TF tutorial 4.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Neural Networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#imports\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Dataset\n",
    "\n",
    "We will be using the MNIST fashion dataset from Keras\n",
    "\n",
    "It contains 70k images:\n",
    "- 60k for training\n",
    "- 10k for testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#load in the dataset, then split into train/test\n",
    "fashion = keras.datasets.fashion_mnist\n",
    "\n",
    "(train_images, train_labels), (test_images, test_labels) = fashion.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(60000, 28, 28)\n",
      "(10000, 28, 28)\n"
     ]
    }
   ],
   "source": [
    "# Look at shape of data\n",
    "\n",
    "print(train_images.shape)\n",
    "print(test_images.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This shows we have 70k images in totalm are they are 28 by 28 pixels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  0   0   0   0   0   1   0   0   0   0  41 188 103  54  48  43  87 168\n",
      "  133  16   0   0   0   0   0   0   0   0]\n",
      " [  0   0   0   1   0   0   0  49 136 219 216 228 236 255 255 255 255 217\n",
      "  215 254 231 160  45   0   0   0   0   0]\n",
      " [  0   0   0   0   0  14 176 222 224 212 203 198 196 200 215 204 202 201\n",
      "  201 201 209 218 224 164   0   0   0   0]\n",
      " [  0   0   0   0   0 188 219 200 198 202 198 199 199 201 196 198 198 200\n",
      "  200 200 200 201 200 225  41   0   0   0]\n",
      " [  0   0   0   0  51 219 199 203 203 212 238 248 250 245 249 246 247 252\n",
      "  248 235 207 203 203 222 140   0   0   0]\n",
      " [  0   0   0   0 116 226 206 204 207 204 101  75  47  73  48  50  45  51\n",
      "   63 113 222 202 206 220 224   0   0   0]\n",
      " [  0   0   0   0 200 222 209 203 215 200   0  70  98   0 103  59  68  71\n",
      "   49   0 219 206 214 210 250  38   0   0]\n",
      " [  0   0   0   0 247 218 212 210 215 214   0 254 243 139 255 174 251 255\n",
      "  205   0 215 217 214 208 220  95   0   0]\n",
      " [  0   0   0  45 226 214 214 215 224 205   0  42  35  60  16  17  12  13\n",
      "   70   0 189 216 212 206 212 156   0   0]\n",
      " [  0   0   0 164 235 214 211 220 216 201  52  71  89  94  83  78  70  76\n",
      "   92  87 206 207 222 213 219 208   0   0]\n",
      " [  0   0   0 106 187 223 237 248 211 198 252 250 248 245 248 252 253 250\n",
      "  252 239 201 212 225 215 193 113   0   0]\n",
      " [  0   0   0   0   0  17  54 159 222 193 208 192 197 200 200 200 200 201\n",
      "  203 195 210 165   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0  47 225 192 214 203 206 204 204 205 206 204\n",
      "  212 197 218 107   0   0   0   0   0   0]\n",
      " [  0   0   0   0   1   6   0  46 212 195 212 202 206 205 204 205 206 204\n",
      "  212 200 218  91   0   3   1   0   0   0]\n",
      " [  0   0   0   0   0   1   0  11 197 199 205 202 205 206 204 205 207 204\n",
      "  205 205 218  77   0   5   0   0   0   0]\n",
      " [  0   0   0   0   0   3   0   2 191 198 201 205 206 205 205 206 209 206\n",
      "  199 209 219  74   0   5   0   0   0   0]\n",
      " [  0   0   0   0   0   2   0   0 188 197 200 207 207 204 207 207 210 208\n",
      "  198 207 221  72   0   4   0   0   0   0]\n",
      " [  0   0   0   0   0   2   0   0 215 198 203 206 208 205 207 207 210 208\n",
      "  200 202 222  75   0   4   0   0   0   0]\n",
      " [  0   0   0   0   0   1   0   0 212 198 209 206 209 206 208 207 211 206\n",
      "  205 198 221  80   0   3   0   0   0   0]\n",
      " [  0   0   0   0   0   1   0   0 204 201 205 208 207 205 211 205 210 210\n",
      "  209 195 221  96   0   3   0   0   0   0]\n",
      " [  0   0   0   0   0   1   0   0 202 201 205 209 207 205 213 206 210 209\n",
      "  210 194 217 105   0   2   0   0   0   0]\n",
      " [  0   0   0   0   0   1   0   0 204 204 205 208 207 205 215 207 210 208\n",
      "  211 193 213 115   0   2   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   0 204 207 207 208 206 206 215 210 210 207\n",
      "  212 195 210 118   0   2   0   0   0   0]\n",
      " [  0   0   0   0   0   1   0   0 198 208 208 208 204 207 212 212 210 207\n",
      "  211 196 207 121   0   1   0   0   0   0]\n",
      " [  0   0   0   0   0   1   0   0 198 210 207 208 206 209 213 212 211 207\n",
      "  210 197 207 124   0   1   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   0 172 210 203 201 199 204 207 205 204 201\n",
      "  205 197 206 127   0   0   0   0   0   0]\n",
      " [  0   0   0   0   0   0   0   0 188 221 214 234 236 238 244 244 244 240\n",
      "  243 214 224 162   0   2   0   0   0   0]\n",
      " [  0   0   0   0   0   1   0   0 139 146 130 135 135 137 125 124 125 121\n",
      "  119 114 130  76   0   0   0   0   0   0]]\n"
     ]
    }
   ],
   "source": [
    "print(train_images[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The pixels seem to be between 0 and 255 which would suggest a greyscale\n",
    "\n",
    "0 equates to a black pixel, whilst 1 equates to a white pixel and anything inbetween is grey\n",
    "\n",
    "This means there are no colour channels\n",
    "\n",
