mnist_ideas.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import scipy\n",
    "from scipy import signal"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'list'>\n"
     ]
    }
   ],
   "source": [
    "from mnist import MNIST # pip install python-mnist\n",
    "mndata = MNIST('MNIST_ORG')\n",
    "image_list, label_list = mndata.load_training()\n",
    "print(type(image_list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "images = np.array(image_list)\n",
    "labels = np.array(label_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(60000, 784) (60000,)\n",
      "(784,)\n",
      "(784, 11872) (11872,)\n"
     ]
    }
   ],
   "source": [
    "print(images.shape, labels.shape)\n",
    "\n",
    "temp = images[0, :]\n",
    "print(temp.shape) #can also reduce m by randomly picking 50% of the data\n",
    "\n",
    "extracted_idx = [ii for ii in range(len(labels)) if(labels[ii]==0 or labels[ii] == 9)]\n",
    "#extracted_labels = [labels[ii] for ii in range(len(labels)) if(labels[ii]==0 or labels[ii] == 9)]\n",
    "\n",
    "X_train = images[extracted_idx, :]\n",
    "Y_train = labels[extracted_idx]\n",
    "\n",
    "X_train = np.transpose(X_train)\n",
    "\n",
    "print(X_train.shape, Y_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "for i in range(6):\n",
    "    val = np.random.randint(len(labels))\n",
    "    plt.subplot(2,3,i+1)\n",
    "    tmp = images[val, :]\n",
    "    plt.tight_layout()\n",
    "    plt.imshow(np.reshape(tmp, [28, 28]), cmap='gray', interpolation='none')\n",
    "    plt.title(\"Ground Truth: {}\".format(labels[val]))\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(14, 14)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f78b589f898>"
      ]
     },
     "execution_count": 107,
     "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": [
    "im = images[10,:]\n",
    "im1 = np.reshape(im, [28,28]) #we are reducing n\n",
    "#tmp1 = ([im[ii] for ii in range(len(im)) if ii % 4 ==0])\n",
    "tmp1 = im1[0::2, 0::2]\n",
    "print(tmp1.shape)\n",
    "#tmp1 = signal.resample(im, 14 * 14, domain='time')\n",
    "#tmp1 = np.reshape(tmp1, [14,14])\n",
    "plt.imshow(tmp1, cmap='gray')\n",
    "#plt.imshow(np.reshape(im, [28,28]), cmap='gray', interpolation='none')"
   ]
  }
 ],
 "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}