - val_loss: 0.4848 - val_categorical_accuracy: 0.9036\n",
      "Epoch 228/400\n",
      "18/18 [==============================] - 1s 78ms/step - loss: 0.4464 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1205 - categorical_accuracy: 0.9689 - val_loss: 0.4464 - val_categorical_accuracy: 0.8929\n",
      "Epoch 229/400\n",
      "18/18 [==============================] - 1s 78ms/step - loss: 0.4676 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1521 - categorical_accuracy: 0.9614 - val_loss: 0.4676 - val_categorical_accuracy: 0.8750\n",
      "Epoch 230/400\n",
      "18/18 [==============================] - 1s 76ms/step - loss: 0.3649 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1197 - categorical_accuracy: 0.9650 - val_loss: 0.3649 - val_categorical_accuracy: 0.9107\n",
      "Epoch 231/400\n",
      "18/18 [==============================] - 1s 76ms/step - loss: 0.4656 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.1398 - categorical_accuracy: 0.9607 - val_loss: 0.4656 - val_categorical_accuracy: 0.8821\n",
      "Epoch 232/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.6119 - categorical_accuracy: 0.8571\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.1153 - categorical_accuracy: 0.9714 - val_loss: 0.6119 - val_categorical_accuracy: 0.8571\n",
      "Epoch 233/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.5321 - categorical_accuracy: 0.8643\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.1212 - categorical_accuracy: 0.9657 - val_loss: 0.5321 - val_categorical_accuracy: 0.8643\n",
      "Epoch 234/400\n",
      "18/18 [==============================] - 1s 78ms/step - loss: 0.6024 - categorical_accuracy: 0.8714\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1045 - categorical_accuracy: 0.9732 - val_loss: 0.6024 - val_categorical_accuracy: 0.8714\n",
      "Epoch 235/400\n",
      "18/18 [==============================] - 1s 76ms/step - loss: 0.5928 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1428 - categorical_accuracy: 0.9661 - val_loss: 0.5928 - val_categorical_accuracy: 0.8821\n",
      "Epoch 236/400\n",
      "18/18 [==============================] - 1s 76ms/step - loss: 0.4941 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.1355 - categorical_accuracy: 0.9646 - val_loss: 0.4941 - val_categorical_accuracy: 0.9036\n",
      "Epoch 237/400\n",
      "18/18 [==============================] - 1s 76ms/step - loss: 0.5126 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1269 - categorical_accuracy: 0.9682 - val_loss: 0.5126 - val_categorical_accuracy: 0.8893\n",
      "Epoch 238/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.3898 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 40s 228ms/step - loss: 0.1166 - categorical_accuracy: 0.9729 - val_loss: 0.3898 - val_categorical_accuracy: 0.9107\n",
      "Epoch 239/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18/18 [==============================] - 1s 75ms/step - loss: 0.5153 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1210 - categorical_accuracy: 0.9643 - val_loss: 0.5153 - val_categorical_accuracy: 0.9179\n",
      "Epoch 240/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.4231 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.1126 - categorical_accuracy: 0.9664 - val_loss: 0.4231 - val_categorical_accuracy: 0.9143\n",
      "Epoch 241/400\n",
      "18/18 [==============================] - 1s 78ms/step - loss: 0.5216 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.0928 - categorical_accuracy: 0.9775 - val_loss: 0.5216 - val_categorical_accuracy: 0.8964\n",
      "Epoch 242/400\n",
      "18/18 [==============================] - 1s 78ms/step - loss: 0.5323 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1304 - categorical_accuracy: 0.9639 - val_loss: 0.5323 - val_categorical_accuracy: 0.8929\n",
      "Epoch 243/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.5700 - categorical_accuracy: 0.8464\n",
      "175/175 [==============================] - 40s 230ms/step - loss: 0.1209 - categorical_accuracy: 0.9696 - val_loss: 0.5700 - val_categorical_accuracy: 0.8464\n",
      "Epoch 244/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.3981 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.1183 - categorical_accuracy: 0.9693 - val_loss: 0.3981 - val_categorical_accuracy: 0.9179\n",
      "Epoch 245/400\n",
      "18/18 [==============================] - 1s 76ms/step - loss: 0.3895 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.1404 - categorical_accuracy: 0.9611 - val_loss: 0.3895 - val_categorical_accuracy: 0.9071\n",
      "Epoch 246/400\n",
      "18/18 [==============================] - 1s 76ms/step - loss: 0.5329 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.0797 - categorical_accuracy: 0.9804 - val_loss: 0.5329 - val_categorical_accuracy: 0.9107\n",
      "Epoch 247/400\n",
      "18/18 [==============================] - 1s 78ms/step - loss: 0.4248 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.1531 - categorical_accuracy: 0.9543 - val_loss: 0.4248 - val_categorical_accuracy: 0.9000\n",
      "Epoch 248/400\n",
      "18/18 [==============================] - 1s 71ms/step - loss: 0.4952 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 40s 229ms/step - loss: 0.1030 - categorical_accuracy: 0.9732 - val_loss: 0.4952 - val_categorical_accuracy: 0.9071\n",
      "Epoch 249/400\n",
      "18/18 [==============================] - 1s 71ms/step - loss: 0.4665 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 39s 225ms/step - loss: 0.1280 - categorical_accuracy: 0.9718 - val_loss: 0.4665 - val_categorical_accuracy: 0.9036\n",
      "Epoch 250/400\n",
      "18/18 [==============================] - 1s 71ms/step - loss: 0.4485 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 39s 225ms/step - loss: 0.1233 - categorical_accuracy: 0.9714 - val_loss: 0.4485 - val_categorical_accuracy: 0.8964\n",
      "Epoch 251/400\n",
      "18/18 [==============================] - 1s 79ms/step - loss: 0.4482 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 40s 226ms/step - loss: 0.1048 - categorical_accuracy: 0.9743 - val_loss: 0.4482 - val_categorical_accuracy: 0.8964\n",
      "Epoch 252/400\n",