    "Next we will look at the labels for the images:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([9, 0, 0, 3, 0, 2, 7, 2, 5, 5, 0, 9, 5, 5, 7], dtype=uint8)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_labels[:15]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have labels between 0 and 9, which corrsepond to different items of clothing\n",
    "\n",
    "Using the datasets documentation we can map the number to it's label by cretaing an array of label names:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "label_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', \n",
    "               'Coat', 'Sandal', 'Shirt', 'Sneaker', 'Bag', \n",
    "               'Ankle boot']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using matplotlib, we can take a look at what one of the images looks like"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.imshow(train_images[3])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The third image looks like a dress!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Preprocessing\n",
    "\n",
    "Neural networks work faster with smaller values, so if we normalise the pixels to be between 0 and 1 rather than between 0 and 255, we should be bale to get a more efficent model\n",
    "\n",
    "A neural network will start with biases and weights that are alrady inbetween 0 and 1 so making our values close to this will help"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_images = train_images / 255.0\n",
    "\n",
    "test_images = test_images / 255.0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Building the model\n",
    "\n",
    "We will use a sequential mode, which is just the most basic of NN's, information goes from left to right sequentially\n",
    "\n",
    "It is as simple as defining layers in network within the sequential function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\sinea\\AppData\\Local\\Programs\\Python\\Python39\\lib\\site-packages\\keras\\src\\layers\\reshaping\\flatten.py:37: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(**kwargs)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "model = keras.Sequential([\n",
    "    keras.layers.Flatten(input_shape=(28, 28)),  # input layer is in the same shape as our image\n",
    "                                                    #flatten will turn the 28x28 image into an array of length 784 (flattening the matrix)\n",
    "\n",
    "    keras.layers.Dense(128, activation = 'relu'),  #First hidden layer, dense means fully connected, we picked 128 randomly, find through experimenting\n",
    "                                                    #relu makes all numbers below 0 equal to 0\n",
    "\n",
    "    keras.layers.Dense(10, activation = 'softmax')  #output layer has 10 neurons, one for each class 0-9\n",
    "                                                    #softmax ensures all our ouputs combined add up to 1\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"sequential\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ flatten (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">784</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)            │       <span style=\"color: #00af00; text-decoration-color: #00af00\">100,480</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">1,290</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m784\u001b[0m)            │             \u001b[38;5;34m0\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense (\u001b[38;5;33mDense\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)            │       \u001b[38;5;34m100,480\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)             │         \u001b[38;5;34m1,290\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">101,770</span> (397.54 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m101,770\u001b[0m (397.54 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">101,770</span> (397.54 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m101,770\u001b[0m (397.54 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Compile the model\n",
    "\n",
    "This is where we choose the optimiser (the function that will perform gradient descent)\n",
    "\n",
    "loss function (how we calculate how good/bad the ouput is compared to the actual class) \n",
    "\n",
    "and metric we will use to evaulate (things like accuracy, precision, recall etc)\n",
    "\n",
    "To choose the most appropriate ones, go to Keras documentation and see what they recommend, and experiment a bit!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer= 'adam',\n",
    "              loss = 'sparse_categorical_crossentropy',\n",
    "              metrics = ['accuracy'])\n",
    "\n",
    "#changing these to get best results is hyperparameter tuning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Training the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.7891 - loss: 0.6162\n",
      "Epoch 2/10\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.8622 - loss: 0.3834\n",
      "Epoch 3/10\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.8765 - loss: 0.3460\n",
      "Epoch 4/10\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2ms/step - accuracy: 0.8830 - loss: 0.3146\n",
      "Epoch 5/10\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.8893 - loss: 0.2950\n",
