{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def read_images(csv_file):\n",
    "    # Returns a df with the images in it\n",
    "    \n",
    "    import pandas as pd\n",
    "    from skimage import io\n",
    "    \n",
    "    df = pd.read_csv(csv_file)\n",
    "    read_images = []\n",
    "    for file in df['image_path']:\n",
    "        read_images.append(io.imread(file))\n",
    "\n",
    "    df = df.assign(image = read_images)\n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_hog(images, canny = False):\n",
    "    # Takes a list of images, returns a list of the hog descriptors\n",
    "    \n",
    "    from skimage.feature import hog\n",
    "\n",
    "    hog_features = []\n",
    "    \n",
    "    if canny: # we need a special case for canny hog because we have to specify the image as type int\n",
    "        for image in images:\n",
    "            hog_features.append(hog(image.astype(int),\n",
    "                                    orientations = 8,\n",
    "                                    pixels_per_cell = (40, 40),\n",
    "                                    visualize = False)\n",
    "                               )\n",
    "    else:\n",
    "        for image in images:\n",
    "            hog_features.append(hog(image,\n",
    "                                    orientations = 8,\n",
    "                                    pixels_per_cell = (40, 40),\n",
    "                                    visualize = False)\n",
    "                               )\n",
    "    \n",
    "    return hog_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "def classify_images(df, random_state = 0):\n",
    "    # From a df, performs image classification and returns the results\n",
    "    # Takes random_state as an input to allow different reproducible results\n",
    "    \n",
    "    import numpy as np\n",
    "    from sklearn.ensemble import RandomForestClassifier\n",
    "    from sklearn.model_selection import train_test_split\n",
    "    \n",
    "    x = ['hog_features', 'image', 'image_path']\n",
    "    train, test, y_train, y_test = train_test_split(df[x],\n",
    "                                                    df['label'],\n",
    "                                                    test_size = 0.2,\n",
    "                                                    random_state = random_state)\n",
    "    \n",
    "    x_train = np.stack(train['hog_features'].values)\n",
    "    x_test = np.stack(test['hog_features'].values)\n",
    "    \n",
    "    random_forest = RandomForestClassifier(n_estimators = 10, max_depth = 7, random_state = random_state)\n",
    "    random_forest.fit(x_train, y_train.values)\n",
    "    predictions = random_forest.predict(x_test)\n",
    "    \n",
    "    return predictions, y_test, test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def print_results(predictions, test):\n",
    "    \n",
    "    import matplotlib.pyplot as plt\n",
    "    import pandas as pd\n",
    "    from skimage import io\n",
    "\n",
    "    i = 0\n",
    "    l = len(test)\n",
    "    \n",
    "    df_test = pd.DataFrame(test)\n",
    "    \n",
    "    for index, row in df_test.iterrows():\n",
    "        plt.subplot(1, l, i+1)\n",
    "        plt.imshow(io.imread(row['image_path']))\n",
    "        plt.title(str(predictions[i]))\n",
    "        i = i + 1\n",
    "    plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}