epynnlive/dummy_image/prepare_dataset.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dummy dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Find this notebook at `EpyNN/epynnlive/dummy_image/prepare_dataset.ipynb`. \n",
    "* Regular python code at `EpyNN/epynnlive/dummy_image/prepare_dataset.py`.\n",
    "\n",
    "Run the notebook online with [Google Colab](https://colab.research.google.com/github/Synthaze/EpyNN/blob/main/epynnlive/dummy_image/prepare_dataset.ipynb).\n",
    "\n",
    "**Level: Intermediate**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook is part of the series on preparing data for Neural Network regression with EpyNN.\n",
    "\n",
    "In addition to the topic-specific content, it contains several explanations about basics or general concepts in programming that are important in the context.\n",
    "\n",
    "Note that elements developed in the following notebooks may not be reviewed herein:\n",
    "\n",
    "* [Boolean dataset](../dummy_boolean/prepare_dataset.ipynb)\n",
    "* [String dataset](../dummy_string/prepare_dataset.ipynb)\n",
    "* [Time-series (numerical)](../dummy_time/prepare_dataset.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What is an image?\n",
    "\n",
    "Instinctively, an image may resemble a 2D plane composed of **WIDTH * HEIGHT** colored units arranged together in a particular manner.\n",
    "\n",
    "In computing, a 2D image is generally a 3D object which is composed of **WIDTH * HEIGHT** units within each plane with respect the the third dimension, which is the **DEPTH** of the image, therefore giving **WIDTH * HEIGHT * DEPTH = N_FEATURES**.\n",
    "\n",
    "Image depth is simply the number of channels which compose the image. You are certainly aware of RGB colors, for instance. In the RGB scheme, one color is written such as ``rgb(int, int, int)`` or ``rgb(255, 0, 0)``, ``rgb(0, 255, 0)`` and ``rgb(0, 0, 255)`` for pure red, green and blue, respectively. One RGB image would therefore have a **DEPTH** equal to 3, because of the three channels within.\n",
    "\n",
    "Note that following this scheme, an image is made of Numerical data, namely integer or ``int``. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Why preparing a dummy dataset of images?\n",
    "\n",
    "In addition to general considerations [reviewed here](../dummy_boolean/prepare_dataset.ipynb#Why-preparing-a-dummy-dataset-with-Boolean-features), this may be a good idea to practically understand what an image is, how to build an image, and how to handle such kind of data overall."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prepare a set of image sample features and related label"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is no specific import for this notebook since we will create images from scratch."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: epynn in /home/synthase/.local/lib/python3.9/site-packages (1.2.6)\n",
      "Requirement already satisfied: numpy==1.21.2 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (1.21.2)\n",
      "Requirement already satisfied: texttable==1.6.4 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (1.6.4)\n",
      "Requirement already satisfied: Pygments==2.10.0 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (2.10.0)\n",
      "Requirement already satisfied: Pillow==8.3.1 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (8.3.1)\n",
      "Requirement already satisfied: cycler==0.10.0 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (0.10.0)\n",
      "Requirement already satisfied: kiwisolver==1.3.2 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (1.3.2)\n",
      "Requirement already satisfied: nbconvert==5.4.1 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (5.4.1)\n",
      "Requirement already satisfied: pyparsing==2.4.7 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (2.4.7)\n",
      "Requirement already satisfied: jupyter==1.0.0 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (1.0.0)\n",
      "Requirement already satisfied: termcolor==1.1.0 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (1.1.0)\n",
      "Requirement already satisfied: utilsovs-pkg in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (0.9.4)\n",
      "Requirement already satisfied: python-dateutil==2.8.2 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (2.8.2)\n",
      "Requirement already satisfied: tabulate==0.8.9 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (0.8.9)\n",
      "Requirement already satisfied: six==1.16.0 in /usr/lib/python3/dist-packages (from epynn) (1.16.0)\n",