      "18/18 [==============================] - 1s 79ms/step - loss: 0.4933 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 41s 232ms/step - loss: 0.1225 - categorical_accuracy: 0.9693 - val_loss: 0.4933 - val_categorical_accuracy: 0.9071\n",
      "Epoch 253/400\n",
      "18/18 [==============================] - 1s 79ms/step - loss: 0.4077 - categorical_accuracy: 0.9286\n",
      "175/175 [==============================] - 41s 232ms/step - loss: 0.1060 - categorical_accuracy: 0.9721 - val_loss: 0.4077 - val_categorical_accuracy: 0.9286\n",
      "Epoch 254/400\n",
      "18/18 [==============================] - 1s 77ms/step - loss: 0.4428 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 41s 232ms/step - loss: 0.1215 - categorical_accuracy: 0.9711 - val_loss: 0.4428 - val_categorical_accuracy: 0.9071\n",
      "Epoch 255/400\n",
      "18/18 [==============================] - 1s 78ms/step - loss: 0.5249 - categorical_accuracy: 0.8857\n",
      "175/175 [==============================] - 41s 232ms/step - loss: 0.1510 - categorical_accuracy: 0.9625 - val_loss: 0.5249 - val_categorical_accuracy: 0.8857\n",
      "Epoch 256/400\n",
      "18/18 [==============================] - 1s 78ms/step - loss: 0.3290 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 44s 253ms/step - loss: 0.1294 - categorical_accuracy: 0.9636 - val_loss: 0.3290 - val_categorical_accuracy: 0.9143\n",
      "Epoch 257/400\n",
      "18/18 [==============================] - 2s 133ms/step - loss: 0.3975 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 163241s 933s/step - loss: 0.0959 - categorical_accuracy: 0.9764 - val_loss: 0.3975 - val_categorical_accuracy: 0.9214\n",
      "Epoch 258/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.4840 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 33s 190ms/step - loss: 0.0987 - categorical_accuracy: 0.9761 - val_loss: 0.4840 - val_categorical_accuracy: 0.8893\n",
      "Epoch 259/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.5147 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.1021 - categorical_accuracy: 0.9725 - val_loss: 0.5147 - val_categorical_accuracy: 0.8929\n",
      "Epoch 260/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6992 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.0947 - categorical_accuracy: 0.9750 - val_loss: 0.6992 - val_categorical_accuracy: 0.8821\n",
      "Epoch 261/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.4010 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.1140 - categorical_accuracy: 0.9739 - val_loss: 0.4010 - val_categorical_accuracy: 0.8821\n",
      "Epoch 262/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.4644 - categorical_accuracy: 0.9000A: 0s - l\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1507 - categorical_accuracy: 0.9625 - val_loss: 0.4644 - val_categorical_accuracy: 0.9000\n",
      "Epoch 263/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 0.4456 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1248 - categorical_accuracy: 0.9711 - val_loss: 0.4456 - val_categorical_accuracy: 0.9000\n",
      "Epoch 264/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.4732 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1096 - categorical_accuracy: 0.9707 - val_loss: 0.4732 - val_categorical_accuracy: 0.8964\n",
      "Epoch 265/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.3921 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.1155 - categorical_accuracy: 0.9693 - val_loss: 0.3921 - val_categorical_accuracy: 0.9107\n",
      "Epoch 266/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4960 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0944 - categorical_accuracy: 0.9729 - val_loss: 0.4960 - val_categorical_accuracy: 0.8964\n",
      "Epoch 267/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4950 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0955 - categorical_accuracy: 0.9729 - val_loss: 0.4950 - val_categorical_accuracy: 0.9000\n",
      "Epoch 268/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5114 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0780 - categorical_accuracy: 0.9818 - val_loss: 0.5114 - val_categorical_accuracy: 0.9036\n",
      "Epoch 269/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4537 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1033 - categorical_accuracy: 0.9779 - val_loss: 0.4537 - val_categorical_accuracy: 0.8964\n",
      "Epoch 270/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.7595 - categorical_accuracy: 0.8500\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0970 - categorical_accuracy: 0.9714 - val_loss: 0.7595 - val_categorical_accuracy: 0.8500\n",
      "Epoch 271/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4059 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1100 - categorical_accuracy: 0.9704 - val_loss: 0.4059 - val_categorical_accuracy: 0.9143\n",
      "Epoch 272/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4929 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1260 - categorical_accuracy: 0.9679 - val_loss: 0.4929 - val_categorical_accuracy: 0.8929\n",
      "Epoch 273/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4557 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1070 - categorical_accuracy: 0.9679 - val_loss: 0.4557 - val_categorical_accuracy: 0.9179\n",
      "Epoch 274/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4473 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0998 - categorical_accuracy: 0.9754 - val_loss: 0.4473 - val_categorical_accuracy: 0.9143\n",
      "Epoch 275/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4284 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1137 - categorical_accuracy: 0.9689 - val_loss: 0.4284 - val_categorical_accuracy: 0.9000\n",
      "Epoch 276/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5779 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0934 - categorical_accuracy: 0.9761 - val_loss: 0.5779 - val_categorical_accuracy: 0.8821\n",
      "Epoch 277/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4073 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1235 - categorical_accuracy: 0.9657 - val_loss: 0.4073 - val_categorical_accuracy: 0.9107\n",
      "Epoch 278/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4301 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0966 - categorical_accuracy: 0.9729 - val_loss: 0.4301 - val_categorical_accuracy: 0.8964\n",
