Hw9.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "843b2c09",
   "metadata": {},
   "outputs": [],
   "source": [
    "# all imports\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import layers, models\n",
    "from tensorflow.keras.datasets import fashion_mnist\n",
    "from tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau\n",
    "import matplotlib.pyplot as plt\n",
    "import keras\n",
    "from keras import optimizers\n",
    "from keras.optimizers import Adam, RMSprop, SGD\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout\n",
    "from keras import utils\n",
    "from tensorflow.keras.layers import Conv2D\n",
    "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "83e50358",
   "metadata": {},
   "outputs": [],
   "source": [
    "# downloading the data\n",
    "(train_x, train_y), (test_x, test_y) = fashion_mnist.load_data()\n",
    "train_x, test_x = train_x / 255.0, test_x / 255.0"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "46d465dc",
   "metadata": {},
   "source": [
    "### Adam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "f6848a7a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Functions for creating and training model\n",
    "def create_model():\n",
    "    model = models.Sequential([\n",
    "        layers.Flatten(input_shape=(28, 28)),\n",
    "        layers.Dense(512, activation='relu'),\n",
    "        layers.Dropout(0.2),\n",
    "        layers.Dense(256, activation='relu'),\n",
    "        layers.Dropout(0.2),\n",
    "        layers.Dense(128, activation='relu'),\n",
    "        layers.Dropout(0.2),\n",
    "        layers.Dense(10, activation='softmax')\n",
    "    ])\n",
    "    return model\n",
    "\n",
    "def train_model_adam(model, epochs=100):\n",
    "    model.compile(optimizer='adam',\n",
    "                  loss='sparse_categorical_crossentropy',\n",
    "                  metrics=['accuracy'])\n",
    "    reduce_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.2, patience=5, min_lr=0.001)\n",
    "    history = model.fit(train_x, train_y, epochs=epochs, batch_size=256, validation_split=0.2, verbose=1, callbacks=[reduce_lr])\n",
    "\n",
    "    return history\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "aefb2f58",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.6509 - accuracy: 0.7707 - val_loss: 0.4225 - val_accuracy: 0.8462 - lr: 0.0010\n",
      "Epoch 2/100\n",
      "188/188 [==============================] - 2s 8ms/step - loss: 0.4233 - accuracy: 0.8469 - val_loss: 0.3812 - val_accuracy: 0.8622 - lr: 0.0010\n",
      "Epoch 3/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.3781 - accuracy: 0.8625 - val_loss: 0.3811 - val_accuracy: 0.8612 - lr: 0.0010\n",
      "Epoch 4/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.3546 - accuracy: 0.8708 - val_loss: 0.3433 - val_accuracy: 0.8792 - lr: 0.0010\n",
      "Epoch 5/100\n",
      "188/188 [==============================] - 2s 8ms/step - loss: 0.3391 - accuracy: 0.8757 - val_loss: 0.3503 - val_accuracy: 0.8734 - lr: 0.0010\n",
      "Epoch 6/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.3198 - accuracy: 0.8823 - val_loss: 0.3314 - val_accuracy: 0.8794 - lr: 0.0010\n",
      "Epoch 7/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.3058 - accuracy: 0.8870 - val_loss: 0.3186 - val_accuracy: 0.8849 - lr: 0.0010\n",
      "Epoch 8/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2940 - accuracy: 0.8928 - val_loss: 0.3087 - val_accuracy: 0.8863 - lr: 0.0010\n",
      "Epoch 9/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2837 - accuracy: 0.8944 - val_loss: 0.3179 - val_accuracy: 0.8854 - lr: 0.0010\n",
      "Epoch 10/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2792 - accuracy: 0.8969 - val_loss: 0.3261 - val_accuracy: 0.8830 - lr: 0.0010\n",
      "Epoch 11/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2683 - accuracy: 0.9004 - val_loss: 0.3118 - val_accuracy: 0.8888 - lr: 0.0010\n",
      "Epoch 12/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2620 - accuracy: 0.9021 - val_loss: 0.3044 - val_accuracy: 0.8895 - lr: 0.0010\n",
      "Epoch 13/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2511 - accuracy: 0.9048 - val_loss: 0.3171 - val_accuracy: 0.8854 - lr: 0.0010\n",
      "Epoch 14/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2515 - accuracy: 0.9057 - val_loss: 0.3006 - val_accuracy: 0.8917 - lr: 0.0010\n",
      "Epoch 15/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2452 - accuracy: 0.9075 - val_loss: 0.3138 - val_accuracy: 0.8880 - lr: 0.0010\n",
      "Epoch 16/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2401 - accuracy: 0.9099 - val_loss: 0.3026 - val_accuracy: 0.8920 - lr: 0.0010\n",
      "Epoch 17/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2355 - accuracy: 0.9094 - val_loss: 0.3332 - val_accuracy: 0.8878 - lr: 0.0010\n",
      "Epoch 18/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2288 - accuracy: 0.9121 - val_loss: 0.3045 - val_accuracy: 0.8944 - lr: 0.0010\n",
      "Epoch 19/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.2252 - accuracy: 0.9146 - val_loss: 0.3031 - val_accuracy: 0.8946 - lr: 0.0010\n",
      "Epoch 20/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2232 - accuracy: 0.9155 - val_loss: 0.2913 - val_accuracy: 0.8978 - lr: 0.0010\n",
      "Epoch 21/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2188 - accuracy: 0.9176 - val_loss: 0.3040 - val_accuracy: 0.8949 - lr: 0.0010\n",
      "Epoch 22/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2183 - accuracy: 0.9175 - val_loss: 0.2967 - val_accuracy: 0.8977 - lr: 0.0010\n",
      "Epoch 23/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2125 - accuracy: 0.9197 - val_loss: 0.3012 - val_accuracy: 0.8950 - lr: 0.0010\n",
      "Epoch 24/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2038 - accuracy: 0.9218 - val_loss: 0.3036 - val_accuracy: 0.8957 - lr: 0.0010\n",
      "Epoch 25/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2005 - accuracy: 0.9241 - val_loss: 0.3099 - val_accuracy: 0.8988 - lr: 0.0010\n",
      "Epoch 26/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.2027 - accuracy: 0.9233 - val_loss: 0.2987 - val_accuracy: 0.8978 - lr: 0.0010\n",
      "Epoch 27/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1972 - accuracy: 0.9257 - val_loss: 0.3019 - val_accuracy: 0.9008 - lr: 0.0010\n",
      "Epoch 28/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1905 - accuracy: 0.9272 - val_loss: 0.3253 - val_accuracy: 0.8957 - lr: 0.0010\n",
      "Epoch 29/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1945 - accuracy: 0.9258 - val_loss: 0.3227 - val_accuracy: 0.8932 - lr: 0.0010\n",
      "Epoch 30/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1909 - accuracy: 0.9274 - val_loss: 0.3075 - val_accuracy: 0.8987 - lr: 0.0010\n",
      "Epoch 31/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1826 - accuracy: 0.9308 - val_loss: 0.3080 - val_accuracy: 0.8985 - lr: 0.0010\n",
      "Epoch 32/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1806 - accuracy: 0.9321 - val_loss: 0.3148 - val_accuracy: 0.9006 - lr: 0.0010\n",
      "Epoch 33/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1801 - accuracy: 0.9314 - val_loss: 0.3182 - val_accuracy: 0.8999 - lr: 0.0010\n",
      "Epoch 34/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1776 - accuracy: 0.9323 - val_loss: 0.3239 - val_accuracy: 0.9003 - lr: 0.0010\n",
      "Epoch 35/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1747 - accuracy: 0.9330 - val_loss: 0.3102 - val_accuracy: 0.9008 - lr: 0.0010\n",
      "Epoch 36/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1722 - accuracy: 0.9347 - val_loss: 0.3224 - val_accuracy: 0.8972 - lr: 0.0010\n",
