model_definer.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import cv2\n",
    "import pickle\n",
    "import os.path\n",
    "import numpy as np\n",
    "from imutils import paths\n",
    "from sklearn.model_selection import train_test_split\n",
    "from keras.models import Sequential\n",
    "from keras.layers.convolutional import Conv2D, MaxPooling2D\n",
    "from keras.layers.core import Flatten, Dense\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import LabelBinarizer\n",
    "\n",
    "from keras.models import load_model\n",
    "import random\n",
    "\n",
    "LABELS_FILENAME = \"model_label.dat\"\n",
    "LETTER_IMAGES_FOLDER = \"extracted_letter_images\"\n",
    "MODEL_FILENAME = \"captcha_model.hdf5\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# start the data and classes array\n",
    "data = []\n",
    "labels = []\n",
    "\n",
    "# check every input image\n",
    "for image_file in paths.list_images(LETTER_IMAGES_FOLDER):    \n",
    "    # load the image and convert it to grayscale\n",
    "    image = cv2.imread(image_file)\n",
    "    image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)\n",
    "\n",
    "    # change the letter size to make is be 20x20\n",
    "    image = cv2.resize(image, (20, 20))\n",
    "    \n",
    "    \n",
    "    # add a third dimension to make a 3D tensor which is the way keras spects images to be\n",
    "    image = np.expand_dims(image, axis=2)\n",
    "    \n",
    "        \n",
    "    # takes the name of the letter from the directory it was in\n",
    "    label = image_file.split(os.path.sep)[-2]\n",
    "\n",
    "    # adds the image and the class to the data and labels a lists\n",
    "    data.append(image)\n",
    "    labels.append(label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# turns the image and the label into numpy arrays and normalizes fields in the image\n",
    "data = np.array(data, dtype=\"float\") / 255.0\n",
    "labels = np.array(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(20, 20)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\olive\\Anaconda3\\lib\\site-packages\\matplotlib\\text.py:1191: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
      "  if s != self._text:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, '[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# devides the data into train and test set\n",
    "(X_train, X_test, Y_train, Y_test) = train_test_split(data, labels, test_size=0.25, random_state=0)\n",
    "\n",
    "# turns the labels into one-hot encoded vectors\n",
    "lb = LabelBinarizer().fit(Y_train)\n",
    "Y_train = lb.transform(Y_train)\n",
    "Y_test = lb.transform(Y_test)\n",
    "\n",
    "# saves the fittet transformer\n",
    "pickle.dump(lb, open(LABELS_FILENAME, 'wb'))\n",
    "\n",
    "# takes away one dimension in order to display the image\n",
    "# result = X_train[50][:, :, 0]\n",
    "\n",
    "# print(result.shape)\n",
    "# plt.imshow(result, cmap = plt.cm.binary)\n",
    "# plt.title(Y_train[50])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\olive\\Anaconda3\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:74: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\olive\\Anaconda3\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:517: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\olive\\Anaconda3\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:4138: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\olive\\Anaconda3\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:3976: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\olive\\Anaconda3\\lib\\site-packages\\keras\\optimizers.py:790: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
      "\n",
      "WARNING:tensorflow:From C:\\Users\\olive\\Anaconda3\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:3295: The name tf.log is deprecated. Please use tf.math.log instead.\n",
      "\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_1 (Conv2D)            (None, 19, 19, 20)        100       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 9, 9, 20)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 8, 8, 50)          4050      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 4, 4, 50)          0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 800)               0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 500)               400500    \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 32)                16032     \n",
      "=================================================================\n",
      "Total params: 420,682\n",
      "Trainable params: 420,682\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Build the Neuralnet!\n",
    "model = Sequential()\n",
    "# Defines the convolutional layers\n",
    "model.add(Conv2D(20, (2,2), input_shape=(20, 20, 1), activation=\"relu\"))\n",
    "model.add(MaxPooling2D(2))\n",
    "\n",
    "model.add(Conv2D(50, (2,2), activation=\"relu\"))\n",
    "model.add(MaxPooling2D(2))\n",
    "\n",
    "model.add(Flatten())\n",
    "model.add(Dense(500, activation=\"relu\"))\n",
    "# output layer with one neuron for each class\n",
    "model.add(Dense(32, activation=\"softmax\"))\n",
    "\n",
    "# compiles the nn\n",
    "model.compile(loss=\"categorical_crossentropy\", \n",
    "              optimizer=\"adam\", \n",
    "              metrics=[\"accuracy\"])\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\olive\\Anaconda3\\lib\\site-packages\\tensorflow\\python\\ops\\math_grad.py:1250: add_dispatch_support.<locals>.wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
      "WARNING:tensorflow:From C:\\Users\\olive\\Anaconda3\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:986: The name tf.assign_add is deprecated. Please use tf.compat.v1.assign_add instead.\n",
      "\n",
      "Train on 23246 samples, validate on 5812 samples\n",
      "Epoch 1/5\n",
      "23246/23246 [==============================] - 20s 880us/step - loss: 0.8746 - acc: 0.8170 - val_loss: 0.0834 - val_acc: 0.9871\n",
      "Epoch 2/5\n",
      "23246/23246 [==============================] - 19s 812us/step - loss: 0.0581 - acc: 0.9894 - val_loss: 0.0426 - val_acc: 0.9917 - loss: 0.0619 - acc: - ET - ETA: 1s - loss: 0.0591 - \n",
      "Epoch 3/5\n",
      "23246/23246 [==============================] - 20s 845us/step - loss: 0.0334 - acc: 0.9934 - val_loss: 0.0331 - val_acc: 0.9935\n",
      "Epoch 4/5\n",
      "23246/23246 [==============================] - 19s 806us/step - loss: 0.0214 - acc: 0.9951 - val_loss: 0.0279 - val_acc: 0.9940\n",
      "Epoch 5/5\n",
      "23246/23246 [==============================] - 19s 816us/step - loss: 0.0138 - acc: 0.9969 - val_loss: 0.0215 - val_acc: 0.9957\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train, Y_train, epochs=5, validation_split=0.2, batch_size=128, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plots the model's performance\n",
    "acc = history.history['acc']\n",
    "val_acc = history.history[\"val_acc\"]\n",
    "epochs = range(len(acc))\n",
    "\n",
    "plt.plot(epochs, acc, 'b', label='Training Accuracy')\n",
    "plt.plot(epochs, val_acc, 'r', label='Validation accuracy')\n",
    "plt.title('Model accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9686/9686 [==============================] - 3s 290us/step\n"
     ]
    }
   ],
   "source": [
    "loss, accuracy = model.evaluate(X_test, Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9947346685938467"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# accuracy should be over 95% pretty good!\n",
    "accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Saves the model for latter use\n",
    "model.save(MODEL_FILENAME)"
   ]
  }
 ],
 "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}