      "Epoch 279/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.3845 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0860 - categorical_accuracy: 0.9786 - val_loss: 0.3845 - val_categorical_accuracy: 0.8929\n",
      "Epoch 280/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4804 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1062 - categorical_accuracy: 0.9696 - val_loss: 0.4804 - val_categorical_accuracy: 0.8893\n",
      "Epoch 281/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3533 - categorical_accuracy: 0.9250\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1085 - categorical_accuracy: 0.9725 - val_loss: 0.3533 - val_categorical_accuracy: 0.9250\n",
      "Epoch 282/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4065 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0999 - categorical_accuracy: 0.9718 - val_loss: 0.4065 - val_categorical_accuracy: 0.9036\n",
      "Epoch 283/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5266 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0872 - categorical_accuracy: 0.9807 - val_loss: 0.5266 - val_categorical_accuracy: 0.9143\n",
      "Epoch 284/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5490 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1019 - categorical_accuracy: 0.9725 - val_loss: 0.5490 - val_categorical_accuracy: 0.9036\n",
      "Epoch 285/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4849 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0940 - categorical_accuracy: 0.9796 - val_loss: 0.4849 - val_categorical_accuracy: 0.8964\n",
      "Epoch 286/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3872 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1044 - categorical_accuracy: 0.9725 - val_loss: 0.3872 - val_categorical_accuracy: 0.9143\n",
      "Epoch 287/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5988 - categorical_accuracy: 0.8714 0s - loss: 0.7164 - categorical_accura\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0928 - categorical_accuracy: 0.9786 - val_loss: 0.5988 - val_categorical_accuracy: 0.8714\n",
      "Epoch 288/400\n",
      "18/18 [==============================] - 1s 69ms/step - loss: 0.4374 - categorical_accuracy: 0.9071A: 0s - los\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1018 - categorical_accuracy: 0.9764 - val_loss: 0.4374 - val_categorical_accuracy: 0.9071\n",
      "Epoch 289/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.4414 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1206 - categorical_accuracy: 0.9696 - val_loss: 0.4414 - val_categorical_accuracy: 0.9071\n",
      "Epoch 290/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.4977 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.0934 - categorical_accuracy: 0.9754 - val_loss: 0.4977 - val_categorical_accuracy: 0.8821\n",
      "Epoch 291/400\n",
      "18/18 [==============================] - 1s 61ms/step - loss: 0.4061 - categorical_accuracy: 0.9250\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1004 - categorical_accuracy: 0.9754 - val_loss: 0.4061 - val_categorical_accuracy: 0.9250\n",
      "Epoch 292/400\n",
      "18/18 [==============================] - 1s 63ms/step - loss: 0.4306 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1089 - categorical_accuracy: 0.9714 - val_loss: 0.4306 - val_categorical_accuracy: 0.9179\n",
      "Epoch 293/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.4123 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1025 - categorical_accuracy: 0.9729 - val_loss: 0.4123 - val_categorical_accuracy: 0.9071\n",
      "Epoch 294/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.6136 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.0918 - categorical_accuracy: 0.9793 - val_loss: 0.6136 - val_categorical_accuracy: 0.8821\n",
      "Epoch 295/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.4189 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1084 - categorical_accuracy: 0.9743 - val_loss: 0.4189 - val_categorical_accuracy: 0.9107\n",
      "Epoch 296/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.4462 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.0981 - categorical_accuracy: 0.9729 - val_loss: 0.4462 - val_categorical_accuracy: 0.9107\n",
      "Epoch 297/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4677 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1289 - categorical_accuracy: 0.9686 - val_loss: 0.4677 - val_categorical_accuracy: 0.9143\n",
      "Epoch 298/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4304 - categorical_accuracy: 0.9286\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0947 - categorical_accuracy: 0.9750 - val_loss: 0.4304 - val_categorical_accuracy: 0.9286\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 299/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4597 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0713 - categorical_accuracy: 0.9829 - val_loss: 0.4597 - val_categorical_accuracy: 0.9107\n",
      "Epoch 300/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5004 - categorical_accuracy: 0.9179A: 0s - loss: 0.0\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0769 - categorical_accuracy: 0.9821 - val_loss: 0.5004 - val_categorical_accuracy: 0.9179\n",
      "Epoch 301/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5379 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0938 - categorical_accuracy: 0.9789 - val_loss: 0.5379 - val_categorical_accuracy: 0.8821\n",
      "Epoch 302/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4222 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1195 - categorical_accuracy: 0.9707 - val_loss: 0.4222 - val_categorical_accuracy: 0.9179\n",
      "Epoch 303/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4929 - categorical_accuracy: 0.9250\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0712 - categorical_accuracy: 0.9829 - val_loss: 0.4929 - val_categorical_accuracy: 0.9250\n",
      "Epoch 304/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4640 - categorical_accuracy: 0.9286\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0669 - categorical_accuracy: 0.9846 - val_loss: 0.4640 - val_categorical_accuracy: 0.9286\n",
