Digit Classification (Binary).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.datasets import load_digits\n",
    "\n",
    "from BASIC_spatial_pooler import SpatialPooler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "digits = load_digits()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Image data shape (1797, 64)\n",
      "Label data shape (1797,)\n"
     ]
    }
   ],
   "source": [
    "# We see that we have 1797 images, each 8x8=64 pixels\n",
    "print(\"Image data shape\", digits.data.shape)\n",
    "\n",
    "# And we see we have the right number of labels (integers from 0-9)\n",
    "print(\"Label data shape\", digits.target.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABHsAAAEKCAYAAACCF1ktAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3X20XXdZJ/DvY1NU3pqovGiLhoqgjrMINjIiLGyxVVTG1nFgcITV4ksrS5x2qaOVtWZaX5aWUbCs0dFGlHSU0TGIrYMoNgsqviCSQlBLASETh4KALJLyJrCA3/xxTjS93OSek5zf3fvufD5r3XXuPWfn2c85ud/k3Of+9t7VWgsAAAAA0/BZQzcAAAAAwOoY9gAAAABMiGEPAAAAwIQY9gAAAABMiGEPAAAAwIQY9gAAAABMiGHPCFTV/auqVdXLV1DrQFV9eBV9wZlONmF85BLGSTZhnGTzzHVGD3vm3/TLfFwxdM9TVDPfV1V3VNVHqupoVe2vqkuG7o1hyObwqurhVfVfqup3q+rQca/1Q4fujWHI5fCq6sKqev78zfb7q+rjVfWOqvqVqvqSoftjGLI5vKq6uKpeUlV3VtUHqupj82y+rKqeOHR/DEM2x6eq7ldVb5+/3m8Zup/etg3dwMB+Yp37rklyTpIXJjm65rGDnfr4SJKvSLKKKel3JPnsFdTZTL+c5Kokh5P8SpL7JXl6kj+uqme11vYO1xoDkc3hPT7JTyZpSd6e5ENJHjBoRwxNLof38sz6/fMkv5Hk00mekNn/oU+vqotaa28csD+GIZvDe3KSJyZ5XZL9Sf4pyc4k/zbJt1fVta215w3XHgORzfF5QZKHDN3EZqnW2tA9jEpVHU7yJUke3lo7PGw30zdfvfPHSe5M8rjW2ofm9z8qyYEkleQRrbX3DNclYyCbm6uqdiY5N8mbWmsfrqoDSS5I8oXyyDFyubmq6rlJXtRae9+a+38myY8n+cvW2uMGaY5Rkc3NVVWf01r72Dr3n5/kjsx+kfng1traH+45w8jmcKrqW5L8QZJnZ7bY4K2ttS8ftqu+zujDuE7VsWMVq+pzq+qn50vBPlFVvzh//POr6tqq+pOqevf8sffOD4f46nXqrXscZVX9/Pz+3VX1XfPDnP5pvnT7N6rqwSfqbc19T5nX+ZGqemxVvbKq7pk/h/1VdcEJnucXV9Vvzvf30fn+/8Px9U7vlUwyC1uS/MSxQU+StNbemuRXM/vP8Zkr2A9nANlcXTZba4dba3/eWnNcNqdFLleay59ZO+iZ+6kkn0rytVX1Oae7H84MsrnSbH7GoGd+/6HMfnl5dmY/4MOGZHOlP2se29fnJ/m1JLcm+c1V1R07w55T91mZLae+IsmfJLkxyV3zxx6T2bK9j2X2DfWCJLcn+ZYkr63lj9390cwGH29L8ktJ/i7JM5K8sqrOWqLOE5K8JrPDMn41sxU1T0pye6051r+qzkvy2iTfldmSwhdmtvrm5iTfs17x44K5zMm/Lpr388p1HvvD+e2TlqgHsrnGKWYTVkku11hxLj+V2SFdxz5gUbK5xiqzWVXnZvY6fiTJO063HmcU2VzjNLN5U2ZD16tO4c9uWWf6OXtOx+dmdv6Kr1pnSeYbkjy0tXbk+Dur6kszO5b3+Um+Zol9fUOSXa21t83rVJJbknxbkm9K8ooF61ya5KmttZce19MPJ/n5JD+QWdCPeX6SL0ryX1trP3Xc9v8jyZ8t0fsJVdVDkmxP8p7W2gfX2eTv5rePXMX+OGPIJoyPXPb1zMzexO5vrX1iE/bHdMjmClXV1yX5xszy+LDMntt9k3yflbIsSTZXpKqemdm5hp7eWntvVd1/lfXHzMqe0/Pj6x1721r7wNrwze9/R5LfT7J7vpRsUT93LHzzOi3Ji+ZfPnaJOq88Pnxze9bWqaoHJPl3Sd6X5OeO37i19pdJ9p2g/qszO/nX9y/Yzznz23tO8Pix+7cvWA+Okc17Wzab0INc3ttKcllVX57ZG+mPJ/nPp1OLM5Zs3tvpZPPrklyX5LmZDWE/neQ/ttZuPoVaIJv3tnQ2q+phSf57kpe21v73on9uKgx7Ts9fneiBqrqoZpdbvHt+HGWrqpbkWfNNvmiJ/RxY5753zm93nE6d+Xly7llT56syW/V1xwmOQV532tpa+0hr7S2ttbuX6Olk6ljpFdXjzCGb96616mzCqZDLe9c67VzODxF5RWa/PLmqtdbrSi5Mm2zeu9YpZ7O19vOttcpsNc+/TvI7SfZV1fOXrQWRzbW1lsrmfIXS3sx+GfLsk289TQ7jOnUfPf6Ewserqmck+Z+ZXd7utiT/N7NjdVtmSzsfl+UuWbfemfs/Ob9d5jjKE10B4JNr6hxbcfPeE2x/ovuXdWzlzjknePyBa7aDRcgmjI9crth80HN7Zid9vdLKAU6RbHbQWvunJH+b5Pvnh4z8UFXd1lr7o177ZHJk8/Q9O7NzBn1Ha+39K6q5pRj2nLqTrTb56SQfSvKY+Vn4/1lVfVlmARyzY+fPecgJHj/R/UuZHzN5NMlDquqB65y358vmt28LLE42YXzkcoWq6ouTvCrJw5N8t0EPp0E2+/vDzE5Ce2ESwx4WJZun79iVyX53tsjnMzxqvhoqSc5urX1yvY22MsOeFauqbZn9lu0164Tv7Iw/fEnyN5lNYC+oqs9ZZ3ndE1a4r1cn+fbMTv619vjMb57fvmqF++MMJZswPnK5vKo6P7P/F89Lcnlr7Yy5hCybRzZX6tz57eR+kGTzyeZS/vQE929LcnlmR48cO8fQJK9k6Zw9KzafCL4ryb+qqi84dn9VfVaSn83st3CjNl8yeEuSB2fNyR6r6t8keep6f66q7ldVXz6/lN6ifnl+e938ZF3Haj0qyfdmtiTxN5aoB+uSzaWzCd3J5XK5rKpHZHYJ3nMzO+mrQQ9dyObS2byw1lk6UFVfkeRH5l/+waL14ERkc/FsttZubq1979qPJM+Zb/Ke4+6f5LDHyp4+fiGzK2P8dVW9LLNJ4dcn2ZnZUs5vPvEfHY0fzmyq+pNV9cQkr8/st4hPS/J/klyWz5yAXjR/7A+SPGWRnbTWbquqPUmuzL+8XvdL8vTMLjf43a2195z+04EksrlwNqvqs5PcdNxdO+e3L6yqf5p//outtfVO6gfLkMsFcllV98ls0PNFSV6b5Cur6vp1Nt3TWnv3qT0NuBfZXPD/zCT7kxyuqtcnuTvJfZI8MsklmZ2r5Gdba689vacC/0w2F8/mGc2wp48XZHbCrOck+e7MVqfcntk37/dlCwSwtfb/quprM5sQf1NmYXxzZkvePjezAK49x86p+v4kd8xvn53Zsr7XJ7mhtXbbivYBiWwu4+x5zbWedtznL8/6V3CAZcjlYu6Tf7m6yuNy4qX6L09i2MMqyObinpvZiWAfn+RBmV1R9j2ZHSJyU2vt1SvYBxwjmyykWnNVa5ZTVS9M8p+SPKG19udD9wPMyCaMj1zCOMkmjJNsro5hDydUVV+0dil4VX1Nktck+UCSL5niWcth7GQTxkcuYZxkE8ZJNvtzGBcnc1dVvSHJnUk+luRR+ZdlgT8gfDAY2YTxkUsYJ9mEcZLNzqzs4YSq6meTfEuSL05y/yRHkvxFkv/WWvuLIXuDM5lswvjIJYyTbMI4yWZ/hj0AAAAAE/JZQzcAAAAAwOoY9gAAAABMiGEPAAAAwIQY9gAAAABMiGEPAAAAwIQY9gAAAABMiGEPAAAAwIQY9gAAAABMiGEPAAAAwIQY9gAAAABMiGEPAAAAwIQY9gAAAABMiGEPAAAAwIQY9gAAAABMiGEPAAAAwIQY9gAAAABMiGEPAAAAwIQY9gAAAABMiGEPAAAAwIQY9gAAAABMiGEPAAAAwIQY9gAAAABMyLYeRauq9ai7WXbs2NG1/rnnntu1/gc/+MGu9ZPkXe96V9f6n/rUp7rW7621VkP3sNZWz2Vvj3zkI7vW37atyz+399I7l/fcc0/X+pvg/a21Bw3dxFqyeXL3v//9u9Z/xCMe0bV+knz0ox/tWv9tb3tb1/qbQDY7eOhDH9q1fu/3sx//+Me71k+Su+66q2v9rf5+NrK5JZ111lld6+/cubNr/SR5xzve0X0fW9xC2ez/08cWdPHFF3etf8MNN3Stv3///q71k+Taa6/tWv/IkSNd68Nae/bs6Vp/+/btXesnyXXXXde1/q233tq1/ib4+6EbYHm7d+/uWv+WW27pWj9JDh482LX+hRde2LX+JpDNDi6//PKu9Xu/nz106FDX+kn/f18m8H5WNregBzzgAV3rP//5z+9aP0kuu+yy7vvY4hbKpsO4AAAAACbEsAcAAABgQgx7AAAAACbEsAcAAABgQgx7AAAAACbEsAcAAABgQgx7AAAAACZkoWFPVT25qt5aVW+vqmt7NwUsRjZhnGQTxkk2YZxkE1Zvw2FPVZ2V5JeSfHOSr0zynVX1lb0bA05ONmGcZBPGSTZhnGQT+lhkZc9jk7y9tXaotfaJJL+d5NK+bQELkE0YJ9mEcZJNGCfZhA4WGfacm+Sdx3199/y+e6mqK6vqQFUdWFVzwEltmE25hEHIJoyTbMI4ySZ0sG2BbWqd+9pn3NHaniR7kqSqPuNxYOU2zKZcwiBkE8ZJNmGcZBM6WGRlz91JHnbc1+cleXefdoAlyCaMk2zCOMkmjJNsQgeLDHten+TLqurhVXWfJE9P8vt92wIWIJswTrIJ4ySbME6yCR1seBhXa+2TVfWcJK9MclaSX2+t3dm9M+CkZBPGSTZhnGQTxkk2oY9FztmT1torkryicy/AkmQTxkk2YZxkE8ZJNmH1FjmMCwAAAIAtwrAHAAAAYEIMewAAAAAmxLAHAAAAYEIMewAAAAAmxLAHAAAAYEIWuvT6meaGG27oWv/888/vWn/Hjh1d6yfJBz7wga71n/a0p3Wtv2/fvq712XqOHj3atf7Xf/3Xd62fJBdddFHX+rfeemvX+mxNu3bt6lr/1a9+ddf699xzT9f6SbJz587u+2Dr6f1+86lPfWrX+ldddVXX+jfddFPX+klywQUXdK2/f//+rvVhPVdccUXX+gcPHuxan9WxsgcAAABgQgx7AAAAACbEsAcAAABgQgx7AAAAACbEsAcAAABgQgx7AAAAACbEsAcAAABgQgx7AAAAACZkw2FPVf16Vb2vqv52MxoCFiObME6yCeMkmzBOsgl9LLKyZ2+SJ3fuA1je3sgmjNHeyCaM0d7IJozR3sgmrNyGw57W2muSfGATegGWIJswTrIJ4ySbME6yCX1sW1WhqroyyZWrqgecPrmEcZJNGCfZhHGSTVjeyoY9rbU9SfYkSVW1VdUFTp1cwjjJJoyTbMI4ySYsz9W4AAAAACbEsAcAAABgQha59PpvJXltkkdV1d1V9T392wI2IpswTrIJ4ySbME6yCX1seM6e1tp3bkYjwHJkE8ZJNmGcZBPGSTahD4dxAQAAAEyIYQ8AAADAhBj2AAAAAEyIYQ8AAADAhBj2AAAAAEyIYQ8AAADAhGx46fUxuuCCC7rWP//887vW/9Iv/dKu9Q8dOtS1fpLcdtttXev3/jvet29f1/qs3q5du7rWv/DCC7vW3wwHDx4cugXOQJdddlnX+m9605u61r/lllu61k+S6667rvs+2Hr27NnTtf7znve8rvUPHDjQtf5mvJ/dv39/933AWtu3b+9a/4orruha/8Ybb+xaP0l27tzZfR89HT58eOgWkljZAwAAADAphj0AAAAAE2LYAwAAADAhhj0AAAAAE2LYAwAAADAhhj0AAAAAE2LYAwAAADAhhj0AAAAAE7LhsKeqHlZVr66qu6rqzqq6ejMaA05ONmGcZBPGSTZhnGQT+ti2wDafTPLDrbU3VNUDktxRVbe11t7cuTfg5GQTxkk2YZxkE8ZJNqGDDVf2tNb+obX2hvnnH0pyV5JzezcGnJxswjjJJoyTbMI4ySb0sdQ5e6pqZ5LHJHldj2aAUyObME6yCeMkmzBOsgmrs8hhXEmSqrp/kt9Nck1r7YPrPH5lkitX2BuwgJNlUy5hOLIJ4ySbME6yCau10LCnqs7OLHgvaa29bL1tWmt7kuyZb99W1iFwQhtlUy5hGLIJ4ySbME6yCau3yNW4KsmvJbmrtfaC/i0Bi5BNGCfZhHGSTRgn2YQ+Fjlnz+OTPDPJk6rq4PzjWzr3BWxMNmGcZBPGSTZhnGQTOtjwMK7W2p8lqU3oBViCbMI4ySaMk2zCOMkm9LHU1bgAAAAAGDfDHgAAAIAJMewBAAAAmBDDHgAAAIAJMewBAAAAmBDDHgAAAIAJ2fDS62O0Y8eOrvXvuOOOrvUPHTrUtf5m6P0asfVcc801Xetff/31Xeufc845Xetvhttvv33oFjgD3XjjjV3rHz58uGv93v0nya233tp9H2w9vd8Pnn/++Vu6/v79+7vWT/r/THHkyJGu9dmarrjiiq71d+7c2bX+3r17u9ZP+v/ffPTo0a71e//csigrewAAAAAmxLAHAAAAYEIMewAAAAAmxLAHAAAAYEIMewAAAAAmxLAHAAAAYEIMewAAAAAmxLAHAAAAYEI2HPZU1edU1V9V1Zuq6s6q+onNaAw4OdmEcZJNGCfZhHGSTehj2wLbfDzJk1prH66qs5P8WVX9YWvtLzv3BpycbMI4ySaMk2zCOMkmdLDhsKe11pJ8eP7l2fOP1rMpYGOyCeMkmzBOsgnjJJvQx0Ln7Kmqs6rqYJL3Jbmttfa6vm0Bi5BNGCfZhHGSTRgn2YTVW2jY01r7VGttV5Lzkjy2qr5q7TZVdWVVHaiqA6tuEljfRtmUSxiGbMI4ySaMk2zC6i11Na7W2tEktyd58jqP7Wmt7W6t7V5Rb8CCTpRNuYRhySaMk2zCOMkmrM4iV+N6UFVtn3/+uUkuTvKW3o0BJyebME6yCeMkmzBOsgl9LHI1ri9McnNVnZXZcOh3Wmsv79sWsADZhHGSTRgn2YRxkk3oYJGrcf11ksdsQi/AEmQTxkk2YZxkE8ZJNqGPpc7ZAwAAAMC4GfYAAAAATIhhDwAAAMCEGPYAAAAATIhhDwAAAMCEGPYAAAAATMiGl14fox07dnStv3///q71p6D338GRI0e61mf1brzxxq719+7d27X+FL7ntm/fPnQLjFDv74trrrmma/3LLrusa/3NcMUVVwzdAmegQ4cOda3/eZ/3eV3r33bbbV3rb8Y+Lrnkkq71p/DeZYwuvfTSrvV/4Rd+oWv9m2++uWv9zXD11Vd3rf+sZz2ra/2xsLIHAAAAYEIMewAAAAAmxLAHAAAAYEIMewAAAAAmxLAHAAAAYEIMewAAAAAmxLAHAAAAYEIMewAAAAAmZOFhT1WdVVVvrKqX92wIWI5swjjJJoyPXMI4ySas3jIre65OclevRoBTJpswTrIJ4yOXME6yCSu20LCnqs5L8q1JXtS3HWAZsgnjJJswPnIJ4ySb0MeiK3tuTPKjST7dsRdgebIJ4ySbMD5yCeMkm9DBhsOeqnpKkve11u7YYLsrq+pAVR1YWXfACS2STbmEzSebMD7ez8I4ySb0s8jKnscn+baqOpzkt5M8qap+c+1GrbU9rbXdrbXdK+4RWN+G2ZRLGIRswvh4PwvjJJvQyYbDntbaj7fWzmut7Uzy9CSvaq09o3tnwEnJJoyTbML4yCWMk2xCP8tcjQsAAACAkdu2zMattduT3N6lE+CUySaMk2zC+MgljJNswmpZ2QMAAAAwIYY9AAAAABNi2AMAAAAwIYY9AAAAABNi2AMAAAAwIYY9AAAAABNi2AMAAAAwIduGbuBUHDlypGv9Cy64oGv93nbs2NF9H71fo3379nWtD1O0a9eurvUPHjzYtT59XH/99V3rX3311V3r93bZZZd138fRo0e77wM2W+/345dccknX+kly0003da3/Yz/2Y13rX3vttV3rn6nuueeeLV3/8ssv71q/9/vNzXDLLbcM3cKmsLIHAAAAYEIMewAAAAAmxLAHAAAAYEIMewAAAAAmxLAHAAAAYEIMewAAAAAmxLAHAAAAYEK2LbJRVR1O8qEkn0ryydba7p5NAYuRTRgn2YRxkk0YJ9mE1Vto2DN3UWvt/d06AU6VbMI4ySaMk2zCOMkmrJDDuAAAAAAmZNFhT0vyx1V1R1Vd2bMhYCmyCeMkmzBOsgnjJJuwYosexvX41tq7q+rBSW6rqre01l5z/AbzUAombK6TZlMuYTCyCeMkmzBOsgkrttDKntbau+e370vye0keu842e1pru51MCzbPRtmUSxiGbMI4ySaMk2zC6m047Kmq+1XVA459nuQbk/xt78aAk5NNGCfZhHGSTRgn2YQ+FjmM6yFJfq+qjm3/v1prf9S1K2ARsgnjJJswTrIJ4ySb0MGGw57W2qEkj96EXoAlyCaMk2zCOMkmjJNsQh8uvQ4AAAAwIYY9AAAAABNi2AMAAAAwIYY9AAAAABNi2AMAAAAwIYY9AAAAABNi2AMAAAAwIduGbuBUHDp0qGv9Cy64oGv9pz71qVu6/mZ43vOeN3QLAJOwd+/ervUvvPDCrvUf/ehHd61/yy23dK2fJLfeemvX+i9+8Yu71u/dP33ccMMNXevv37+/a/0dO3Z0rZ8kF198cdf6+/bt61qfPm6//fau9bdv3961/q5du7rW7/36JMnNN9/ctf7Ro0e71h8LK3sAAAAAJsSwBwAAAGBCDHsAAAAAJsSwBwAAAGBCDHsAAAAAJsSwBwAAAGBCDHsAAAAAJsSwBwAAAGBCFhr2VNX2qnppVb2lqu6qqsf1bgzYmGzCOMkmjJNswjjJJqzetgW3e2GSP2qt/fuquk+S+3bsCVicbMI4ySaMk2zCOMkmrNiGw56qemCSJya5Iklaa59I8om+bQEbkU0YJ9mEcZJNGCfZhD4WOYzr/CT/mOTFVfXGqnpRVd1v7UZVdWVVHaiqAyvvEljPhtmUSxiEbMI4ySaMk2xCB4sMe7Yl+eokv9xae0ySjyS5du1GrbU9rbXdrbXdK+4RWN+G2ZRLGIRswjjJJoyTbEIHiwx77k5yd2vtdfOvX5pZGIFhySaMk2zCOMkmjJNsQgcbDntaa+9J8s6qetT8rm9I8uauXQEbkk0YJ9mEcZJNGCfZhD4WvRrXDyZ5yfzM6IeSPKtfS8ASZBPGSTZhnGQTxkk2YcUWGva01g4mcXwkjIxswjjJJoyTbMI4ySas3iLn7AEAAABgizDsAQAAAJgQwx4AAACACTHsAQAAAJgQwx4AAACACTHsAQAAAJiQhS69PjaHDh3qWv/aa6/tWv+GG27oWv+OO+7oWj9Jdu92ZUQ219GjR7vWv/XWW7vWv/TSS7vWT5ILL7ywa/29e/d2rU8fBw8e7Fp/165dW7r+9ddf37V+0j//hw8f7lq/97+P9HHkyJGu9W+66aau9TfDvn37uta/6qqrutaH9fR+z3zOOed0rZ94z7kqVvYAAAAATIhhDwAAAMCEGPYAAAAATIhhDwAAAMCEGPYAAAAATIhhDwAAAMCEGPYAAAAATIhhDwAAAMCEbDjsqapHVdXB4z4+WFXXbEZzwInJJoyTbMI4ySaMk2xCH9s22qC19tYku5Kkqs5K8q4kv9e5L2ADsgnjJJswTrIJ4ySb0Meyh3F9Q5J3tNb+vkczwCmTTRgn2YRxkk0YJ9mEFdlwZc8aT0/yW+s9UFVXJrnytDsCTsW62ZRLGJxswjjJJoyTbMKKLLyyp6ruk+Tbkuxb7/HW2p7W2u7W2u5VNQds7GTZlEsYjmzCOMkmjJNswmotcxjXNyd5Q2vtvb2aAU6JbMI4ySaMk2zCOMkmrNAyw57vzAkO4QIGJZswTrIJ4ySbME6yCSu00LCnqu6b5JIkL+vbDrAM2YRxkk0YJ9mEcZJNWL2FTtDcWvtoks/v3AuwJNmEcZJNGCfZhHGSTVi9ZS+9DgAAAMCIGfYAAAAATIhhDwAAAMCEGPYAAAAATIhhDwAAAMCEGPYAAAAATEi11lZftOofk/z9En/kC5K8f+WNbB79D2ts/X9Ja+1BQzex1hmYy2TrPwf9r5ZsjsNW7z/Z+s9hbP3L5jjof3hjew6yOQ76H9YY+18om12GPcuqqgOttd1D93Gq9D+srd7/WE3hdd3qz0H/rGerv65bvf9k6z+Hrd7/WG3111X/w5vCcxijrf666n9YW7l/h3EBAAAATIhhDwAAAMCEjGXYs2foBk6T/oe11fsfqym8rlv9Oeif9Wz113Wr959s/eew1fsfq63+uup/eFN4DmO01V9X/Q9ry/Y/inP2AAAAALAaY1nZAwAAAMAKGPYAAAAATMigw56qenJVvbWq3l5V1w7Zy7Kq6mFV9eqququq7qyqq4fu6VRU1VlV9caqevnQvSyrqrZX1Uur6i3zv4fHDd3TVMjm8GST9cjm8GST9cjm8GST9cjm8GRzOIOds6eqzkrytiSXJLk7yeuTfGdr7c2DNLSkqvrCJF/YWntDVT0gyR1JLtsq/R9TVT+UZHeSB7bWnjJ0P8uoqpuT/Glr7UVVdZ8k922tHR26r61ONsdBNllLNsdBNllLNsdBNllLNsdBNocz5MqexyZ5e2vtUGvtE0l+O8mlA/azlNbaP7TW3jD//ENJ7kpy7rBdLaeqzkvyrUleNHQvy6qqByZ5YpJfS5LW2ie2UvBGTjYHJpucgGwOTDY5AdkcmGxyArI5MNkc1pDDnnOTvPO4r+/OFvvmPaaqdiZ5TJLXDdvJ0m5M8qNJPj10I6fg/CT/mOTF82WBL6qq+w3d1ETI5vBkk/XI5vBkk/XI5vBkk/XI5vBkc0BDDntqnfu23HXgq+r+SX43yTWttQ8O3c+iquopSd7XWrtj6F5O0bYkX53kl1trj0nykSRb6jjcEZPNAckmJyGbA5JNTkI2BySbnIRsDkg2hzfksOfuJA877uvzkrx7oF5OSVWdnVnwXtJae9nQ/Szp8Um+raoOZ7ak8UlV9ZvDtrSUu5Pc3Vo7Nt1+aWZh5PTJ5rBkkxORzWHJJicim8OSTU5ENoclmwMbctjz+iRfVlUPn5/s6OlJfn/AfpZSVZXZ8Xt3tdbmuiVpAAAA30lEQVReMHQ/y2qt/Xhr7bzW2s7MXvtXtdaeMXBbC2utvSfJO6vqUfO7viHJljpZ2YjJ5oBkk5OQzQHJJichmwOSTU5CNgckm8PbNtSOW2ufrKrnJHllkrOS/Hpr7c6h+jkFj0/yzCR/U1UH5/c9t7X2igF7OtP8YJKXzP/xPpTkWQP3MwmyyQrIZgeyyQrIZgeyyQrIZgeyyQps6WwOdul1AAAAAFZvyMO4AAAAAFgxwx4AAACACTHsAQAAAJgQwx4AAACACTHsAQAAAJgQwx4AAACACTHsAQAAAJiQ/w+a7ymTTBodXgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x288 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show a few training images\n",
    "plt.figure(figsize=(20,4))\n",
    "for index, (image,label) in enumerate(zip(digits.data[0:5], digits.target[0:5])):\n",
    "    plt.subplot(1, 5, index + 1)\n",
    "    plt.imshow(numpy.reshape(image, (8,8)), cmap=plt.cm.gray)\n",