      "Epoch 6/10\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.8959 - loss: 0.2805\n",
      "Epoch 7/10\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.9014 - loss: 0.2685\n",
      "Epoch 8/10\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.9049 - loss: 0.2561\n",
      "Epoch 9/10\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.9055 - loss: 0.2465\n",
      "Epoch 10/10\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.9102 - loss: 0.2375\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.src.callbacks.history.History at 0x28d70f7bbe0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(train_images, train_labels, epochs=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Testing the model\n",
    "\n",
    "Above we see we got a final accuracy of 91%, but this isn't the true accuracy as we haven't used the model on unseen data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 898us/step - accuracy: 0.8821 - loss: 0.3428\n",
      "Test Accuracy: 0.8823999762535095\n"
     ]
    }
   ],
   "source": [
    "test_loss, test_acc = model.evaluate(test_images, test_labels, verbose=1)\n",
    "\n",
    "print('Test Accuracy:', test_acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We got an accuracy of 88% on unseen data, which is a good starting point to improve on\n",
    "\n",
    "We have a lower accuracy than on the training data as we have some overfitting, we can adjust this with hyperparamer tuning\n",
    "\n",
    "Let's try with a lower epoch:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/2\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.9262 - loss: 0.1966\n",
      "Epoch 2/2\n",
      "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 1ms/step - accuracy: 0.9271 - loss: 0.1921\n",
      "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 936us/step - accuracy: 0.8857 - loss: 0.3525\n",
      "Test Accuracy: 0.886900007724762\n"
     ]
    }
   ],
   "source": [
    "model.fit(train_images, train_labels, epochs=2)\n",
    "\n",
    "test_loss, test_acc = model.evaluate(test_images, test_labels, verbose=1)\n",
    "\n",
    "print('Test Accuracy:', test_acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We only imrpvoed by about 0.4% but it is something!\n",
    "\n",
    "Generally a lower epoch will be better as the data will be seen less and the model won't tune to specific images as much"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Making predictions\n",
    "\n",
    "Neural networks work better at making many predictions at once rather than just one, so usuually you would want to predict with an array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step  \n"
     ]
    }
   ],
   "source": [
    "predictions = model.predict(test_images)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at the predictions on the first image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([8.4482528e-09, 8.0702589e-12, 8.3643361e-12, 1.8321981e-11,\n",
       "       9.0247315e-10, 1.7849190e-04, 8.0890928e-08, 1.5342067e-02,\n",
       "       8.6642954e-10, 9.8447925e-01], dtype=float32)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This would predict the 9th class as this has the highest probability in the distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<function matplotlib.pyplot.show(close=None, block=None)>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.bar(np.arange(len(predictions[0])), predictions[0])\n",
    "plt.show"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The chart above shows that the 9th class dominates the prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ankle boot\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(label_names[np.argmax(predictions[0])])\n",
    "\n",
    "plt.figure()\n",
    "plt.imshow(test_images[0])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Verifying predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Try again\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 19ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def predict(model, image, correct_label):\n",
    "    label_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', \n",
    "               'Coat', 'Sandal', 'Shirt', 'Sneaker', 'Bag', \n",
    "               'Ankle boot']\n",
    "    \n",
    "    prediction = model.predict(np.array([image]))\n",
    "    predicted_class = label_names[np.argmax(prediction)]\n",
    "\n",
    "    show_image(image, label_names[correct_label], predicted_class)\n",
    "\n",
    "\n",
    "def show_image(img, label, guess):\n",
    "    plt.figure()\n",
    "    plt.imshow(img)\n",
    "    plt.title('Expected: ' + label)\n",
    "    plt.xlabel('Guess: ' + guess)\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "def get_number():\n",
    "    while True:\n",
    "        num = input('Pick a number: ')\n",
    "        if num.isdigit():\n",
    "            num = int(num)\n",
    "            if 0 <= num <= 1000:\n",
    "                return int(num)\n",
    "        else:\n",
    "            print('Try again')\n",
    "\n",
    "\n",
    "num = get_number()\n",
    "image = test_images[num]\n",
    "label = test_labels[num]\n",
    "predict(model, image, label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.9.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}