      "Epoch 305/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5066 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0925 - categorical_accuracy: 0.9736 - val_loss: 0.5066 - val_categorical_accuracy: 0.9000\n",
      "Epoch 306/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4929 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0792 - categorical_accuracy: 0.9818 - val_loss: 0.4929 - val_categorical_accuracy: 0.9179\n",
      "Epoch 307/400\n",
      "18/18 [==============================] - 1s 71ms/step - loss: 0.4878 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1097 - categorical_accuracy: 0.9739 - val_loss: 0.4878 - val_categorical_accuracy: 0.8821\n",
      "Epoch 308/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5304 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0810 - categorical_accuracy: 0.9796 - val_loss: 0.5304 - val_categorical_accuracy: 0.8893\n",
      "Epoch 309/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.6432 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.1136 - categorical_accuracy: 0.9721 - val_loss: 0.6432 - val_categorical_accuracy: 0.8929\n",
      "Epoch 310/400\n",
      "18/18 [==============================] - 1s 82ms/step - loss: 0.4332 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 36s 204ms/step - loss: 0.0715 - categorical_accuracy: 0.9839 - val_loss: 0.4332 - val_categorical_accuracy: 0.9214\n",
      "Epoch 311/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.4807 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 33s 191ms/step - loss: 0.0795 - categorical_accuracy: 0.9807 - val_loss: 0.4807 - val_categorical_accuracy: 0.9071\n",
      "Epoch 312/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.4376 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.0930 - categorical_accuracy: 0.9754 - val_loss: 0.4376 - val_categorical_accuracy: 0.9179\n",
      "Epoch 313/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.6061 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0914 - categorical_accuracy: 0.9789 - val_loss: 0.6061 - val_categorical_accuracy: 0.9000\n",
      "Epoch 314/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5479 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1008 - categorical_accuracy: 0.9743 - val_loss: 0.5479 - val_categorical_accuracy: 0.8750\n",
      "Epoch 315/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5602 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0744 - categorical_accuracy: 0.9818 - val_loss: 0.5602 - val_categorical_accuracy: 0.8786\n",
      "Epoch 316/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.6768 - categorical_accuracy: 0.8857\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0887 - categorical_accuracy: 0.9804 - val_loss: 0.6768 - val_categorical_accuracy: 0.8857\n",
      "Epoch 317/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5410 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0951 - categorical_accuracy: 0.9779 - val_loss: 0.5410 - val_categorical_accuracy: 0.9036\n",
      "Epoch 318/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3605 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1042 - categorical_accuracy: 0.9750 - val_loss: 0.3605 - val_categorical_accuracy: 0.9179\n",
      "Epoch 319/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4136 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0955 - categorical_accuracy: 0.9779 - val_loss: 0.4136 - val_categorical_accuracy: 0.9143\n",
      "Epoch 320/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5586 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0790 - categorical_accuracy: 0.9818 - val_loss: 0.5586 - val_categorical_accuracy: 0.8964\n",
      "Epoch 321/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4596 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0834 - categorical_accuracy: 0.9779 - val_loss: 0.4596 - val_categorical_accuracy: 0.9143\n",
      "Epoch 322/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5079 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0912 - categorical_accuracy: 0.9775 - val_loss: 0.5079 - val_categorical_accuracy: 0.8821\n",
      "Epoch 323/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3804 - categorical_accuracy: 0.9250\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0664 - categorical_accuracy: 0.9864 - val_loss: 0.3804 - val_categorical_accuracy: 0.9250\n",
      "Epoch 324/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5826 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1077 - categorical_accuracy: 0.9739 - val_loss: 0.5826 - val_categorical_accuracy: 0.8786\n",
      "Epoch 325/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4257 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0923 - categorical_accuracy: 0.9729 - val_loss: 0.4257 - val_categorical_accuracy: 0.9179\n",
      "Epoch 326/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5058 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0746 - categorical_accuracy: 0.9807 - val_loss: 0.5058 - val_categorical_accuracy: 0.9000\n",
      "Epoch 327/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4696 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1086 - categorical_accuracy: 0.9707 - val_loss: 0.4696 - val_categorical_accuracy: 0.8964\n",
      "Epoch 328/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.5483 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1040 - categorical_accuracy: 0.9750 - val_loss: 0.5483 - val_categorical_accuracy: 0.9000\n",
      "Epoch 329/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18/18 [==============================] - 1s 69ms/step - loss: 0.4179 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0834 - categorical_accuracy: 0.9811 - val_loss: 0.4179 - val_categorical_accuracy: 0.9214\n",
      "Epoch 330/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.3969 - categorical_accuracy: 0.9464\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0624 - categorical_accuracy: 0.9886 - val_loss: 0.3969 - val_categorical_accuracy: 0.9464\n",
      "Epoch 331/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4592 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.1009 - categorical_accuracy: 0.9764 - val_loss: 0.4592 - val_categorical_accuracy: 0.9000\n",