    "    plt.title('Training: %i\\n' % label, fontsize=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sklearn.utils.Bunch"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(digits)\n",
    "binary_digits = load_digits()\n",
    "type(binary_digits)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.,  0.,  5., 13.,  9.,  1.,  0.,  0.,  0.,  0., 13., 15., 10.,\n",
       "       15.,  5.,  0.,  0.,  3., 15.,  2.,  0., 11.,  8.,  0.,  0.,  4.,\n",
       "       12.,  0.,  0.,  8.,  8.,  0.,  0.,  5.,  8.,  0.,  0.,  9.,  8.,\n",
       "        0.,  0.,  4., 11.,  0.,  1., 12.,  7.,  0.,  0.,  2., 14.,  5.,\n",
       "       10., 12.,  0.,  0.,  0.,  0.,  6., 13., 10.,  0.,  0.,  0.])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "binary_digits.data[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a spatial pooler object\n",
    "inputDimensions = (8,8)\n",
    "columnDimensions = (16,16)\n",
    "potentialRadius = 3\n",
    "potentialPct = 0.5\n",
    "sparsity = 0.03\n",
    "stimulusThreshold = 2\n",
    "synPermInactiveDec = 0.008 #0.008\n",
    "synPermActiveInc = 0.05 #0.05\n",
    "synConnectedPermThreshold = 0.1\n",
    "minPctActiveDutyCycles=0.01\n",
    "dutyCyclePeriod=20\n",
    "boostStrength=3.0\n",
    "learn=True\n",
    "seed=12345\n",
    "debug=False\n",
    "sp = SpatialPooler(inputDimensions, columnDimensions, potentialRadius, potentialPct, sparsity, stimulusThreshold, synPermInactiveDec, synPermActiveInc, synConnectedPermThreshold, minPctActiveDutyCycles, dutyCyclePeriod, boostStrength, learn, seed, debug)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x288 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Now, let's threshold the data\n",
    "threshold = 12\n",
    "#binary_digits = numpy.copy(digits)\n",
    "binary_digits.data[digits.data < threshold] = 0.0\n",
    "binary_digits.data[digits.data > 0.0] = 1.0\n",
    "\n",
    "# Show the same training images after thresholding\n",
    "plt.figure(figsize=(20,4))\n",
    "for index, (image,label) in enumerate(zip(binary_digits.data[0:5], binary_digits.target[0:5])):\n",
    "    plt.subplot(1, 5, index + 1)\n",
    "    plt.imshow(numpy.reshape(image, (8,8)), cmap=plt.cm.gray)\n",