      "Epoch 37/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1700 - accuracy: 0.9352 - val_loss: 0.3204 - val_accuracy: 0.8985 - lr: 0.0010\n",
      "Epoch 38/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1634 - accuracy: 0.9379 - val_loss: 0.3354 - val_accuracy: 0.8966 - lr: 0.0010\n",
      "Epoch 39/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1637 - accuracy: 0.9375 - val_loss: 0.3024 - val_accuracy: 0.9022 - lr: 0.0010\n",
      "Epoch 40/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1653 - accuracy: 0.9384 - val_loss: 0.3191 - val_accuracy: 0.9002 - lr: 0.0010\n",
      "Epoch 41/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1597 - accuracy: 0.9390 - val_loss: 0.3277 - val_accuracy: 0.8967 - lr: 0.0010\n",
      "Epoch 42/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1598 - accuracy: 0.9386 - val_loss: 0.3386 - val_accuracy: 0.8979 - lr: 0.0010\n",
      "Epoch 43/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1567 - accuracy: 0.9402 - val_loss: 0.3341 - val_accuracy: 0.8972 - lr: 0.0010\n",
      "Epoch 44/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1534 - accuracy: 0.9411 - val_loss: 0.3407 - val_accuracy: 0.9010 - lr: 0.0010\n",
      "Epoch 45/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1547 - accuracy: 0.9400 - val_loss: 0.3404 - val_accuracy: 0.8971 - lr: 0.0010\n",
      "Epoch 46/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1500 - accuracy: 0.9423 - val_loss: 0.3338 - val_accuracy: 0.9003 - lr: 0.0010\n",
      "Epoch 47/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1498 - accuracy: 0.9425 - val_loss: 0.3541 - val_accuracy: 0.8958 - lr: 0.0010\n",
      "Epoch 48/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1475 - accuracy: 0.9426 - val_loss: 0.3355 - val_accuracy: 0.9004 - lr: 0.0010\n",
      "Epoch 49/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1485 - accuracy: 0.9433 - val_loss: 0.3445 - val_accuracy: 0.8978 - lr: 0.0010\n",
      "Epoch 50/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1423 - accuracy: 0.9446 - val_loss: 0.3321 - val_accuracy: 0.9007 - lr: 0.0010\n",
      "Epoch 51/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1458 - accuracy: 0.9434 - val_loss: 0.3379 - val_accuracy: 0.9018 - lr: 0.0010\n",
      "Epoch 52/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1445 - accuracy: 0.9455 - val_loss: 0.3576 - val_accuracy: 0.8970 - lr: 0.0010\n",
      "Epoch 53/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1388 - accuracy: 0.9469 - val_loss: 0.3494 - val_accuracy: 0.8981 - lr: 0.0010\n",
      "Epoch 54/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1390 - accuracy: 0.9463 - val_loss: 0.3564 - val_accuracy: 0.8992 - lr: 0.0010\n",
      "Epoch 55/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1398 - accuracy: 0.9474 - val_loss: 0.3451 - val_accuracy: 0.9009 - lr: 0.0010\n",
      "Epoch 56/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1345 - accuracy: 0.9484 - val_loss: 0.3461 - val_accuracy: 0.9032 - lr: 0.0010\n",
      "Epoch 57/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1356 - accuracy: 0.9477 - val_loss: 0.3553 - val_accuracy: 0.9011 - lr: 0.0010\n",
      "Epoch 58/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1353 - accuracy: 0.9484 - val_loss: 0.3520 - val_accuracy: 0.9010 - lr: 0.0010\n",
      "Epoch 59/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1304 - accuracy: 0.9489 - val_loss: 0.3602 - val_accuracy: 0.9003 - lr: 0.0010\n",
      "Epoch 60/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1295 - accuracy: 0.9506 - val_loss: 0.3456 - val_accuracy: 0.9024 - lr: 0.0010\n",
      "Epoch 61/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1297 - accuracy: 0.9506 - val_loss: 0.3739 - val_accuracy: 0.9003 - lr: 0.0010\n",
      "Epoch 62/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1298 - accuracy: 0.9507 - val_loss: 0.3493 - val_accuracy: 0.9017 - lr: 0.0010\n",
      "Epoch 63/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1237 - accuracy: 0.9529 - val_loss: 0.3788 - val_accuracy: 0.8992 - lr: 0.0010\n",
      "Epoch 64/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1226 - accuracy: 0.9524 - val_loss: 0.3679 - val_accuracy: 0.9017 - lr: 0.0010\n",
      "Epoch 65/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1258 - accuracy: 0.9525 - val_loss: 0.3579 - val_accuracy: 0.8992 - lr: 0.0010\n",
      "Epoch 66/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1220 - accuracy: 0.9528 - val_loss: 0.3443 - val_accuracy: 0.9033 - lr: 0.0010\n",
      "Epoch 67/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1190 - accuracy: 0.9544 - val_loss: 0.3781 - val_accuracy: 0.9009 - lr: 0.0010\n",
      "Epoch 68/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1187 - accuracy: 0.9545 - val_loss: 0.3609 - val_accuracy: 0.9005 - lr: 0.0010\n",
      "Epoch 69/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1225 - accuracy: 0.9526 - val_loss: 0.3692 - val_accuracy: 0.8998 - lr: 0.0010\n",
      "Epoch 70/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1222 - accuracy: 0.9538 - val_loss: 0.3526 - val_accuracy: 0.9000 - lr: 0.0010\n",
      "Epoch 71/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1149 - accuracy: 0.9560 - val_loss: 0.3657 - val_accuracy: 0.9023 - lr: 0.0010\n",
      "Epoch 72/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1164 - accuracy: 0.9554 - val_loss: 0.3688 - val_accuracy: 0.9014 - lr: 0.0010\n",
      "Epoch 73/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1141 - accuracy: 0.9555 - val_loss: 0.3843 - val_accuracy: 0.8997 - lr: 0.0010\n",
      "Epoch 74/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1159 - accuracy: 0.9563 - val_loss: 0.3702 - val_accuracy: 0.9004 - lr: 0.0010\n",
      "Epoch 75/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1106 - accuracy: 0.9575 - val_loss: 0.3711 - val_accuracy: 0.9025 - lr: 0.0010\n",
      "Epoch 76/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1136 - accuracy: 0.9555 - val_loss: 0.3855 - val_accuracy: 0.9007 - lr: 0.0010\n",
      "Epoch 77/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1071 - accuracy: 0.9600 - val_loss: 0.4093 - val_accuracy: 0.9018 - lr: 0.0010\n",
      "Epoch 78/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1115 - accuracy: 0.9580 - val_loss: 0.3788 - val_accuracy: 0.9026 - lr: 0.0010\n",
      "Epoch 79/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1094 - accuracy: 0.9587 - val_loss: 0.3842 - val_accuracy: 0.9010 - lr: 0.0010\n",
      "Epoch 80/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1127 - accuracy: 0.9570 - val_loss: 0.3798 - val_accuracy: 0.9035 - lr: 0.0010\n",
      "Epoch 81/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1104 - accuracy: 0.9590 - val_loss: 0.4011 - val_accuracy: 0.8977 - lr: 0.0010\n",
      "Epoch 82/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1076 - accuracy: 0.9594 - val_loss: 0.3895 - val_accuracy: 0.9008 - lr: 0.0010\n",
      "Epoch 83/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1055 - accuracy: 0.9596 - val_loss: 0.3846 - val_accuracy: 0.9018 - lr: 0.0010\n",
      "Epoch 84/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.1069 - accuracy: 0.9591 - val_loss: 0.3907 - val_accuracy: 0.9001 - lr: 0.0010\n",
      "Epoch 85/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1052 - accuracy: 0.9607 - val_loss: 0.3902 - val_accuracy: 0.9001 - lr: 0.0010\n",
      "Epoch 86/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1065 - accuracy: 0.9583 - val_loss: 0.3898 - val_accuracy: 0.9028 - lr: 0.0010\n",
      "Epoch 87/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1002 - accuracy: 0.9623 - val_loss: 0.4104 - val_accuracy: 0.9009 - lr: 0.0010\n",
      "Epoch 88/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1000 - accuracy: 0.9627 - val_loss: 0.4151 - val_accuracy: 0.9007 - lr: 0.0010\n",