      "Epoch 332/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4330 - categorical_accuracy: 0.9250\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0795 - categorical_accuracy: 0.9786 - val_loss: 0.4330 - val_categorical_accuracy: 0.9250\n",
      "Epoch 333/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4727 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0731 - categorical_accuracy: 0.9825 - val_loss: 0.4727 - val_categorical_accuracy: 0.9000\n",
      "Epoch 334/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4489 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1151 - categorical_accuracy: 0.9693 - val_loss: 0.4489 - val_categorical_accuracy: 0.9000\n",
      "Epoch 335/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4651 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0656 - categorical_accuracy: 0.9839 - val_loss: 0.4651 - val_categorical_accuracy: 0.9071\n",
      "Epoch 336/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4090 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0832 - categorical_accuracy: 0.9793 - val_loss: 0.4090 - val_categorical_accuracy: 0.9143\n",
      "Epoch 337/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5419 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0875 - categorical_accuracy: 0.9775 - val_loss: 0.5419 - val_categorical_accuracy: 0.9071\n",
      "Epoch 338/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3821 - categorical_accuracy: 0.9321\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.1197 - categorical_accuracy: 0.9686 - val_loss: 0.3821 - val_categorical_accuracy: 0.9321\n",
      "Epoch 339/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.3819 - categorical_accuracy: 0.9250\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1119 - categorical_accuracy: 0.9729 - val_loss: 0.3819 - val_categorical_accuracy: 0.9250\n",
      "Epoch 340/400\n",
      "18/18 [==============================] - 1s 69ms/step - loss: 0.4460 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0693 - categorical_accuracy: 0.9854 - val_loss: 0.4460 - val_categorical_accuracy: 0.9214\n",
      "Epoch 341/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.6074 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0903 - categorical_accuracy: 0.9779 - val_loss: 0.6074 - val_categorical_accuracy: 0.8893\n",
      "Epoch 342/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.4515 - categorical_accuracy: 0.9143A: 3s - loss: 0.1098 - categorica\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1092 - categorical_accuracy: 0.9725 - val_loss: 0.4515 - val_categorical_accuracy: 0.9143\n",
      "Epoch 343/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4278 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1025 - categorical_accuracy: 0.9768 - val_loss: 0.4278 - val_categorical_accuracy: 0.9214\n",
      "Epoch 344/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4657 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0662 - categorical_accuracy: 0.9850 - val_loss: 0.4657 - val_categorical_accuracy: 0.9214\n",
      "Epoch 345/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.3662 - categorical_accuracy: 0.9357\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0928 - categorical_accuracy: 0.9789 - val_loss: 0.3662 - val_categorical_accuracy: 0.9357\n",
      "Epoch 346/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4617 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0639 - categorical_accuracy: 0.9871 - val_loss: 0.4617 - val_categorical_accuracy: 0.8893\n",
      "Epoch 347/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5815 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0659 - categorical_accuracy: 0.9846 - val_loss: 0.5815 - val_categorical_accuracy: 0.9036\n",
      "Epoch 348/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4946 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.0667 - categorical_accuracy: 0.9854 - val_loss: 0.4946 - val_categorical_accuracy: 0.9143\n",
      "Epoch 349/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5452 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.0960 - categorical_accuracy: 0.9754 - val_loss: 0.5452 - val_categorical_accuracy: 0.8750\n",
      "Epoch 350/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3897 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0864 - categorical_accuracy: 0.9775 - val_loss: 0.3897 - val_categorical_accuracy: 0.9214\n",
      "Epoch 351/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4336 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0741 - categorical_accuracy: 0.9829 - val_loss: 0.4336 - val_categorical_accuracy: 0.9179\n",
      "Epoch 352/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5408 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0646 - categorical_accuracy: 0.9839 - val_loss: 0.5408 - val_categorical_accuracy: 0.9071\n",
      "Epoch 353/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4136 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0674 - categorical_accuracy: 0.9843 - val_loss: 0.4136 - val_categorical_accuracy: 0.9214\n",
      "Epoch 354/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5673 - categorical_accuracy: 0.8857\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1036 - categorical_accuracy: 0.9746 - val_loss: 0.5673 - val_categorical_accuracy: 0.8857\n",
      "Epoch 355/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4636 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0826 - categorical_accuracy: 0.9775 - val_loss: 0.4636 - val_categorical_accuracy: 0.8964\n",
      "Epoch 356/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5319 - categorical_accuracy: 0.8857\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0673 - categorical_accuracy: 0.9850 - val_loss: 0.5319 - val_categorical_accuracy: 0.8857\n",
      "Epoch 357/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.6632 - categorical_accuracy: 0.8786\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0820 - categorical_accuracy: 0.9804 - val_loss: 0.6632 - val_categorical_accuracy: 0.8786\n",
      "Epoch 358/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.5281 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0890 - categorical_accuracy: 0.9764 - val_loss: 0.5281 - val_categorical_accuracy: 0.9214\n",
