test_human.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import PIL\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "proposals = pd.read_csv('proposals.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "def invert_image(input_image):\n",
    "    \"\"\"inverts the input image color values. Returns image\n",
    "    of same dim.\"\"\"\n",
    "    if not type(input_image) == type(np.array([0])):\n",
    "        input_image = np.array(input_image)/256\n",
    "    return (input_image - 256) * -1 \n",
    "\n",
    "def normalize_image(input_image):\n",
    "    \"\"\"Takes a PIL image and \"normalizes\" its pixel values;\n",
    "    returns a numpy array of same shape, with min value 0 and\n",
    "    max value 256.\n",
    "    \"\"\"\n",
    "    if not type(input_image) == type(np.array([0])):\n",
    "        array = np.array(input_image)/256\n",
    "    else:\n",
    "        array = input_image/256\n",
    "    min_ = np.min(np.min(array))\n",
    "    array = array - min_\n",
    "    max_ = np.max(np.max(array))\n",
    "    array = array * 1/max_ * 256\n",
    "    return array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_image(id_number):\n",
    "    \"\"\"Takes id number and returns an image.\"\"\"\n",
    "    path = './tp_images/' + id_number + '.bmp'\n",
    "    try:\n",
    "        img = PIL.Image.open(path)\n",
    "    except:\n",
    "        try:\n",
    "            path = './fp_images/' + id_number + '.bmp'\n",
    "            img = PIL.Image.open(path)\n",
    "        except:\n",
    "            raise('Error: No file associated with ', id_number)\n",
    "    return np.array(img)\n",
    "\n",
    "def remove_ticks(ax_obj):\n",
    "    \"\"\"takes an ax object from matplotlib and removes ticks.\"\"\"\n",
    "    ax_obj.tick_params(\n",
    "        axis='both', \n",
    "        which='both', \n",
    "        bottom='off', \n",
    "        top='off', \n",
    "        labelbottom='off', \n",
    "        right='off', \n",
    "        left='off', \n",
    "        labelleft='off'\n",
    "        )\n",
    "    return ax_obj\n",
    "\n",
    "def show_example(id_number):\n",
    "    \"\"\"Takes example id number and shows it for user inspection.\"\"\"\n",
    "    img = get_image(id_number)\n",
    "    #Plot \"zoomed in\"\n",
    "    fig, ax = plt.subplots(nrows=1, ncols=2)\n",
    "    ax[0].set_title('natural')\n",
    "    ax[0].imshow(img, cmap='Greys')\n",
    "    ax[0] = remove_ticks(ax[0])\n",
    "    ax[1].set_title('inverted')\n",
    "    ax[1].imshow(invert_image(img), cmap='Greys')\n",
    "    ax[1] = remove_ticks(ax[1])\n",
    "    plt.tight_layout()\n",
    "    \n",
    "    #Plot \"actual size\"\n",
    "    fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(2, 28/80))\n",
    "    ax[0].set_title('natural')\n",
    "    ax[0].imshow(img, cmap='Greys')\n",
    "    ax[0] = remove_ticks(ax[0])\n",
    "    ax[1].set_title('inverted')\n",
    "    ax[1].imshow(invert_image(img), cmap='Greys')\n",
    "    ax[1] = remove_ticks(ax[1])\n",
    "    return"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa7ad7b0320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa7ad516a58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_example(proposals['id'].loc[10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4910"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(proposals)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Environment (conda_ann)",
   "language": "python",
   "name": "conda_ann"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}