      "Epoch 89/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.0993 - accuracy: 0.9623 - val_loss: 0.3929 - val_accuracy: 0.9016 - lr: 0.0010\n",
      "Epoch 90/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1008 - accuracy: 0.9609 - val_loss: 0.3934 - val_accuracy: 0.9033 - lr: 0.0010\n",
      "Epoch 91/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.0953 - accuracy: 0.9636 - val_loss: 0.3975 - val_accuracy: 0.9007 - lr: 0.0010\n",
      "Epoch 92/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.1032 - accuracy: 0.9615 - val_loss: 0.4131 - val_accuracy: 0.9019 - lr: 0.0010\n",
      "Epoch 93/100\n",
      "188/188 [==============================] - 2s 10ms/step - loss: 0.0999 - accuracy: 0.9625 - val_loss: 0.4053 - val_accuracy: 0.9003 - lr: 0.0010\n",
      "Epoch 94/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.0984 - accuracy: 0.9627 - val_loss: 0.4080 - val_accuracy: 0.8988 - lr: 0.0010\n",
      "Epoch 95/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.0987 - accuracy: 0.9627 - val_loss: 0.4079 - val_accuracy: 0.9012 - lr: 0.0010\n",
      "Epoch 96/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.0969 - accuracy: 0.9634 - val_loss: 0.4279 - val_accuracy: 0.8973 - lr: 0.0010\n",
      "Epoch 97/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.0974 - accuracy: 0.9631 - val_loss: 0.4141 - val_accuracy: 0.9037 - lr: 0.0010\n",
      "Epoch 98/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.0975 - accuracy: 0.9627 - val_loss: 0.4037 - val_accuracy: 0.8984 - lr: 0.0010\n",
      "Epoch 99/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.0930 - accuracy: 0.9647 - val_loss: 0.4153 - val_accuracy: 0.9028 - lr: 0.0010\n",
      "Epoch 100/100\n",
      "188/188 [==============================] - 2s 9ms/step - loss: 0.0947 - accuracy: 0.9643 - val_loss: 0.4126 - val_accuracy: 0.9022 - lr: 0.0010\n"
     ]
    }
   ],
   "source": [
    "# Create the first model (adam optimization)\n",
    "model_adam_optim = create_model()\n",
    "history_adam = train_model_adam(model_adam_optim)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "df515bfe",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 0s 1ms/step - loss: 0.4435 - accuracy: 0.8972\n",
      "Accurancy on test data optimizer_Adam 0.8971999883651733\n",
      "Loss on test data optimizer_Adam 0.44346514344215393\n",
      "1875/1875 [==============================] - 2s 923us/step - loss: 0.1276 - accuracy: 0.9633\n",
      "Accurancy on train data optimizer_Adam: 0.9632999897003174\n",
      "Loss on train data optimizer_Adam: 0.12758523225784302\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualization of result and fing loss and accuracy - Adam optimazer\n",
    "\n",
    "test_loss, test_acc = model_adam_optim.evaluate(test_x, test_y)\n",
    "print('Accurancy on test data optimizer_Adam', test_acc)\n",
    "print('Loss on test data optimizer_Adam', test_loss)\n",
    "\n",
    "train_loss, train_acc = model_adam_optim.evaluate(train_x, train_y)\n",
    "print('Accurancy on train data optimizer_Adam:', train_acc)\n",
    "print('Loss on train data optimizer_Adam:', train_loss)\n",
    "\n",
    "plt.plot(history_adam.history['accuracy'], label='Accurancy on train')\n",
    "plt.plot(history_adam.history['val_accuracy'], label='Accurancy on test')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accurancy')\n",
    "plt.title('Adam optimizer')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.plot(history_adam.history['loss'], label='loss on train')\n",
    "plt.plot(history_adam.history['val_loss'], label='loss on test')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Adam optimizer')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59d58cad",
   "metadata": {},
   "source": [
    "### RMSprop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "165a17fb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train (study) funcion for RMS model\n",
    "def train_model_rmsprop(model, epochs=100):\n",
    "    model.compile(optimizer='rmsprop', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "    early_stopping = EarlyStopping(monitor='val_accuracy', patience=6, restore_best_weights=True)\n",
    "    history = model.fit(train_x, train_y, epochs=epochs, batch_size=64, validation_split=0.2, callbacks=[early_stopping])\n",
    "    return history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "eb01e0d7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.6002 - accuracy: 0.7824 - val_loss: 0.4393 - val_accuracy: 0.8330\n",
      "Epoch 2/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4438 - accuracy: 0.8396 - val_loss: 0.3850 - val_accuracy: 0.8627\n",
      "Epoch 3/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4175 - accuracy: 0.8549 - val_loss: 0.3832 - val_accuracy: 0.8679\n",
      "Epoch 4/100\n",
      "750/750 [==============================] - 3s 5ms/step - loss: 0.4123 - accuracy: 0.8588 - val_loss: 0.3888 - val_accuracy: 0.8657\n",
      "Epoch 5/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4051 - accuracy: 0.8620 - val_loss: 0.3865 - val_accuracy: 0.8629\n",
      "Epoch 6/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4015 - accuracy: 0.8649 - val_loss: 0.3991 - val_accuracy: 0.8698\n",
      "Epoch 7/100\n",
      "750/750 [==============================] - 3s 5ms/step - loss: 0.3970 - accuracy: 0.8664 - val_loss: 0.4793 - val_accuracy: 0.8585\n",
      "Epoch 8/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.3983 - accuracy: 0.8689 - val_loss: 0.3793 - val_accuracy: 0.8757\n",
      "Epoch 9/100\n",
      "750/750 [==============================] - 3s 5ms/step - loss: 0.4004 - accuracy: 0.8691 - val_loss: 0.3961 - val_accuracy: 0.8752\n",
      "Epoch 10/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4090 - accuracy: 0.8682 - val_loss: 0.4229 - val_accuracy: 0.8798\n",
      "Epoch 11/100\n",
      "750/750 [==============================] - 3s 5ms/step - loss: 0.4031 - accuracy: 0.8709 - val_loss: 0.4053 - val_accuracy: 0.8819\n",
      "Epoch 12/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4033 - accuracy: 0.8711 - val_loss: 0.4336 - val_accuracy: 0.8679\n",
      "Epoch 13/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4014 - accuracy: 0.8710 - val_loss: 0.4331 - val_accuracy: 0.8772\n",
      "Epoch 14/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4086 - accuracy: 0.8715 - val_loss: 0.4221 - val_accuracy: 0.8835\n",
      "Epoch 15/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4122 - accuracy: 0.8722 - val_loss: 0.4699 - val_accuracy: 0.8742\n",
      "Epoch 16/100\n",
      "750/750 [==============================] - 3s 5ms/step - loss: 0.4026 - accuracy: 0.8741 - val_loss: 0.4251 - val_accuracy: 0.8813\n",
      "Epoch 17/100\n",
      "750/750 [==============================] - 3s 5ms/step - loss: 0.4102 - accuracy: 0.8705 - val_loss: 0.4601 - val_accuracy: 0.8765\n",
      "Epoch 18/100\n",
      "750/750 [==============================] - 3s 5ms/step - loss: 0.4115 - accuracy: 0.8721 - val_loss: 0.4523 - val_accuracy: 0.8806\n",
      "Epoch 19/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4248 - accuracy: 0.8716 - val_loss: 0.4724 - val_accuracy: 0.8865\n",
      "Epoch 20/100\n",
      "750/750 [==============================] - 3s 5ms/step - loss: 0.4124 - accuracy: 0.8725 - val_loss: 0.4475 - val_accuracy: 0.8839\n",
      "Epoch 21/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4301 - accuracy: 0.8689 - val_loss: 0.4683 - val_accuracy: 0.8698\n",
      "Epoch 22/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4073 - accuracy: 0.8742 - val_loss: 0.6377 - val_accuracy: 0.8650\n",
      "Epoch 23/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4165 - accuracy: 0.8728 - val_loss: 0.5046 - val_accuracy: 0.8741\n",
      "Epoch 24/100\n",