      "Epoch 359/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5130 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0695 - categorical_accuracy: 0.9818 - val_loss: 0.5130 - val_categorical_accuracy: 0.9179\n",
      "Epoch 360/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.7719 - categorical_accuracy: 0.8679\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0972 - categorical_accuracy: 0.9771 - val_loss: 0.7719 - val_categorical_accuracy: 0.8679\n",
      "Epoch 361/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.4300 - categorical_accuracy: 0.9214ETA: 12s - lo\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.1293 - categorical_accuracy: 0.9661 - val_loss: 0.4300 - val_categorical_accuracy: 0.9214\n",
      "Epoch 362/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4110 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0949 - categorical_accuracy: 0.9761 - val_loss: 0.4110 - val_categorical_accuracy: 0.9179\n",
      "Epoch 363/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.6660 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0688 - categorical_accuracy: 0.9843 - val_loss: 0.6660 - val_categorical_accuracy: 0.9000\n",
      "Epoch 364/400\n",
      "18/18 [==============================] - 1s 82ms/step - loss: 0.4766 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 35s 200ms/step - loss: 0.0705 - categorical_accuracy: 0.9871 - val_loss: 0.4766 - val_categorical_accuracy: 0.9214\n",
      "Epoch 365/400\n",
      "18/18 [==============================] - 1s 62ms/step - loss: 0.3361 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 192ms/step - loss: 0.1068 - categorical_accuracy: 0.9729 - val_loss: 0.3361 - val_categorical_accuracy: 0.9143\n",
      "Epoch 366/400\n",
      "18/18 [==============================] - 1s 64ms/step - loss: 0.2992 - categorical_accuracy: 0.9393\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0764 - categorical_accuracy: 0.9843 - val_loss: 0.2992 - val_categorical_accuracy: 0.9393\n",
      "Epoch 367/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4071 - categorical_accuracy: 0.9250\n",
      "175/175 [==============================] - 34s 193ms/step - loss: 0.0750 - categorical_accuracy: 0.9818 - val_loss: 0.4071 - val_categorical_accuracy: 0.9250\n",
      "Epoch 368/400\n",
      "18/18 [==============================] - 1s 69ms/step - loss: 0.3444 - categorical_accuracy: 0.9250\n",
      "175/175 [==============================] - 34s 194ms/step - loss: 0.0658 - categorical_accuracy: 0.9864 - val_loss: 0.3444 - val_categorical_accuracy: 0.9250\n",
      "Epoch 369/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.6320 - categorical_accuracy: 0.8821\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0934 - categorical_accuracy: 0.9743 - val_loss: 0.6320 - val_categorical_accuracy: 0.8821\n",
      "Epoch 370/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4727 - categorical_accuracy: 0.9107\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0819 - categorical_accuracy: 0.9796 - val_loss: 0.4727 - val_categorical_accuracy: 0.9107\n",
      "Epoch 371/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.4992 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0820 - categorical_accuracy: 0.9796 - val_loss: 0.4992 - val_categorical_accuracy: 0.9000\n",
      "Epoch 372/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4291 - categorical_accuracy: 0.9286A: 3s - loss: 0.0486 - catego - ETA: 1s -\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0482 - categorical_accuracy: 0.9907 - val_loss: 0.4291 - val_categorical_accuracy: 0.9286\n",
      "Epoch 373/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5689 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0750 - categorical_accuracy: 0.9796 - val_loss: 0.5689 - val_categorical_accuracy: 0.8929\n",
      "Epoch 374/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4211 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0782 - categorical_accuracy: 0.9836 - val_loss: 0.4211 - val_categorical_accuracy: 0.9179\n",
      "Epoch 375/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.6188 - categorical_accuracy: 0.8750\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0787 - categorical_accuracy: 0.9843 - val_loss: 0.6188 - val_categorical_accuracy: 0.8750\n",
      "Epoch 376/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4507 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0795 - categorical_accuracy: 0.9829 - val_loss: 0.4507 - val_categorical_accuracy: 0.9214\n",
      "Epoch 377/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4494 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0663 - categorical_accuracy: 0.9836 - val_loss: 0.4494 - val_categorical_accuracy: 0.8964\n",
      "Epoch 378/400\n",
      "18/18 [==============================] - 1s 65ms/step - loss: 0.4170 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0608 - categorical_accuracy: 0.9861 - val_loss: 0.4170 - val_categorical_accuracy: 0.9214\n",
      "Epoch 379/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4887 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0824 - categorical_accuracy: 0.9782 - val_loss: 0.4887 - val_categorical_accuracy: 0.9214\n",
      "Epoch 380/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.5033 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0656 - categorical_accuracy: 0.9882 - val_loss: 0.5033 - val_categorical_accuracy: 0.9071\n",
      "Epoch 381/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.7418 - categorical_accuracy: 0.8893\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0546 - categorical_accuracy: 0.9875 - val_loss: 0.7418 - val_categorical_accuracy: 0.8893\n",
      "Epoch 382/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4403 - categorical_accuracy: 0.9250\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1122 - categorical_accuracy: 0.9725 - val_loss: 0.4403 - val_categorical_accuracy: 0.9250\n",
      "Epoch 383/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4420 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0926 - categorical_accuracy: 0.9771 - val_loss: 0.4420 - val_categorical_accuracy: 0.9214\n",
      "Epoch 384/400\n",