      "Requirement already satisfied: wget==3.2 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (3.2)\n",
      "Requirement already satisfied: matplotlib==3.4.3 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (3.4.3)\n",
      "Requirement already satisfied: scipy==1.6.3 in /home/synthase/.local/lib/python3.9/site-packages (from epynn) (1.6.3)\n",
      "Requirement already satisfied: notebook in /home/synthase/.local/lib/python3.9/site-packages (from jupyter==1.0.0->epynn) (6.4.11)\n",
      "Requirement already satisfied: qtconsole in /home/synthase/.local/lib/python3.9/site-packages (from jupyter==1.0.0->epynn) (5.3.0)\n",
      "Requirement already satisfied: jupyter-console in /home/synthase/.local/lib/python3.9/site-packages (from jupyter==1.0.0->epynn) (6.4.3)\n",
      "Requirement already satisfied: ipywidgets in /home/synthase/.local/lib/python3.9/site-packages (from jupyter==1.0.0->epynn) (7.7.0)\n",
      "Requirement already satisfied: ipykernel in /home/synthase/.local/lib/python3.9/site-packages (from jupyter==1.0.0->epynn) (6.13.0)\n",
      "Requirement already satisfied: bleach in /home/synthase/.local/lib/python3.9/site-packages (from nbconvert==5.4.1->epynn) (4.1.0)\n",
      "Requirement already satisfied: entrypoints>=0.2.2 in /home/synthase/.local/lib/python3.9/site-packages (from nbconvert==5.4.1->epynn) (0.4)\n",
      "Requirement already satisfied: pandocfilters>=1.4.1 in /home/synthase/.local/lib/python3.9/site-packages (from nbconvert==5.4.1->epynn) (1.5.0)\n",
      "Requirement already satisfied: testpath in /home/synthase/.local/lib/python3.9/site-packages (from nbconvert==5.4.1->epynn) (0.6.0)\n",
      "Requirement already satisfied: mistune>=0.8.1 in /home/synthase/.local/lib/python3.9/site-packages (from nbconvert==5.4.1->epynn) (0.8.4)\n",
      "Requirement already satisfied: defusedxml in /home/synthase/.local/lib/python3.9/site-packages (from nbconvert==5.4.1->epynn) (0.7.1)\n",
      "Requirement already satisfied: nbformat>=4.4 in /home/synthase/.local/lib/python3.9/site-packages (from nbconvert==5.4.1->epynn) (5.3.0)\n",
      "Requirement already satisfied: jinja2 in /home/synthase/.local/lib/python3.9/site-packages (from nbconvert==5.4.1->epynn) (3.0.3)\n",
      "Requirement already satisfied: jupyter-core in /home/synthase/.local/lib/python3.9/site-packages (from nbconvert==5.4.1->epynn) (4.10.0)\n",
      "Requirement already satisfied: traitlets>=4.2 in /home/synthase/.local/lib/python3.9/site-packages (from nbconvert==5.4.1->epynn) (5.1.1)\n",
      "Requirement already satisfied: fastjsonschema in /home/synthase/.local/lib/python3.9/site-packages (from nbformat>=4.4->nbconvert==5.4.1->epynn) (2.15.3)\n",
      "Requirement already satisfied: jsonschema>=2.6 in /usr/lib/python3/dist-packages (from nbformat>=4.4->nbconvert==5.4.1->epynn) (3.2.0)\n",
      "Requirement already satisfied: packaging in /home/synthase/.local/lib/python3.9/site-packages (from bleach->nbconvert==5.4.1->epynn) (21.3)\n",
      "Requirement already satisfied: webencodings in /home/synthase/.local/lib/python3.9/site-packages (from bleach->nbconvert==5.4.1->epynn) (0.5.1)\n",
      "Requirement already satisfied: ipython>=7.23.1 in /home/synthase/.local/lib/python3.9/site-packages (from ipykernel->jupyter==1.0.0->epynn) (8.2.0)\n",
      "Requirement already satisfied: debugpy>=1.0 in /home/synthase/.local/lib/python3.9/site-packages (from ipykernel->jupyter==1.0.0->epynn) (1.6.0)\n",
      "Requirement already satisfied: jupyter-client>=6.1.12 in /home/synthase/.local/lib/python3.9/site-packages (from ipykernel->jupyter==1.0.0->epynn) (7.2.2)\n",
      "Requirement already satisfied: psutil in /usr/lib/python3/dist-packages (from ipykernel->jupyter==1.0.0->epynn) (5.8.0)\n",
      "Requirement already satisfied: matplotlib-inline>=0.1 in /home/synthase/.local/lib/python3.9/site-packages (from ipykernel->jupyter==1.0.0->epynn) (0.1.3)\n",
      "Requirement already satisfied: nest-asyncio in /home/synthase/.local/lib/python3.9/site-packages (from ipykernel->jupyter==1.0.0->epynn) (1.5.5)\n",
      "Requirement already satisfied: tornado>=6.1 in /home/synthase/.local/lib/python3.9/site-packages (from ipykernel->jupyter==1.0.0->epynn) (6.1)\n",
      "Requirement already satisfied: backcall in /home/synthase/.local/lib/python3.9/site-packages (from ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (0.2.0)\n",
      "Requirement already satisfied: decorator in /home/synthase/.local/lib/python3.9/site-packages (from ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (5.1.1)\n",
      "Requirement already satisfied: jedi>=0.16 in /home/synthase/.local/lib/python3.9/site-packages (from ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (0.18.1)\n",