      "750/750 [==============================] - 4s 5ms/step - loss: 0.4311 - accuracy: 0.8723 - val_loss: 0.5536 - val_accuracy: 0.8799\n",
      "Epoch 25/100\n",
      "750/750 [==============================] - 3s 5ms/step - loss: 0.4354 - accuracy: 0.8720 - val_loss: 0.4950 - val_accuracy: 0.8789\n"
     ]
    }
   ],
   "source": [
    "# Create second model RMS\n",
    "\n",
    "model_rms = create_model()\n",
    "history_rmsprop = train_model_rmsprop(model_rms)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "62406c33",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 0s 929us/step - loss: 0.5296 - accuracy: 0.8733\n",
      "Accurancy on test data optimizer_RMSprop 0.8733000159263611\n",
      "Loss on test data optimizer_RMSprop 0.5295857191085815\n",
      "1875/1875 [==============================] - 2s 894us/step - loss: 0.3051 - accuracy: 0.9014\n",
      "Accurancy on train data optimizer_RMSprop : 0.9013500213623047\n",
      "Loss on train data optimizer_RMSprop: 0.30512794852256775\n"
     ]
    }
   ],
   "source": [
    "# Find accuracy and loss for model RMS\n",
    "test_loss, test_acc = model_rms.evaluate(test_x, test_y)\n",
    "print('Accurancy on test data optimizer_RMSprop', test_acc)\n",
    "print('Loss on test data optimizer_RMSprop', test_loss)\n",
    "\n",
    "train_loss, train_acc = model_rms.evaluate(train_x, train_y)\n",
    "print('Accurancy on train data optimizer_RMSprop :', train_acc)\n",
    "print('Loss on train data optimizer_RMSprop:', train_loss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "da158ddc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize result (accuracy and loss) for RMS optimazer\n",
    "\n",
    "plt.plot(history_rmsprop.history['accuracy'], label='Accurancy on train')\n",
    "plt.plot(history_rmsprop.history['val_accuracy'], label='Accurancy on test')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accurancy')\n",
    "plt.title('RMSprop optimizer')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.plot(history_rmsprop.history['loss'], label='loss on train')\n",
    "plt.plot(history_rmsprop.history['val_loss'], label='loss on test')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('RMSprop optimizer')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "640b8655",
   "metadata": {},
   "source": [
    "### SGD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "b93f54aa",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Study function for SGD optimizator \n",
    "def train_model_sgd(model, epochs=100):\n",
    "    model.compile(optimizer='sgd',loss='sparse_categorical_crossentropy',metrics=['accuracy'])\n",
    "\n",
    "    early_stopping = EarlyStopping(monitor='val_accuracy', patience=6, restore_best_weights=True)\n",
    "    history = model.fit(train_x, train_y, epochs=epochs, batch_size=64, validation_split=0.2, callbacks=[early_stopping])\n",
    "\n",
    "    return history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "992a64b8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 1.0730 - accuracy: 0.6236 - val_loss: 0.6315 - val_accuracy: 0.7797\n",
      "Epoch 2/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.6695 - accuracy: 0.7645 - val_loss: 0.5310 - val_accuracy: 0.8133\n",
      "Epoch 3/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.5848 - accuracy: 0.7974 - val_loss: 0.4871 - val_accuracy: 0.8288\n",
      "Epoch 4/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.5358 - accuracy: 0.8135 - val_loss: 0.4588 - val_accuracy: 0.8352\n",
      "Epoch 5/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.4985 - accuracy: 0.8264 - val_loss: 0.4373 - val_accuracy: 0.8418\n",
      "Epoch 6/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.4736 - accuracy: 0.8345 - val_loss: 0.4198 - val_accuracy: 0.8488\n",
      "Epoch 7/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.4547 - accuracy: 0.8386 - val_loss: 0.4078 - val_accuracy: 0.8535\n",
      "Epoch 8/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.4393 - accuracy: 0.8453 - val_loss: 0.4016 - val_accuracy: 0.8518\n",
      "Epoch 9/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.4278 - accuracy: 0.8491 - val_loss: 0.3903 - val_accuracy: 0.8603\n",
      "Epoch 10/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.4144 - accuracy: 0.8525 - val_loss: 0.3784 - val_accuracy: 0.8627\n",
      "Epoch 11/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.4042 - accuracy: 0.8559 - val_loss: 0.3738 - val_accuracy: 0.8673\n",
      "Epoch 12/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.3938 - accuracy: 0.8584 - val_loss: 0.3679 - val_accuracy: 0.8674\n",
      "Epoch 13/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.3857 - accuracy: 0.8635 - val_loss: 0.3623 - val_accuracy: 0.8714\n",
      "Epoch 14/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.3779 - accuracy: 0.8659 - val_loss: 0.3588 - val_accuracy: 0.8719\n",
      "Epoch 15/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.3707 - accuracy: 0.8675 - val_loss: 0.3497 - val_accuracy: 0.8742\n",
      "Epoch 16/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.3651 - accuracy: 0.8691 - val_loss: 0.3528 - val_accuracy: 0.8729\n",
      "Epoch 17/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.3567 - accuracy: 0.8730 - val_loss: 0.3486 - val_accuracy: 0.8748\n",
      "Epoch 18/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.3528 - accuracy: 0.8736 - val_loss: 0.3438 - val_accuracy: 0.8765\n",
      "Epoch 19/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.3471 - accuracy: 0.8746 - val_loss: 0.3411 - val_accuracy: 0.8785\n",
      "Epoch 20/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.3416 - accuracy: 0.8775 - val_loss: 0.3370 - val_accuracy: 0.8791\n",
      "Epoch 21/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.3372 - accuracy: 0.8791 - val_loss: 0.3317 - val_accuracy: 0.8811\n",
      "Epoch 22/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.3310 - accuracy: 0.8798 - val_loss: 0.3303 - val_accuracy: 0.8817\n",
      "Epoch 23/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.3260 - accuracy: 0.8829 - val_loss: 0.3302 - val_accuracy: 0.8808\n",
      "Epoch 24/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.3253 - accuracy: 0.8833 - val_loss: 0.3261 - val_accuracy: 0.8831\n",
      "Epoch 25/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.3198 - accuracy: 0.8838 - val_loss: 0.3246 - val_accuracy: 0.8835\n",
      "Epoch 26/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.3155 - accuracy: 0.8876 - val_loss: 0.3223 - val_accuracy: 0.8839\n",
      "Epoch 27/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.3119 - accuracy: 0.8885 - val_loss: 0.3197 - val_accuracy: 0.8855\n",
      "Epoch 28/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.3106 - accuracy: 0.8887 - val_loss: 0.3168 - val_accuracy: 0.8867\n",
      "Epoch 29/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.3049 - accuracy: 0.8913 - val_loss: 0.3161 - val_accuracy: 0.8850\n",
      "Epoch 30/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.3017 - accuracy: 0.8920 - val_loss: 0.3131 - val_accuracy: 0.8870\n",
      "Epoch 31/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2960 - accuracy: 0.8939 - val_loss: 0.3141 - val_accuracy: 0.8859\n",
      "Epoch 32/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.2944 - accuracy: 0.8926 - val_loss: 0.3151 - val_accuracy: 0.8872\n",
      "Epoch 33/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.2891 - accuracy: 0.8960 - val_loss: 0.3114 - val_accuracy: 0.8883\n",
      "Epoch 34/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.2884 - accuracy: 0.8954 - val_loss: 0.3086 - val_accuracy: 0.8883\n",
      "Epoch 35/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.2851 - accuracy: 0.8975 - val_loss: 0.3094 - val_accuracy: 0.8870\n",