      "18/18 [==============================] - 1s 69ms/step - loss: 0.5333 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0520 - categorical_accuracy: 0.9904 - val_loss: 0.5333 - val_categorical_accuracy: 0.9214\n",
      "Epoch 385/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4884 - categorical_accuracy: 0.8929\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0957 - categorical_accuracy: 0.9764 - val_loss: 0.4884 - val_categorical_accuracy: 0.8929\n",
      "Epoch 386/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.4163 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0753 - categorical_accuracy: 0.9832 - val_loss: 0.4163 - val_categorical_accuracy: 0.9179\n",
      "Epoch 387/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4710 - categorical_accuracy: 0.8857\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1502 - categorical_accuracy: 0.9632 - val_loss: 0.4710 - val_categorical_accuracy: 0.8857\n",
      "Epoch 388/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4285 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0812 - categorical_accuracy: 0.9775 - val_loss: 0.4285 - val_categorical_accuracy: 0.9179\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 389/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3409 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0710 - categorical_accuracy: 0.9825 - val_loss: 0.3409 - val_categorical_accuracy: 0.9214\n",
      "Epoch 390/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4554 - categorical_accuracy: 0.9036\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.1096 - categorical_accuracy: 0.9768 - val_loss: 0.4554 - val_categorical_accuracy: 0.9036\n",
      "Epoch 391/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.6681 - categorical_accuracy: 0.8857 0s - loss: 1.1024 - \n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0714 - categorical_accuracy: 0.9800 - val_loss: 0.6681 - val_categorical_accuracy: 0.8857\n",
      "Epoch 392/400\n",
      "18/18 [==============================] - 1s 71ms/step - loss: 0.4055 - categorical_accuracy: 0.9393\n",
      "175/175 [==============================] - 34s 196ms/step - loss: 0.0593 - categorical_accuracy: 0.9882 - val_loss: 0.4055 - val_categorical_accuracy: 0.9393\n",
      "Epoch 393/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.3754 - categorical_accuracy: 0.9214\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0689 - categorical_accuracy: 0.9857 - val_loss: 0.3754 - val_categorical_accuracy: 0.9214\n",
      "Epoch 394/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.6528 - categorical_accuracy: 0.9000\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0656 - categorical_accuracy: 0.9861 - val_loss: 0.6528 - val_categorical_accuracy: 0.9000\n",
      "Epoch 395/400\n",
      "18/18 [==============================] - 1s 66ms/step - loss: 0.3722 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0688 - categorical_accuracy: 0.9854 - val_loss: 0.3722 - val_categorical_accuracy: 0.9179\n",
      "Epoch 396/400\n",
      "18/18 [==============================] - 1s 70ms/step - loss: 0.4890 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0688 - categorical_accuracy: 0.9832 - val_loss: 0.4890 - val_categorical_accuracy: 0.9143\n",
      "Epoch 397/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.6033 - categorical_accuracy: 0.9071\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0539 - categorical_accuracy: 0.9875 - val_loss: 0.6033 - val_categorical_accuracy: 0.9071\n",
      "Epoch 398/400\n",
      "18/18 [==============================] - 1s 70ms/step - loss: 0.4368 - categorical_accuracy: 0.9179\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0778 - categorical_accuracy: 0.9782 - val_loss: 0.4368 - val_categorical_accuracy: 0.9179\n",
      "Epoch 399/400\n",
      "18/18 [==============================] - 1s 68ms/step - loss: 0.5495 - categorical_accuracy: 0.8964\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0707 - categorical_accuracy: 0.9818 - val_loss: 0.5495 - val_categorical_accuracy: 0.8964\n",
      "Epoch 400/400\n",
      "18/18 [==============================] - 1s 67ms/step - loss: 0.4688 - categorical_accuracy: 0.9143\n",
      "175/175 [==============================] - 34s 195ms/step - loss: 0.0483 - categorical_accuracy: 0.9900 - val_loss: 0.4688 - val_categorical_accuracy: 0.9143\n"
     ]
    }
   ],
   "source": [
    "import math\n",
    "\n",
    "history = indusNetModel.fit_generator(\n",
    "      train_generator,\n",
    "      steps_per_epoch=math.ceil(number_of_training_samples//batch_size),\n",
    "      epochs=400,\n",
    "      callbacks=callbacks_list,\n",
    "      validation_data=validation_generator,\n",
    "      validation_steps=math.ceil(number_of_validation_samples//batch_size))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "acc=history.history['categorical_accuracy']\n",
    "val_acc=history.history['val_categorical_accuracy']\n",
    "loss=history.history['loss']\n",
    "val_loss=history.history['val_loss']\n",
    "epochs=range(1,len(acc)+1)\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.plot(epochs, acc, 'bo', label='Training acc')\n",
    "plt.plot(epochs, val_acc, 'b', label='Validation acc')\n",
    "plt.title('Training and validation accuracy')\n",
    "plt.legend()\n",
    "plt.figure()\n",
    "\n",
    "plt.plot(epochs, loss, 'bo', label='Training loss')\n",
    "plt.plot(epochs, val_loss, 'b', label='Validation loss')\n",
    "plt.title('Training and validation loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.python.keras.models import load_model\n",
    "\n",
    "best_model = load_model('IndusNet_7_classes_100_100_2nd Experiment.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "35/35 [==============================] - 4s 122ms/step - loss: 0.4053 - categorical_accuracy: 0.9232\n",
      "Test-set accuracy: 92.32%\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.python.keras.models import load_model\n",
    "\n",
    "steps_test=int(number_of_test_samples/batch_size)\n",
    "result = best_model.evaluate_generator(test_generator, steps=steps_test,verbose=1)\n",
    "print(\"Test-set accuracy: {0:.2%}\".format(result[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of test examples:  (560, 100, 100, 3)\n",