    "    plt.title('Training: %i\\n' % label, fontsize=20)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split data into training and testing sets\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "x_train, x_test, y_train, y_test = train_test_split(binary_digits.data, binary_digits.target, test_size=0.25, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# train the spatial pooler on the training data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "type(x_train)\n",
    "x_train.shape\n",
    "num_training_images = 1347\n",
    "\n",
    "x_train[0].shape\n",
    "for image in x_train:\n",
    "    sp.compute(image.reshape((8,8)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# now turn off learning for spatial pooler and generate a\n",
    "# new dataset of spatial pooler representations for the \n",
    "# classifier to learn from (still over training data)\n",
    "sp._learn = False\n",
    "\n",
    "SDR_train = numpy.zeros((1347,256))\n",
    "index = 0\n",
    "for image in x_train:\n",
    "    sp.compute(image.reshape((8,8)))\n",
    "    SDR = sp._columnActivations\n",
    "    SDR_train[index] = SDR.reshape(-1)\n",
    "    index = index + 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Now repeat to create testing dataset\n",
    "x_test.shape # (450, 64)\n",
    "\n",
    "SDR_test = numpy.zeros((450,256))\n",
    "index = 0\n",
    "for image in x_test:\n",
    "    sp.compute(image.reshape((8,8)))\n",
    "    SDR = sp._columnActivations\n",
    "    SDR_test[index] = SDR.reshape(-1)\n",
    "    index = index + 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import scikit learn logistic regression ML model\n",
    "from sklearn.linear_model import LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "          penalty='l2', random_state=None, solver='liblinear', tol=0.0001,\n",
       "          verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a model instance\n",
    "LRmodel = LogisticRegression()\n",
    "\n",
    "# train model and store information learned\n",
    "# x is digit images and y is labels\n",
    "#LRmodel.fit(x_train, y_train)\n",
    "LRmodel.fit(SDR_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([4])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# predict labels for first test image\n",
    "#LRmodel.predict(x_test[0].reshape(1,-1))\n",
    "LRmodel.predict(SDR_test[0].reshape(1,-1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get all test predictions\n",
    "predictions = LRmodel.predict(SDR_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.08444444444444445\n"
     ]
    }
   ],
   "source": [
    "# compute accuracy\n",
    "score = LRmodel.score(SDR_test, y_test)\n",