      "Epoch 36/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2814 - accuracy: 0.8974 - val_loss: 0.3109 - val_accuracy: 0.8881\n",
      "Epoch 37/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2786 - accuracy: 0.8995 - val_loss: 0.3074 - val_accuracy: 0.8892\n",
      "Epoch 38/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2770 - accuracy: 0.8997 - val_loss: 0.3107 - val_accuracy: 0.8873\n",
      "Epoch 39/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2720 - accuracy: 0.9021 - val_loss: 0.3047 - val_accuracy: 0.8891\n",
      "Epoch 40/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2672 - accuracy: 0.9028 - val_loss: 0.3069 - val_accuracy: 0.8903\n",
      "Epoch 41/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2661 - accuracy: 0.9031 - val_loss: 0.3068 - val_accuracy: 0.8907\n",
      "Epoch 42/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2635 - accuracy: 0.9053 - val_loss: 0.3012 - val_accuracy: 0.8915\n",
      "Epoch 43/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2603 - accuracy: 0.9055 - val_loss: 0.3002 - val_accuracy: 0.8895\n",
      "Epoch 44/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2600 - accuracy: 0.9048 - val_loss: 0.3061 - val_accuracy: 0.8897\n",
      "Epoch 45/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2580 - accuracy: 0.9071 - val_loss: 0.3017 - val_accuracy: 0.8898\n",
      "Epoch 46/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2519 - accuracy: 0.9086 - val_loss: 0.2980 - val_accuracy: 0.8907\n",
      "Epoch 47/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2505 - accuracy: 0.9100 - val_loss: 0.2986 - val_accuracy: 0.8932\n",
      "Epoch 48/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2483 - accuracy: 0.9089 - val_loss: 0.2995 - val_accuracy: 0.8927\n",
      "Epoch 49/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2468 - accuracy: 0.9111 - val_loss: 0.2976 - val_accuracy: 0.8939\n",
      "Epoch 50/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2430 - accuracy: 0.9117 - val_loss: 0.2962 - val_accuracy: 0.8929\n",
      "Epoch 51/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2407 - accuracy: 0.9138 - val_loss: 0.2956 - val_accuracy: 0.8925\n",
      "Epoch 52/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2384 - accuracy: 0.9139 - val_loss: 0.2982 - val_accuracy: 0.8947\n",
      "Epoch 53/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2373 - accuracy: 0.9147 - val_loss: 0.3073 - val_accuracy: 0.8925\n",
      "Epoch 54/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2339 - accuracy: 0.9146 - val_loss: 0.2980 - val_accuracy: 0.8948\n",
      "Epoch 55/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2323 - accuracy: 0.9158 - val_loss: 0.3025 - val_accuracy: 0.8906\n",
      "Epoch 56/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2281 - accuracy: 0.9170 - val_loss: 0.2972 - val_accuracy: 0.8936\n",
      "Epoch 57/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2286 - accuracy: 0.9162 - val_loss: 0.2937 - val_accuracy: 0.8960\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 58/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.2244 - accuracy: 0.9186 - val_loss: 0.2933 - val_accuracy: 0.8942\n",
      "Epoch 59/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.2244 - accuracy: 0.9186 - val_loss: 0.2951 - val_accuracy: 0.8963\n",
      "Epoch 60/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.2226 - accuracy: 0.9192 - val_loss: 0.2937 - val_accuracy: 0.8964\n",
      "Epoch 61/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.2168 - accuracy: 0.9199 - val_loss: 0.2951 - val_accuracy: 0.8943\n",
      "Epoch 62/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.2159 - accuracy: 0.9231 - val_loss: 0.2910 - val_accuracy: 0.8982\n",
      "Epoch 63/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.2147 - accuracy: 0.9221 - val_loss: 0.2963 - val_accuracy: 0.8973\n",
      "Epoch 64/100\n",
      "750/750 [==============================] - 2s 3ms/step - loss: 0.2117 - accuracy: 0.9233 - val_loss: 0.2945 - val_accuracy: 0.8983\n",
      "Epoch 65/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.2104 - accuracy: 0.9247 - val_loss: 0.2909 - val_accuracy: 0.8978\n",
      "Epoch 66/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.2097 - accuracy: 0.9233 - val_loss: 0.2973 - val_accuracy: 0.8962\n",
      "Epoch 67/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.2045 - accuracy: 0.9241 - val_loss: 0.2945 - val_accuracy: 0.8981\n",
      "Epoch 68/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.2021 - accuracy: 0.9257 - val_loss: 0.3005 - val_accuracy: 0.8985\n",
      "Epoch 69/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.2024 - accuracy: 0.9266 - val_loss: 0.3033 - val_accuracy: 0.8943\n",
      "Epoch 70/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.1996 - accuracy: 0.9269 - val_loss: 0.2986 - val_accuracy: 0.8979\n",
      "Epoch 71/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.1969 - accuracy: 0.9285 - val_loss: 0.3045 - val_accuracy: 0.8946\n",
      "Epoch 72/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.1953 - accuracy: 0.9286 - val_loss: 0.2995 - val_accuracy: 0.8978\n",
      "Epoch 73/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.1931 - accuracy: 0.9302 - val_loss: 0.2905 - val_accuracy: 0.8987\n",
      "Epoch 74/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.1928 - accuracy: 0.9294 - val_loss: 0.2940 - val_accuracy: 0.8979\n",
      "Epoch 75/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.1906 - accuracy: 0.9305 - val_loss: 0.2997 - val_accuracy: 0.8972\n",
      "Epoch 76/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.1856 - accuracy: 0.9333 - val_loss: 0.3009 - val_accuracy: 0.8976\n",
      "Epoch 77/100\n",
      "750/750 [==============================] - 3s 3ms/step - loss: 0.1891 - accuracy: 0.9313 - val_loss: 0.2963 - val_accuracy: 0.9012\n",
      "Epoch 78/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.1846 - accuracy: 0.9323 - val_loss: 0.2983 - val_accuracy: 0.8987\n",
      "Epoch 79/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.1852 - accuracy: 0.9321 - val_loss: 0.2967 - val_accuracy: 0.8973\n",
      "Epoch 80/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.1828 - accuracy: 0.9335 - val_loss: 0.2973 - val_accuracy: 0.8967\n",
      "Epoch 81/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.1804 - accuracy: 0.9349 - val_loss: 0.2957 - val_accuracy: 0.9003\n",
      "Epoch 82/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.1775 - accuracy: 0.9350 - val_loss: 0.3020 - val_accuracy: 0.8963\n",
      "Epoch 83/100\n",
      "750/750 [==============================] - 3s 4ms/step - loss: 0.1763 - accuracy: 0.9364 - val_loss: 0.3007 - val_accuracy: 0.8992\n"
     ]
    }
   ],
   "source": [
    "# create SGD model\n",
    "model_sgd = create_model()\n",
    "history_sgd = train_model_sgd(model_sgd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "fb8f10f3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 0s 883us/step - loss: 0.3255 - accuracy: 0.8893\n",
      "Accurancy on test data optimizer_SGD 0.8892999887466431\n",
      "Loss on test data optimizer_SGD 0.32552215456962585\n",
      "1875/1875 [==============================] - 2s 878us/step - loss: 0.1729 - accuracy: 0.9408\n",
      "Accurancy on train data optimizer_SGD : 0.9407833218574524\n",
      "Loss on train data optimizer_SGD: 0.17288757860660553\n"
     ]
    }
   ],
   "source": [
    "# Find accuracy and loss for model SGD\n",
    "test_loss, test_acc = model_sgd.evaluate(test_x, test_y)\n",
    "print('Accurancy on test data optimizer_SGD', test_acc)\n",
    "print('Loss on test data optimizer_SGD', test_loss)\n",
    "\n",