      "Requirement already satisfied: stack-data in /home/synthase/.local/lib/python3.9/site-packages (from ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (0.2.0)\n",
      "Requirement already satisfied: pickleshare in /home/synthase/.local/lib/python3.9/site-packages (from ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (0.7.5)\n",
      "Requirement already satisfied: setuptools>=18.5 in /usr/lib/python3/dist-packages (from ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (52.0.0)\n",
      "Requirement already satisfied: prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0 in /home/synthase/.local/lib/python3.9/site-packages (from ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (3.0.29)\n",
      "Requirement already satisfied: pexpect>4.3 in /home/synthase/.local/lib/python3.9/site-packages (from ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (4.8.0)\n",
      "Requirement already satisfied: parso<0.9.0,>=0.8.0 in /home/synthase/.local/lib/python3.9/site-packages (from jedi>=0.16->ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (0.8.3)\n",
      "Requirement already satisfied: pyzmq>=22.3 in /home/synthase/.local/lib/python3.9/site-packages (from jupyter-client>=6.1.12->ipykernel->jupyter==1.0.0->epynn) (22.3.0)\n",
      "Requirement already satisfied: ptyprocess>=0.5 in /home/synthase/.local/lib/python3.9/site-packages (from pexpect>4.3->ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (0.7.0)\n",
      "Requirement already satisfied: wcwidth in /home/synthase/.local/lib/python3.9/site-packages (from prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0->ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (0.2.5)\n",
      "Requirement already satisfied: widgetsnbextension~=3.6.0 in /home/synthase/.local/lib/python3.9/site-packages (from ipywidgets->jupyter==1.0.0->epynn) (3.6.0)\n",
      "Requirement already satisfied: jupyterlab-widgets>=1.0.0 in /home/synthase/.local/lib/python3.9/site-packages (from ipywidgets->jupyter==1.0.0->epynn) (1.1.0)\n",
      "Requirement already satisfied: ipython-genutils~=0.2.0 in /home/synthase/.local/lib/python3.9/site-packages (from ipywidgets->jupyter==1.0.0->epynn) (0.2.0)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: terminado>=0.8.3 in /home/synthase/.local/lib/python3.9/site-packages (from notebook->jupyter==1.0.0->epynn) (0.13.3)\n",
      "Requirement already satisfied: argon2-cffi in /home/synthase/.local/lib/python3.9/site-packages (from notebook->jupyter==1.0.0->epynn) (21.3.0)\n",
      "Requirement already satisfied: Send2Trash>=1.8.0 in /home/synthase/.local/lib/python3.9/site-packages (from notebook->jupyter==1.0.0->epynn) (1.8.0)\n",
      "Requirement already satisfied: prometheus-client in /home/synthase/.local/lib/python3.9/site-packages (from notebook->jupyter==1.0.0->epynn) (0.14.1)\n",
      "Requirement already satisfied: argon2-cffi-bindings in /home/synthase/.local/lib/python3.9/site-packages (from argon2-cffi->notebook->jupyter==1.0.0->epynn) (21.2.0)\n",
      "Requirement already satisfied: cffi>=1.0.1 in /home/synthase/.local/lib/python3.9/site-packages/cffi-1.15.0-py3.9-linux-x86_64.egg (from argon2-cffi-bindings->argon2-cffi->notebook->jupyter==1.0.0->epynn) (1.15.0)\n",
      "Requirement already satisfied: pycparser in /home/synthase/.local/lib/python3.9/site-packages/pycparser-2.21-py3.9.egg (from cffi>=1.0.1->argon2-cffi-bindings->argon2-cffi->notebook->jupyter==1.0.0->epynn) (2.21)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /home/synthase/.local/lib/python3.9/site-packages (from jinja2->nbconvert==5.4.1->epynn) (2.0.1)\n",
      "Requirement already satisfied: qtpy>=2.0.1 in /home/synthase/.local/lib/python3.9/site-packages (from qtconsole->jupyter==1.0.0->epynn) (2.0.1)\n",
      "Requirement already satisfied: pure-eval in /home/synthase/.local/lib/python3.9/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (0.2.2)\n",
      "Requirement already satisfied: executing in /home/synthase/.local/lib/python3.9/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (0.8.2)\n",
      "Requirement already satisfied: asttokens in /home/synthase/.local/lib/python3.9/site-packages (from stack-data->ipython>=7.23.1->ipykernel->jupyter==1.0.0->epynn) (2.0.5)\n"
     ]
    }
   ],
   "source": [
    "# EpyNN/epynnlive/dummy_image/prepare_dataset.ipynb\n",
    "# Install dependencies\n",
    "!pip3 install --upgrade-strategy only-if-needed epynn\n",
    "\n",
    "# Standard library imports\n",
    "import random\n",
    "\n",