      "Total number of labels: (560, 1)\n"
     ]
    }
   ],
   "source": [
    "# Preparing test data\n",
    "import glob\n",
    "from pathlib import Path\n",
    "from tensorflow.python.keras.utils import to_categorical\n",
    "import cv2\n",
    "\n",
    "apple_cases_dir = Path('D:\\\\TensorFlow Programs\\\\fruit_image_classification_Indus Fruit Dataset\\\\Indus Fruit Dataset_Resized Images_100_100\\\\Test\\\\apple')\n",
    "banana_cases_dir = Path('D:\\\\TensorFlow Programs\\\\fruit_image_classification_Indus Fruit Dataset\\\\Indus Fruit Dataset_Resized Images_100_100\\\\Test\\\\banana')\n",
    "orange_cases_dir = Path('D:\\\\TensorFlow Programs\\\\fruit_image_classification_Indus Fruit Dataset\\\\Indus Fruit Dataset_Resized Images_100_100\\\\Test\\\\orange')\n",
    "watermelon_cases_dir = Path('D:\\\\TensorFlow Programs\\\\fruit_image_classification_Indus Fruit Dataset\\\\Indus Fruit Dataset_Resized Images_100_100\\\\Test\\\\watermelon')\n",
    "mango_cases_dir = Path('D:\\\\TensorFlow Programs\\\\fruit_image_classification_Indus Fruit Dataset\\\\Indus Fruit Dataset_Resized Images_100_100\\\\Test\\\\mango')\n",
    "grape_cases_dir = Path('D:\\\\TensorFlow Programs\\\\fruit_image_classification_Indus Fruit Dataset\\\\Indus Fruit Dataset_Resized Images_100_100\\\\Test\\\\grape')\n",
    "strawberry_cases_dir = Path('D:\\\\TensorFlow Programs\\\\fruit_image_classification_Indus Fruit Dataset\\\\Indus Fruit Dataset_Resized Images_100_100\\\\Test\\\\strawberry')\n",
    "\n",
    "apple_cases = apple_cases_dir.glob('*.jpg')\n",
    "banana_cases = banana_cases_dir.glob('*.jpg')\n",
    "orange_cases = orange_cases_dir.glob('*.jpg')\n",
    "watermelon_cases = watermelon_cases_dir.glob('*.jpg')\n",
    "mango_cases = mango_cases_dir.glob('*.jpg')\n",
    "grape_cases = grape_cases_dir.glob('*.jpg')\n",
    "strawberry_cases = strawberry_cases_dir.glob('*.jpg')\n",
    "\n",
    "test_data = []\n",
    "test_labels = []\n",
    "\n",
    "for img in apple_cases:\n",
    "    img = cv2.imread(str(img))\n",
    "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
    "    img = img.astype(np.float32)/255.\n",
    "    label = [0]\n",
    "    test_data.append(img)\n",
    "    test_labels.append(label)\n",
    "\n",
    "for img in banana_cases:\n",
    "    img = cv2.imread(str(img))\n",
    "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
    "    img = img.astype(np.float32)/255.\n",
    "    label = [1]\n",
    "    test_data.append(img)\n",
    "    test_labels.append(label)\n",
    "\n",
    "for img in grape_cases:\n",
    "    img = cv2.imread(str(img))\n",
    "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
    "    img = img.astype(np.float32)/255.\n",
    "    label = [2]\n",
    "    test_data.append(img)\n",
    "    test_labels.append(label)\n",
    "    \n",
    "for img in mango_cases:\n",
    "    img = cv2.imread(str(img))\n",
    "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
    "    img = img.astype(np.float32)/255.\n",
    "    label = [3]\n",
    "    test_data.append(img)\n",
    "    test_labels.append(label)\n",
    "\n",
    "for img in orange_cases:\n",
    "    img = cv2.imread(str(img))\n",
    "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
    "    img = img.astype(np.float32)/255.\n",
    "    label = [4]\n",
    "    test_data.append(img)\n",
    "    test_labels.append(label)\n",
    "\n",
    "for img in strawberry_cases:\n",
    "    img = cv2.imread(str(img))\n",
    "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
    "    img = img.astype(np.float32)/255.\n",
    "    label = [5]\n",
    "    test_data.append(img)\n",
    "    test_labels.append(label)\n",
    "\n",
    "for img in watermelon_cases:\n",
    "    img = cv2.imread(str(img))\n",
    "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
    "    img = img.astype(np.float32)/255.\n",
    "    label = [6]\n",
    "    test_data.append(img)\n",
    "    test_labels.append(label)\n",
    "\n",
    "test_data = np.array(test_data)\n",
    "test_labels = np.array(test_labels)\n",
    "test_labels_one_hot_encoding=to_categorical(test_labels)\n",
    "\n",
    "print(\"Total number of test examples: \", test_data.shape)\n",
    "print(\"Total number of labels:\", test_labels.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "560/560 [==============================] - 3s 5ms/sample - loss: 0.4053 - categorical_accuracy: 0.9232\n",
      "Loss on test set:  0.40531098887856515\n",
      "Accuracy on test set:  0.92321426\n"
     ]
    }
   ],
   "source": [
    "# Evaluation on test dataset\n",
    "\n",
    "test_loss, test_score = best_model.evaluate(test_data, test_labels_one_hot_encoding, batch_size=8)\n",
    "print(\"Loss on test set: \", test_loss)\n",
    "print(\"Accuracy on test set: \", test_score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "560/560 [==============================] - 2s 4ms/sample\n",
      "(560, 1)\n",
      "(560, 1)\n"
     ]
    }
   ],
   "source": [
    "# Get predictions\n",
    "preds = best_model.predict(test_data, batch_size=16,verbose=1)\n",
    "preds=np.around(preds)\n",
    "orig_test_labels=test_labels\n",
    "predicted_labels=np.argmax(preds,axis=1)\n",
    "predicted_labels=np.reshape(predicted_labels,(560,1))\n",
    "print(predicted_labels.shape)\n",
    "print(orig_test_labels.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 3600x2400 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "# Get the confusion matrix\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from mlxtend.plotting import plot_confusion_matrix\n",
    "\n",
    "cm  = confusion_matrix(orig_test_labels, predicted_labels)\n",
    "plt.figure(dpi=600)\n",
    "plot_confusion_matrix(cm,figsize=(12,10), hide_ticks=True,cmap=plt.cm.Blues)\n",
    "plt.xticks(range(7), ['apple', 'banana','grape','mango','orange','strawberry','watermelon'], fontsize=12)\n",
    "plt.yticks(range(7), ['apple', 'banana','grape','mango','orange','strawberry','watermelon'], fontsize=12)\n",
    "plt.xlabel('Predicted Label',fontsize=14)\n",
    "plt.ylabel('True Label',fontsize=14)\n",
    "plt.savefig('IndusNet_7 classes_100_100_2nd Experiment.jpeg',dpi=1200)\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}