    "train_loss, train_acc = model_sgd.evaluate(train_x, train_y)\n",
    "print('Accurancy on train data optimizer_SGD :', train_acc)\n",
    "print('Loss on train data optimizer_SGD:', train_loss)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "00f9c4b5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize result (accuracy and loss) for SGD optimazer\n",
    "\n",
    "plt.plot(history_sgd.history['accuracy'], label='Accurancy on train')\n",
    "plt.plot(history_sgd.history['val_accuracy'], label='Accurancy on test')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accurancy')\n",
    "plt.title('SGD optimizer')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.plot(history_sgd.history['loss'], label='loss on train')\n",
    "plt.plot(history_sgd.history['val_loss'], label='loss on test')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('SGD optimizer')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae9d786f",
   "metadata": {},
   "source": [
    "### CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "67e155e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_model_cnn():\n",
    "    return keras.Sequential([\n",
    "    Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)),\n",
    "    MaxPooling2D(2, 2),\n",
    "    Conv2D(64, (3, 3), activation='relu'),\n",
    "    MaxPooling2D(2, 2),\n",
    "    Flatten(),\n",
    "    Dense(512, activation='relu'),\n",
    "    Dropout(0.3),\n",
    "    Dense(10, activation='softmax')\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "c137b1c9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_16\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d_1 (Conv2D)           (None, 26, 26, 32)        320       \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 13, 13, 32)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " conv2d_2 (Conv2D)           (None, 11, 11, 64)        18496     \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 5, 5, 64)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " flatten_15 (Flatten)        (None, 1600)              0         \n",
      "                                                                 \n",
      " dense_44 (Dense)            (None, 512)               819712    \n",
      "                                                                 \n",
      " dropout_41 (Dropout)        (None, 512)               0         \n",
      "                                                                 \n",
      " dense_45 (Dense)            (None, 10)                5130      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 843,658\n",
      "Trainable params: 843,658\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model_cnn = create_model_cnn()\n",
    "# Compile the model\n",
    "model_cnn.compile(optimizer='adam',\n",
    "              loss='sparse_categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "# Summary of the model architecture\n",
    "model_cnn.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "0fc18e51",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "1875/1875 [==============================] - 24s 13ms/step - loss: 0.4450 - accuracy: 0.8365 - val_loss: 0.3326 - val_accuracy: 0.8810\n",
      "Epoch 2/10\n",
      "1875/1875 [==============================] - 23s 13ms/step - loss: 0.3008 - accuracy: 0.8888 - val_loss: 0.2829 - val_accuracy: 0.8921\n",
      "Epoch 3/10\n",
      "1875/1875 [==============================] - 23s 12ms/step - loss: 0.2578 - accuracy: 0.9042 - val_loss: 0.2772 - val_accuracy: 0.8967\n",
      "Epoch 4/10\n",
      "1875/1875 [==============================] - 23s 12ms/step - loss: 0.2256 - accuracy: 0.9153 - val_loss: 0.2520 - val_accuracy: 0.9066\n",
      "Epoch 5/10\n",
      "1875/1875 [==============================] - 23s 12ms/step - loss: 0.1997 - accuracy: 0.9245 - val_loss: 0.2601 - val_accuracy: 0.9064\n",
      "Epoch 6/10\n",
      "1875/1875 [==============================] - 23s 12ms/step - loss: 0.1782 - accuracy: 0.9316 - val_loss: 0.2683 - val_accuracy: 0.9070\n",
      "Epoch 7/10\n",
      "1875/1875 [==============================] - 23s 12ms/step - loss: 0.1596 - accuracy: 0.9395 - val_loss: 0.2710 - val_accuracy: 0.9075\n",
      "Epoch 8/10\n",
      "1875/1875 [==============================] - 23s 12ms/step - loss: 0.1400 - accuracy: 0.9466 - val_loss: 0.2668 - val_accuracy: 0.9116\n",
      "Epoch 9/10\n",
      "1875/1875 [==============================] - 23s 12ms/step - loss: 0.1279 - accuracy: 0.9517 - val_loss: 0.2651 - val_accuracy: 0.9142\n",
      "Epoch 10/10\n",
      "1875/1875 [==============================] - 23s 12ms/step - loss: 0.1119 - accuracy: 0.9574 - val_loss: 0.2912 - val_accuracy: 0.9120\n"
     ]
    }
   ],
   "source": [
    "history = model_cnn.fit(train_x, train_y, batch_size=32, epochs=10, validation_data=(test_x, test_y),\n",
    "                                     shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "6bdbe535",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 1s 4ms/step - loss: 0.2912 - accuracy: 0.9120\n",
      "Accurancy on test data optimizer_CNN 0.9120000004768372\n",
      "Loss on test data optimizer_CNN 0.29122892022132874\n",
      "1875/1875 [==============================] - 8s 4ms/step - loss: 0.0861 - accuracy: 0.9681\n",
      "Accurancy on train data optimizer_CNN: 0.9681166410446167\n",
      "Loss on train data optimizer_CNN: 0.08612586557865143\n"
     ]
    }
   ],
   "source": [
    "# Find accuracy and loss for model SNN\n",
    "test_loss, test_acc = model_cnn.evaluate(test_x, test_y)\n",
    "print('Accurancy on test data optimizer_CNN', test_acc)\n",
    "print('Loss on test data optimizer_CNN', test_loss)\n",
    "\n",
    "train_loss, train_acc = model_cnn.evaluate(train_x, train_y)\n",
    "print('Accurancy on train data optimizer_CNN:', train_acc)\n",
    "print('Loss on train data optimizer_CNN:', train_loss)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "5fb73fcf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize result (accuracy and loss) for CNN optimazer\n",
    "\n",
    "plt.plot(history.history['accuracy'], label='Accurancy on train')\n",
    "plt.plot(history.history['val_accuracy'], label='Accurancy on test')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accurancy')\n",
    "plt.title('CNN optimizer')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.plot(history.history['loss'], label='loss on train')\n",
    "plt.plot(history.history['val_loss'], label='loss on test')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('CNN optimizer')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c14fb4d1",
   "metadata": {},
   "source": [
    "### Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7789cd8",
   "metadata": {},
   "source": [
    "The most efficient model - **CNN**.\n",
    "\n",
    "Over the course of 10 epochs, the model's loss consistently decreases, and its accuracy on the training data steadily increases, indicating that it is learning to fit the training data better. Additionally, the validation loss and accuracy metrics show a similar trend, with the model generalizing well to unseen data. By the end of training, the model achieves an accuracy of approximately **91.2%** on the test set, which is quite high and suggests that it can make accurate predictions on new, previously unseen data. \n",
    "\n",
    "However, the model's performance should also be evaluated on other relevant metrics, and further testing on a diverse range of data should be conducted to ensure its robustness and reliability in real-world applications."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "316418f6",
   "metadata": {},
   "source": [
    "### Test the CNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "8a10e292",
   "metadata": {},
   "outputs": [],
   "source": [