    "# Related third party imports\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Seeding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "random.seed(0)\n",
    "np.random.seed(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For reproducibility, as detailed [here](../dummy_boolean/prepare_dataset.ipynb#seeding)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To generate an image, we need:\n",
    "\n",
    "* Dimensions: ``HEIGHT`` and  ``WIDTH``.\n",
    "* Number of channels: ``DEPTH`` which is equal to ``1`` herein because we will prepare grayscale images. This would be *3* for *RGB* images, *4* for *CMYK* images, etc.\n",
    "* Number of tones per channel: ``N_TONES``.\n",
    "* The actual palette of shades: ``GSCALE``.\n",
    "\n",
    "The actual function to generate such image:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def features_image(WIDTH=28, HEIGHT=28):\n",
    "    \"\"\"Generate dummy image features.\n",
    "\n",
    "    :param WIDTH: Image width, defaults to 28.\n",
    "    :type WIDTH: int\n",
    "\n",
    "    :param HEIGHT: Image height, defaults to 28.\n",
    "    :type HEIGHT: int\n",
    "\n",
    "    :return: Random image features of size N_FEATURES.\n",
    "    :rtype: :class:`numpy.ndarray`\n",
    "\n",
    "    :return: Non-random image features of size N_FEATURES.\n",
    "    :rtype: :class:`numpy.ndarray`\n",
    "    \"\"\"\n",
    "    # Number of channels is one for greyscale images\n",
    "    DEPTH = 1\n",
    "\n",
    "    # Number of features describing a sample\n",
    "    N_FEATURES = WIDTH * HEIGHT * DEPTH\n",
    "\n",
    "    # Number of distinct tones in features\n",
    "    N_TONES = 16\n",
    "\n",
    "    # Shades of grey\n",
    "    GSCALE = [i for i in range(N_TONES)]\n",
    "\n",
    "    # Random choice of shades for N_FEATURES iterations\n",
    "    features = [random.choice(GSCALE) for j in range(N_FEATURES)]\n",
    "\n",
    "    # Vectorization of features\n",
    "    features = np.array(features).reshape(HEIGHT, WIDTH, DEPTH)\n",
    "\n",
    "    # Masked features\n",
    "    mask_on_features = features.copy()\n",
    "    mask_on_features[np.random.randint(0, HEIGHT)] = np.zeros_like(features[0])\n",
    "    mask_on_features[:, np.random.randint(0, WIDTH)] = np.zeros_like(features[:, 0])\n",
    "\n",
    "    # Random choice between random image or masked image\n",
    "    features = random.choice([features, mask_on_features])\n",
    "\n",
    "    return features, mask_on_features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In addition to constructing one image with randomly selected tones, this function achieves a random choice between image ``features`` and a modified version named ``mask_on_features``.\n",
    "\n",
    "The latter consists of the former, with modifications: one column and one row were randomly selected and had values for corresponding data points set to zero, visually corresponding to the following."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(2, 2)\n",
    "\n",
    "for i in range(2):\n",
    "\n",
    "    features, mask_on_features = features_image()\n",
    "    \n",
    "    title = 'Random' if np.sum(features) != np.sum(mask_on_features) else 'Non-random'\n",
    "\n",
    "    ax[0, i].imshow(features, cmap='gray')\n",
    "    ax[0, i].set_title(title)\n",
    "\n",
    "    ax[1, i].imshow(features - mask_on_features, cmap='gray')\n",
    "    ax[1, i].set_title('Diff. with Non-random')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These are image ``features`` we retrieve for two samples (first row).\n",
    "\n",
    "Below is the difference between ``features`` and ``mask_on_features``.\n",
    "\n",
    "When the difference renders an all black image, it means this difference is equal to zero and so the function has returned ``mask_on_features`` from the random choice within ``[features, mask_on_features]``.\n",
    "\n",
    "Said another way, when the difference between ``features`` and ``mask_on_features`` is zero then ``features = mask_on_features`` (non-random image)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate label"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Label is associated with ``features`` depending on if the corresponding image is random or not random."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def label_features(features, mask_on_features):\n",
    "    \"\"\"Prepare label associated with features.\n",
    "\n",
    "    The dummy law is:\n",
    "\n",
    "    Image is NOT random = positive\n",
    "    Image is random = negative\n",
    "\n",
    "    :param features: Random image features of size N_FEATURES\n",