    "# list of class names\n",
    "class_names = [\n",
    "    \"T-shirt/top\",\n",
    "    \"Trouser\",\n",
    "    \"Pullover\",\n",
    "    \"Dress\",\n",
    "    \"Coat\",\n",
    "    \"Sandal\",\n",
    "    \"Shirt\",\n",
    "    \"Sneaker\",\n",
    "    \"Bag\",\n",
    "    \"Ankle boot\"\n",
    "]\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "6123fe88",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 17ms/step\n",
      "Image 1:\n",
      "True Class: Pullover\n",
      "Predicted Class: Pullover\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAANoAAADPCAYAAACTBs/zAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAAnCUlEQVR4nO2de1gV1frHv6DA5uY2FQQUES8kZiliKnjBa0RmUh41MYW8pCHeSE3TX1DHxCw9WCoaIWhHTU1MT3oUSgHNS2qQJh4xxVuBHtO4yUXg/f3Bs/dhZs+G2dvNhg3v53nmj7Vmzax3ZtY7a613XV4zIiIwDFOnmNe3AAzTFGBFYxgjwIrGMEaAFY1hjAArGsMYAVY0hjECrGgMYwRY0RjGCLCiMYwRYEVjGCPAisY0KdLS0jB69Gi4uLjAzMwM3377ba3XpKamwtvbGwqFAp06dcKmTZt0zpcVjWlSFBUVoWfPnli/fr2s9NnZ2XjppZcwaNAgpKen47333sPcuXOxd+9enfI140nFTFPFzMwM+/btQ2BgoNY07777Lg4cOIDLly+r42bNmoVffvkFp06dkp1X8ycRlGGMQUlJCcrKyrSeJyKYmZkJ4qysrGBlZfXEeZ86dQovvPCCIM7f3x9xcXF4/PgxLCwsZN2HFY1p0JSUlMDa2rrGNHZ2digsLBTERUREIDIy8onzz83NRdu2bQVxbdu2RXl5Oe7fvw9nZ2dZ92FFYxo0qppMXGNVp7CwELdv30aLFi3UcYaozVSI81b1tmqSSQwrGmMSmJubSxZsIkJFRQVatGghUDRD4eTkhNzcXEHcvXv30Lx5c7Ru3Vr2fdjqyJgE5ubmWo+6xMfHB8nJyYK4pKQk9OnTR3b/DGBFY0wEMzMzrYcuFBYWIiMjAxkZGQCqzPcZGRm4desWAGDp0qWYMmWKOv2sWbNw8+ZNhIeH4/Lly9iyZQvi4uKwcOFC3R6AGKYBk5eXRwBIoVCQtbW1xqFQKAgA5eXlybrfsWPHCIDGERwcTEREwcHB5OfnJ7gmJSWFvLy8yNLSkjp27EgxMTE6PwePozENmvz8fCiVStjY2Gjtoz169Ah5eXl10kczFGwMYUwCfZqJDQlWNMYkqMnqaAqwojEmASsawxgBbjoyjBEwMzOTHDOrrKysB2l0hxWNMQmMMThdl9SZ5Bs3boS7uzsUCgW8vb1x/PjxusqKaQIYasC6vqiTGm3Xrl2YP38+Nm7ciAEDBmDz5s0ICAhAZmYmOnToUOO1lZWV+OOPP2Bvb28yL5HRHyJCQUEBXFxcaqyxTL1Gq5MB6379+qF3796IiYlRx3l6eiIwMBBRUVE1Xnvnzh24uroaWiSmgXP79m20b99eI141YO3k5KS1j5abm9v0BqzLyspw/vx5LFmyRBD/wgsv4OTJkxrpS0tLUVpaqg6r9F687KGhc+jQIUHY399fI02zZs3qJO/Hjx8LwikpKRpp+vbtKwgrlco6kUVX8vPz4erqCnt7+xrTaWsmmkqrx+CKdv/+fVRUVEgulhMvNwCAqKgofPDBBxrxdbXsoa6wsbERhKVkN5ai2draaqQRy9PQ3m1tCmPqTcc6szpKLZaTeplLly5FeHi4Oqz6w9UFFRUVGnFyCr+4dj59+rRGmps3bwrC0dHRGmn69esnCCsUCkG4Xbt2GteUl5cLwlevXtVIc/78+Rplkcqre/fuGmnEzXoPDw+NNPUFK5qINm3aoFmzZpKL5cS1HGC4vR2Yxo2pNx0N/ouwtLSEt7e3xmK55ORk+Pr6Gjo7polQXws/DUWdNB3Dw8MxefJk9OnTBz4+Pvjiiy9w69YtzJo1qy6yY5oApqRUUtSJok2YMAF//vknPvzwQ+Tk5KBHjx44dOgQ3Nzc6iI7polgKs1EKRrcwk/VuIkhxkXE8+Dk/BH9/Pw04rKysgRhqU1ZxFuiiS2BgGZBEV8jNlgAVdutVSc/P18jjdjII9XnFb+LvLy8Wu/z0UcfaaQJDg7WiHsSavveqvNdunSRNFxVVFTgt99+a3rjaAxTF2hrOjawekIrrGiMSWDqVkdWNMYk4BqtASPnb/fhhx8KwlIDwuI5eOJBZECzTyYn74KCAkH4r7/+0kgj7pdI3VfcJ5NaoyUukG3atNFIU30qHAB89tlnGmkM3UeTi6EVbePGjfjkk0+Qk5ODZ555BtHR0Rg0aJDW9Nu3b8fq1atx9epVKJVKvPjii/j0009lb6JquvZSpklhyGUyqtUly5YtQ3p6OgYNGoSAgAD13o5iTpw4gSlTpmDatGm4dOkS9uzZg7Nnz2L69Omy82RFY0wCQw5Yr127FtOmTcP06dPh6emJ6OhouLq6ClabVOf06dPo2LEj5s6dC3d3dwwcOBAzZ87EuXPn5Muvs5QMUw/Upmj5+fmCQ9wMVqFaXSJ2xaRtdQkA+Pr64s6dOzh06BCICHfv3sU333yDUaNGyZdfdkqGqUdqazq6urpCqVSqD23rHnVdXQJUKdr27dsxYcIEWFpawsnJCS1btsTnn38uW/4mbwzJzMwUhKUGjcWGDqkOuDgvqbzFRormzYWvX85KAilDh3igWSpvcRNLakBdbFR58OCBRhqxl0sfHx/twhqQ2owhurptkru6BKgqI3PnzsX7778Pf39/5OTkYNGiRZg1axbi4uJkyd+oFY1pPNS2C5bc9Yu6ri4BqpYPDRgwAIsWLQIAPPfcc7C1tcWgQYOwYsUKWc4IuenImASGsjrqs7rk0aNHGkquan3IHV7gGo0xCbQ1HfWxOta2umTp0qX4/fffsW3bNgDA6NGjMWPGDMTExKibjvPnz0ffvn3h4uIiK88mr2jisRNxvwnQHKCWckAn/rPJ6cfJOS/OW07BkspbfJ1UXuK+ntR99u3bJwjXdx9NH0WrbXVJTk6OoFyEhISgoKAA69evxzvvvIOWLVti2LBh+Pjjj2Xn2eQVjTENDD3XMTQ0FKGhoZLnEhISNOLmzJmDOXPm6JUXwIrGmAiGrNHqA1Y0xiRgRWMYI8DLZEwIqVXF4vEUqYFOsZFAamBZ/Gc11PINcV5y7iv1lxffR8roU1RUJAhLFWKprfaMAddoDGMEWNEYxghw05FhjIC2KVisaA0Q8YRYoGp6TXXEe+hLpZHqoxnigxuq0EjdR2qQXYy4Lyp1zYULF/QX7AngpiPDGAFuOjKMEeAajWGMACsawxgBbjqaEGfOnKk1jZQPNfEgsdR2c2IDiT5bo0mtnpZzD3GcnLxrWkSpQsroI14cKbU9X9euXTXinhSu0RjGCLCiMYwR4KYjwxgBrtFMiMuXL2vEiSfXSvWT5Pw1xdfJ2QVLjJwJw3JWbkv1reT0/8rKygRhsVspqfwvXryokYb7aJo0KUVjTBtTaSZKwYrGmARcozGMETB1RdNZyrS0NIwePRouLi4wMzPDt99+KzhPRIiMjISLiwusra0xZMgQXLp0yVDyMk0UQ3qTAarcNrm7u0OhUMDb2xvHjx+vMX1paSmWLVsGNzc3WFlZoXPnztiyZYvs/HSu0YqKitCzZ0+8+eabGDt2rMb51atXY+3atUhISICHhwdWrFiBkSNH4sqVK7C3t9c1O4Pyxx9/aMSJZ6hLDUaLDQly/I/JGfjWBzlGDUtLS4048XNJpSksLBSEpVab37t3TxDW5kzC0BiyRlO5bdq4cSMGDBiAzZs3IyAgAJmZmejQoYPkNePHj8fdu3cRFxeHLl264N69e5JlRRs6K1pAQAACAgIkzxERoqOjsWzZMrz22msAgK1bt6Jt27bYsWMHZs6cqXFNaWmp4GNJOUNnmNoUTVxurKystO6/X91tEwBER0fjyJEjiImJkXSOcfjwYaSmpuL69eto1aoVAKBjx466ya9T6lrIzs5Gbm6uwCWOlZUV/Pz8tLrEiYqKEngBcXV1NaRITCPBUN5k9HHbdODAAfTp0werV69Gu3bt4OHhgYULF6K4uFi2/AY1hqg2upFyiXPz5k3Ja5YuXYrw8HB1OD8/n5WN0aC2Gk2uNxl93DZdv34dJ06cgEKhwL59+3D//n2EhobiwYMHsvtpdWJ11MUlTk1VvKGR+gOJP55U30ocJydNfSJVIMV9Kak+mpy+qPg7anNHa2hqUzS53mRU6FJGKysrYWZmhu3bt0OpVAKoan7+7W9/w4YNGyQH9jXklC2ZDJycnABobuFWk0schpGDoayO+rhtcnZ2Rrt27dRKBgCenp4gIty5c0dWvgZVNHd3dzg5OQlc4pSVlSE1NVWrSxyGkYOhfFjr47ZpwIAB+OOPPwRW2aysLJibm6N9+/by5NdJSlSZgDMyMpCRkQGgygCSkZGBW7duwczMDPPnz8fKlSuxb98+/PrrrwgJCYGNjQ2CgoJ0zYph1Kh2wRIf+oyjhYeH48svv8SWLVtw+fJlLFiwQMNt05QpU9Tpg4KC0Lp1a7z55pvIzMxEWloaFi1ahKlTp8pqNgJ69NHOnTuHoUOHCoQGgODgYCQkJGDx4sUoLi5GaGgoHj58iH79+iEpKanex9AY08aQy2R0ddtkZ2eH5ORkzJkzB3369EHr1q0xfvx4rFixQr78ZKi9qw1Efn4+lEol8vLydOrcykFq7ENsFCgpKdFII35FUttpi38kUj6ia0NOoZHzuar3JVSIt/uWeoZr164Jwq1bt9ZIIzYovfLKKxppvvjii1plVFHb91adnz17tqTRrLS0FBs2bKiT8mJIeK4jYxLwwk+GMQKmPqmYFY0xCVjRTIg///xTI07s7FvORFGpwV45iPtX+vTJpAbGxSuqpe4rLpBSzym2oEn1V8X31mVi7ZPATUeGMQJcozGMEWBFYxgjwE1HE0Jquoy4DyQ1TqXPwk85aeT8jQ0lnzgvKTfD06ZNE4RjYmI00shZKFsXcI3GMEaAHREyjBHgpiPDGAFuOjKMEWBFa8CIJ8lKDfaKJ/9KbTyUmpoqCMvZ+UkfF0xyJgxLGR/EA+hSeYuNGFLGEHd3d0G4TZs2GmnEu2DpM3laH7jpyDBGgGs0hjECrGgMYwS46diAycnJEYTl7B5sa2urkUZ8ndTHlTNgLUb8N5Yz0Cz1DGJ5pO4j7sdJ1QQPHjyoNS8xxlo3zDUawxgBVjSGMQKm3nQ0jd8B0+Qx5C5YgO7eZFT8+OOPaN68OXr16qVTfqxojElgqH0dgf95k1m2bBnS09MxaNAgBAQE1Lrrcl5eHqZMmYLhw4frLr/OV5gQjx8/FhyVlZUaR0VFheCwtLTUOFQeb1RHTbvmPonPLin55EBEgkMOUgVW/C7kPJchCr4cDPmuq3uT8fT0RHR0NFxdXSVXK1Rn5syZCAoKgo+Pj855NmpFYxoPtdVo+fn5gkPb7B19vMkAQHx8PK5du4aIiAj95NfrKoYxMrUpmly3Tfp4k7l69SqWLFmC7du3S+6HKQe2OjImQW1WR7lum8TXqdDmTaaiogJBQUH44IMP4OHhoY/oABq5oslxFCcelJXyzSbu94h3nQLqbqWxnInH+kxOlipUYte6UvcRTyI21jiWodw26epNpqCgAOfOnUN6ejrCwsIAVPWliQjNmzdHUlIShg0bVmu+jVrRmMaDoQasq3uTefXVV9XxycnJGDNmjEb6Fi1a4OLFi4K4jRs34ujRo/jmm280VjxogxWNMQkMOWAdHh6OyZMno0+fPvDx8cEXX3yh4U3m999/x7Zt22Bubo4ePXoIrnd0dIRCodCIrwlWNMYkMOQULF29yRgCVjTGJDD05jyhoaEIDQ2VPJeQkFDjtZGRkYiMjNQpv0ataPps59auXTuNNOKPKefjSuVdm8FE6hp9tgSXYzCR2tZcbOiQMvqI4+QOqj8ppj7XsVErGtN44Nn7DGMETF3RdJIyKioKzz//POzt7eHo6IjAwEBcuXJFkIaIEBkZCRcXF1hbW2PIkCG4dOmSQYVmmh6GnldqbHRStNTUVMyePRunT59GcnIyysvL8cILLwjctq5evRpr167F+vXrcfbsWTg5OWHkyJEoKCgwuPC1IWeyq3hCroWFhcYhnmQsnnwr1W+SM0FYnLfUfcWH+Bq51+kzMVpqupM4b30nQuuKIWfv1wc6NR0PHz4sCMfHx8PR0RHnz5/H4MGDQUSIjo7GsmXL8NprrwEAtm7dirZt22LHjh2YOXOm4SRnmhRNqukoRrU3YKtWrQAA2dnZyM3NFcyMtrKygp+fn9aZ0aWlpRozrxlGTJNqOlaHiBAeHo6BAweqR8hV88d0mRkdFRUlmHUtNdeQYUy96ai3lGFhYbhw4QJ27typcU7uzGigarpLXl6e+rh9+7a+IjGNGFNXNL3M+3PmzMGBAweQlpYm8Dnm5OQEoKpmc3Z2VsdrmxkNVDUta1vSoC9SCi9Gjl9pOX6b5cyYr22WvRzf01LoM6Au9Qz3798XhMXbiEvdmwes5aHT74CIEBYWhsTERBw9elRj5rK7uzucnJyQnJysjisrK0Nqaip8fX0NIzHTJGlSNdrs2bOxY8cO7N+/H/b29up+l1KphLW1NczMzDB//nysXLkSXbt2RdeuXbFy5UrY2NggKCioTh6AaRo0KUeEqs1LhgwZIoiPj49HSEgIAGDx4sUoLi5GaGgoHj58iH79+iEpKQn29vYGEZhpmph601EnRZO7clef2c11gT6Tfx89eqSRpqysTBCW2jZcjBwf1uKBbqlJvHK25ZbTF5WD+Dml+mjivp2cVeyGwNTH0XiuI2MSsKIxjBFoUk1HhqkvuEZjGCPAitaAkeNbTNzhl7PKWWoTTXGclEFCbCARGz/kNIOsra014sTXyTGDt2zZUiNN9VUYgPT7Evu1rqtt9sRw05FhjICp12imISXT5DH0zBBd3DYlJiZi5MiRcHBwQIsWLeDj44MjR47oJr9eUjKMkTHkMhld3TalpaVh5MiROHToEM6fP4+hQ4di9OjRSE9Pl51no246ivtAUn0O8W5QUn0r8RIfqTTiPpqUNxNxf0b8N5Ya5C4pKakxHwBQKBQ1hgHAxsZGEM7KytJII2dLbfGz19WEcDGGnIJV3W0TAERHR+PIkSOIiYmRdI4RHR0tCK9cuRL79+/Hv/71L3h5ecnKk2s0xiSob7dN1amsrERBQYF6wbMs+WWnZJh6pLamY126bRKzZs0aFBUVYfz48bLlb9RNR6bxUJvVsa7cNonZuXMnIiMjsX//fjg6OsoRHQArGmMi1Jfbpurs2rUL06ZNw549ezBixAgdpG/kiiY2Ckj95cTGhp49e2qkyc7OFoSvXbumkUY8o1+qjyDeclts/FBtdlTTfaSMIeJBd6nB6Oor4QGgS5cuGmnExiOpdyGWWWxkqSsMNWCtq9smFTt37sTUqVOxc+dOjBo1Sqc8gUauaEzjwZAD1rq4bQKqlGzKlClYt24d+vfvr64Nra2toVQqZeXJisaYBPXptmnz5s0oLy/H7NmzMXv2bHV8cHBwrZ5nVLCiMSaBoec66uK2KSUlRa88qtOoFU3c/xL7aJaKe/jwoUYaT09PQbhjx45PLpwJILWNu/idGgtTn+vYqBWNaTw0qc15GKa+4GUyDGMEuOnIMEaAFa0BIx5wHTZsmEYa8SCtnNnY4oFnQD9/2XWFVHNKzipxMQMGDNCIu3DhgiDcr18/HaXTD246MowR4BqNYYwAK5qBUTVxDOGQUHwPqSafeO6eVL7i65pK01G8czGguXhWalxNl2+nSlvbuzH1pqMZGevry+TOnTvsjLAJcvv2bY2Jz0CVIiqVSly9elXSf0NBQQG6du2KvLw8WbP364sGV6O5uLjg9u3bsLe3R0FBAVxdXTXWGjGGIT8/v97fLxGhoKAALi4uNabjpqOBMTc3V//ZVM0CuWuNGP2o7/crZwa8qTcdG5yiMYwUPAWLYYwANx3rECsrK0RERBhtS7Omhim9X1NvOjY4qyPDVEdldbxz545kPzI/Px/t27dnqyPDGAJTr9FY0RiTgPtoDGMEWNEYxgiYetOxwf4OdHGrw2gnKioKzz//POzt7eHo6IjAwEBcuXJFkIaIEBkZCRcXF1hbW2PIkCG4dOlSPUksTX26bQKA1NRUeHt7Q6FQoFOnTti0aZNuGVID5OuvvyYLCwuKjY2lzMxMmjdvHtna2tLNmzfrWzSTw9/fn+Lj4+nXX3+ljIwMGjVqFHXo0IEKCwvVaVatWkX29va0d+9eunjxIk2YMIGcnZ0pPz+/HiWvIi8vjwDQgwcPqLy8XON48OABAaC8vDzZ99S1fF2/fp1sbGxo3rx5lJmZSbGxsWRhYUHffPON7DwbpKL17duXZs2aJYjr1q0bLVmypJ4kajzcu3ePAFBqaioREVVWVpKTkxOtWrVKnaakpISUSiVt2rSpvsRUo1K0hw8fUkVFhcbx8OFDnRVN1/K1ePFi6tatmyBu5syZ1L9/f9l5NrimoyHc6jDaUW07rnI5lJ2djdzcXMH7trKygp+fX4N634WFhVoPoG7dNp06dUojvb+/P86dOye5ZEqKBqdohnCrw0hDRAgPD8fAgQPRo0cPAP9zsthQ37elpSWcnJw03DKpDldXV9jZ2dWp26bc3FzJ9OXl5bh//76s52iwVkd93eow2gkLC8OFCxdw4sQJjXMN9X0rFApkZ2dLLkJVISWrod02SaWXitdGg1O0J3Grw2hnzpw5OHDgANLS0gQLLJ2cnABU/bWdnZ3V8Q3pfSsUCkl3wfqgT/lycnKSTN+8eXO0bt1aVr4NrulY3a1OdZKTk+Hr61tPUpkuRISwsDAkJibi6NGjcHd3F5x3d3eHk5OT4H2XlZUhNTW1Ub5vfcqXj4+PRvqkpCT06dNHw2WWVmSbTYyIyvwaFxdHmZmZNH/+fLK1taUbN27Ut2gmx9tvv01KpZJSUlIoJydHfTx69EidZtWqVaRUKikxMZEuXrxIEydObDDm/bqgtvK1ZMkSmjx5sjq9yry/YMECyszMpLi4uMZh3ici2rBhA7m5uZGlpSX17t1bbY5mdAOA5BEfH69OU1lZSREREeTk5ERWVlY0ePBgunjxYv0JbQRqKl/BwcHk5+cnSJ+SkkJeXl5kaWlJHTt2pJiYGJ3y42UyDGMEGlwfjWEaI6xoDGMEWNEYxgiwojGMEWBFYxgjwIrGMEaAFY1hjAArGsMYgQataCkpKTAzM8Nff/0FAEhISEDLli3rVaaGSmRkJHr16qUOh4SEIDAw0Ohy3LhxA2ZmZsjIyGgQ92ko1JmiqV6U6njqqacwePBgpKam1lWWDY6QkBD181tYWKBTp05YuHAhioqK6jzvdevWISEhQVba+ijUv/32G9588020b98eVlZWcHd3x8SJE3Hu3DmjyaALqp++6nBwcEBAQAB++eUXWdfXeY32/fffIycnB6mpqWjRogVeeuklZGdn13W2dQYRoby8XHb6F198ETk5Obh+/TpWrFiBjRs3YuHChZJp5a7WlYNSqWywtf+5c+fg7e2NrKwsbN68GZmZmdi3bx+6deuGd955p77Fq5ErV64gJycHBw8exMOHD/Hiiy+qV63XyJAhQ8jOzo7s7e2pd+/edPbsWSIiunHjBr388svUsmVLsrGxoe7du9PBgwdlT6LMzs4mAJSenq6Ou3PnDgGgTZs2SZ5X7f9w7NgxIiI6duyYer8IIqL4+HhSKpWCfDZu3EidOnUiCwsL8vDwoG3btqnPvf766zRhwgRB+rKyMmrdujVt2bKFiKom1H788cfk7u5OCoWCnnvuOdqzZ486vUqGw4cPk7e3N1lYWNDRo0dlvYPg4GAaM2aMIG769Onk5OREREQRERHUs2dPiouLI3d3dzIzM6PKykr666+/aMaMGeTg4ED29vY0dOhQysjIENwnKiqKHB0dyc7OjqZOnUrvvvsu9ezZU2veFRUVtGrVKurcuTNZWlqSq6srrVixgog0Jx5Xn1C7ZcsW6tatG1lZWdHTTz9NGzZsEMhx5swZ6tWrF1lZWZG3tzclJiZqfNfqVFZW0jPPPEPe3t5UUVGhcV71rcXlo7y8nKZOnUodO3YkhUJBHh4eFB0dLbj22LFj9Pzzz5ONjQ0plUry9fVVz8jPyMggbWVdDuKySER04sQJddmQOp+enk4AKDs7m5q3b98eMTExaNasGTIyMtTra2bPno2ysjKkpaXB1tYWmZmZsLOzUytox44dERISgsjISNl/AxsbGwCG+3Pv27cP8+bNQ3R0NEaMGIHvvvtO3RwZOnQoJk2ahPHjx6OwsFAt+5EjR1BUVISxY8cCAJYvX47ExETExMSga9euSEtLwxtvvAEHBwf4+fmp81q8eDE+/fRTdOrU6YlqCmtra8Hz//bbb9i9ezf27t2LZs2aAQBGjRqFVq1a4dChQ1Aqldi8eTOGDx+OrKwstGrVCrt370ZERAQ2bNiAQYMG4auvvsJnn32GTp06ac136dKliI2NxT/+8Q8MHDgQOTk5+M9//gMA+Omnn9C3b198//33eOaZZ2BpaQkAiI2NRUREBNavXw8vLy+kp6djxowZsLW1RXBwMIqKivDyyy9j2LBh+Oc//4ns7GzMmzevxufPyMjApUuXsGPHDsmt4rS928rKSrRv3x67d+9GmzZtcPLkSbz11ltwdnbG+PHjUV5ejsDAQMyYMQM7d+5EWVkZfvrpJ/UK6EmTJsHLy0uyrANVK6Xj4+MREhJSo/zVsba2BlBVnmt1FJKQkCCpwc8++yxFRkZq1fBhw4bR559/rvW8+I9UWFhIM2fOpGbNmtGFCxcMUqP5+vrSjBkzBPmOGzeOXnrpJSKqqr3atGkjqOUmTpxI48aNU8ukUCjo5MmTgntMmzaNJk6cKJDh22+/1fqs2hDXKmfOnKHWrVvT+PHjiaiqRrOwsKB79+6p0/zwww/UokULKikpEdyrc+fOtHnzZiIi8vHx0djFqV+/flprtPz8fLKysqLY2FhJOaW+BRGRq6sr7dixQxD397//nXx8fIiIaPPmzdSqVSsqKipSn4+JiamxRtu1axcBoJ9//lnyfG0yVSc0NJTGjh1LRER//vknAaCUlBTJtPb29qStrBMRPf3005SYmKj1vLgs3r9/n1555RWyt7enu3fv1l6jTZ8+HV999RVGjBiBcePGoXPnzgCAuXPn4u2330ZSUhJGjBiBsWPH4rnnnlMr6A8//CBL6319fWFubo5Hjx7B2dkZCQkJePbZZ3Hjxg25Pw6tXL58GW+99ZYgbsCAAVi3bh0AwMLCAuPGjcP27dsxefJkFBUVYf/+/dixYwcAIDMzEyUlJRg5cqTgHmVlZfDy8hLE9enTRy8Zv/vuO9jZ2aG8vByPHz/GmDFj8Pnnn6vPu7m5wcHBQR0+f/48CgsLNZbIFxcX49q1a+rnnjVrluC8j48Pjh07JinD5cuXUVpaiuHDh8uW+7///S9u376NadOmYcaMGer48vJytYfOy5cvo2fPnuqWikqOmiAd99qozqZNm/Dll1/i5s2bKC4uRllZmdrS2qpVK4SEhMDf3x8jR47EiBEjMH78ePX2DOHh4dBW1gGoa/faUG0DUVRUhK5du2LPnj1wdHREZmZmjdc1v3TpEg4ePIh///vfiIiIwNdff41XX30V06dPh7+/Pw4ePIikpCRERUVhzZo1mDNnjk4vZ9euXejevTtatmwpKDyqZgNVWw6nT5Oytk1WJk2aBD8/P9y7dw/JyclQKBQICAgAUNUcAYCDBw+iXbt2gvuImwK2trY6ywYAQ4cORUxMDCwsLODi4qKx9F1838rKSjg7OyMlJUXjXvo2WVVNHF1QvZvY2Fj069dPcE7VxCU9ljJ6eHgAqFLS6sMRtbF7924sWLAAa9asgY+PD+zt7fHJJ5/gzJkz6jTx8fGYO3cuDh8+jF27dmH58uVITk5G//79ERkZiaCgIMmyrgvHjx9HixYt4ODgIHATVVt5Nvfw8MCCBQuQlJSE1157DfHx8eqTrq6umDVrFhITE/HOO+8gNjZWJ6FU9+jcubPGH1r1F8/JyVHH6Wpe9vT01NjR6eTJk/D09FSHfX194erqil27dmH79u0YN26cug/SvXt3WFlZ4datW+jSpYvgcHV11UkWbdja2qJLly5wc3OTtb9E7969kZubi+bNm2vI1KZNG/Vznz59WnCdOFydrl27wtraWmsrRPU+Kioq1HFt27ZFu3btcP36dQ05VPuOdO/eHb/88guKi4tlyQEAvXr1Qvfu3bFmzRq1MldHNWYq5vjx4/D19UVoaCi8vLzQpUsXdQ1fHS8vLyxduhQnT55Ejx491K0XoErJtZV1ubi7u6Nz584avthqK8/mKSkpuHnzJn788UecPXtWXUjnz5+PI0eOIDs7Gz///DOOHj0qKMDDhw/H+vXrdRZUhbW1Nfr3749Vq1YhMzMTaWlpWL58uU73WLRoERISErBp0yZcvXoVa9euRWJiosB8bmZmhqCgIGzatAnJycl444031Ofs7e2xcOFCLFiwAFu3bsW1a9eQnp6ODRs2YOvWrXo/25MwYsQI+Pj4IDAwEEeOHMGNGzdw8uRJLF++XD3GNG/ePGzZsgVbtmxBVlYWIiIiatwrX6FQ4N1338XixYuxbds2XLt2DadPn0ZcXBwAwNHREdbW1jh8+DDu3r2rNldHRkYiKioK69atQ1ZWFi5evIj4+HisXbsWABAUFARzc3NMmzYNmZmZOHToED799NMan09ldMjKysLgwYNx6NAhXL9+HRcuXMBHH32EMWPGSF7XpUsXnDt3DkeOHEFWVhb+7//+D2fPnlWfz87OxtKlS3Hq1CncvHkTSUlJyMrKgqenJ4qLixEWFgZtZR0AunXrhn379sn4QtKofs6RkZHIysrCwYMHsWbNmv8lcHV1JUtLS3JxcaGwsDAqLi4mIqKwsDDq3LkzWVlZkYODA02ePJnu37+v7ui5ublRRESE1s6jnM5sZmYm9e/fn6ytralXr16UlJRkUPO+ikuXLhEAcnNzo8rKSsG5yspKWrduHT399NNkYWFBDg4O5O/vr95DQqqTW/35VLJKIWXer47KvC8mPz+f5syZQy4uLmRhYUGurq40adIkunXrljrNRx99RG3atCE7OzsKDg6mxYsX12reX7FiBbm5uZGFhQV16NCBVq5cqT4fGxtLrq6uZG5uLjDvb9++nXr16kWWlpb01FNP0eDBgwVGg1OnTlHPnj3J0tKSevXqRXv37q31uxMRXblyhaZMmUIuLi5kaWlJbm5uNHHiRLWRRFx+SkpKKCQkhJRKJbVs2ZLefvttWrJkifqZc3NzKTAwkJydndX3e//996miooJKS0vp9ddfJ21lnYg09lERo60cVOfEiRP07LPPkkKhoEGDBtGePXvUxhDeM0RPUlJS8Oqrr+L69et46qmn6lscpoHToOc6NmQOHz6M9957j5WMkQXXaAxjBLhGYxgjwIrGMEaAFY1hjAArGsMYAVY0hjECrGgMYwRY0RjGCLCiMYwRYEVjGCPw/3c1xwLBJlzkAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 200x200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 20ms/step\n",
      "Image 2:\n",
      "True Class: Shirt\n",
      "Predicted Class: Coat\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 23ms/step\n",
      "Image 3:\n",
      "True Class: Shirt\n",
      "Predicted Class: Shirt\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 18ms/step\n",
      "Image 4:\n",
      "True Class: T-shirt/top\n",
      "Predicted Class: T-shirt/top\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 19ms/step\n",
      "Image 5:\n",
      "True Class: Pullover\n",
      "Predicted Class: Pullover\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "random_indices = np.random.choice(test_x.shape[0], 5, replace=False)\n",
    "\n",
    "# Display each image along with its predicted and true classes\n",
    "for i, idx in enumerate(random_indices):\n",
    "    image = test_images[idx]\n",
    "    true_label = test_labels[idx]\n",
    "\n",
    "    # Predict the class of the image\n",
    "    predictions = model_adam.predict(np.expand_dims(image, axis=0))\n",
    "    predicted_label = np.argmax(predictions)\n",
    "\n",
    "    print(f\"Image {i+1}:\")\n",
    "    print(\"True Class:\", class_names[true_label])\n",
    "    print(\"Predicted Class:\", class_names[predicted_label])\n",
    "\n",
    "    plt.figure(figsize=(2, 2))\n",
    "    plt.imshow(image, cmap=plt.cm.binary)\n",
    "    plt.xlabel(f\"True Class: {class_names[true_label]}, Predicted Class: {class_names[predicted_label]}\")\n",
    "    plt.colorbar()\n",
    "    plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}