    "    :type features: :class:`numpy.ndarray`\n",
    "\n",
    "    :param mask_on_features: Non-random image features of size N_FEATURES\n",
    "    :type mask_on_features: :class:`numpy.ndarray`\n",
    "\n",
    "    :return: Single-digit label with respect to features\n",
    "    :rtype: int\n",
    "    \"\"\"\n",
    "    # Single-digit positive and negative labels\n",
    "    p_label = 0\n",
    "    n_label = 1\n",
    "\n",
    "    # Test if image is not random (0)\n",
    "    if np.sum(features) == np.sum(mask_on_features):\n",
    "        label = p_label\n",
    "\n",
    "    # Test if image is random (1)\n",
    "    elif np.sum(features) != np.sum(mask_on_features):\n",
    "        label = n_label\n",
    "\n",
    "    return label"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The code above is commented and self explaining.\n",
    "\n",
    "Let's check the function we made for a few iterations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 5436 5436\n",
      "1 5500 5500\n",
      "0 5375 5375\n",
      "1 5412 5412\n",
      "1 5531 5531\n"
     ]
    }
   ],
   "source": [
    "for i in range(5):\n",
    "    features, mask_on_features = features_image()\n",
    "    label = label_features(features, mask_on_features)\n",
    "\n",
    "    print(label, np.sum(mask_on_features), np.sum(mask_on_features))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prepare dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The generic function we use to actually prepare a full labeled dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_dataset(N_SAMPLES=100):\n",
    "    \"\"\"Prepare a set of dummy time sample features and label.\n",
    "\n",
    "    :param N_SAMPLES: Number of samples to generate, defaults to 100.\n",
    "    :type N_SAMPLES: int\n",
    "\n",
    "    :return: Set of sample features.\n",
    "    :rtype: tuple[:class:`numpy.ndarray`]\n",
    "\n",
    "    :return: Set of single-digit sample label.\n",
    "    :rtype: tuple[int]\n",
    "    \"\"\"\n",
    "    # Initialize X and Y datasets\n",
    "    X_features = []\n",
    "    Y_label = []\n",
    "\n",
    "   # Iterate over N_SAMPLES\n",
    "    for i in range(N_SAMPLES):\n",
    "\n",
    "        # Compute random string features\n",
    "        features, mask_on_features = features_image()\n",
    "\n",
    "        # Retrieve label associated with features\n",
    "        label = label_features(features, mask_on_features)\n",
    "\n",
    "        # Append sample features to X_features\n",
    "        X_features.append(features)\n",
    "\n",
    "        # Append sample label to Y_label\n",
    "        Y_label.append(label)\n",
    "\n",
    "    # Prepare X-Y pairwise dataset\n",
    "    dataset = list(zip(X_features, Y_label))\n",
    "\n",
    "    # Shuffle dataset\n",
    "    random.shuffle(dataset)\n",
    "\n",
    "    # Separate X-Y pairs\n",
    "    X_features, Y_label = zip(*dataset)\n",
    "\n",
    "    return X_features, Y_label"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can test this function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "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"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "X_features, Y_label = prepare_dataset(N_SAMPLES=5)\n",
    "\n",
    "for features, label in zip(X_features, Y_label):\n",
    "    plt.imshow(features, cmap='gray')\n",
    "    plt.title('Digit = label = %s' % label)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that on this run we got a deviation from the expected mean: 4 images were associated with label 0 while only one was with label 1. According to the code, we expect a balanced dataset.\n",
    "\n",
    "Just to make sure."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "54\n",
      "46\n"
     ]
    }
   ],
   "source": [
    "X_features, Y_label = prepare_dataset(N_SAMPLES=100)\n",
    "\n",
    "print(Y_label.count(0))\n",
    "print(Y_label.count(1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The output of the code then fits our expectations."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Live examples"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The function ``prepare_dataset()`` presented herein is used in the following live examples:\n",
    "\n",
    "* Notebook at`EpyNN/epynnlive/dummy_image/train.ipynb` or following [this link](train.ipynb). \n",
    "* Regular python code at `EpyNN/epynnlive/dummy_image/train.py`."
   ]
  }
 ],
 "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": "3.9.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}