    "print(score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x288 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Are SDR's actually changing?\n",
    "# Show the same training images after thresholding\n",
    "plt.figure(figsize=(20,4))\n",
    "for index, (image,label) in enumerate(zip(SDR_test[0:5], y_test[0:5])):\n",
    "    plt.subplot(1, 5, index + 1)\n",
    "    plt.imshow(numpy.reshape(image, (16,16)), cmap=plt.cm.gray)\n",
    "    plt.title('Testing: %i\\n' % label, fontsize=20)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# check first two test cases\n",
    "#SDR1 = SDR_test[0]\n",
    "#SDR2 = SDR_test[1]\n",
    "\n",
    "#print(numpy.argwhere(SDR1 > 0))\n",
    "#print(numpy.argwhere(SDR2 > 0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#sp._boostFactors\n",
    "#sp._activeDutyCycles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 1., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 1.,\n",
       "       1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0.])"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sp._columnActivations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.082, 0.082, 0.082, ..., 0.   , 0.   , 0.   ],\n",
       "       [0.062, 0.062, 0.062, ..., 0.   , 0.   , 0.   ],\n",
       "       [0.092, 0.092, 0.092, ..., 0.   , 0.   , 0.   ],\n",
       "       ...,\n",
       "       [0.   , 0.   , 0.   , ..., 0.092, 0.092, 0.092],\n",
       "       [0.   , 0.   , 0.   , ..., 0.092, 0.092, 0.092],\n",
       "       [0.   , 0.   , 0.   , ..., 0.072, 0.072, 0.072]])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sp._permanences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8604"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(numpy.argwhere(sp._permanences.reshape(-1) > 0.00))\n",
    "#len(sp._permanences.reshape(-1)) # 16384"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f44f6f605c0>]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualize active duty cycles\n",
    "plt.plot(numpy.array(list(range(sp._numColumns))), sp._activeDutyCycles)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f44fac3e860>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(numpy.reshape(sp._activeDutyCycles, (16,16)), cmap=plt.cm.gray)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0  0  0  0 37  0  0  0  0  0]\n",
      " [ 0  0  0  0 43  0  0  0  0  0]\n",
      " [ 0  0  0  0 44  0  0  0  0  0]\n",
      " [ 0  0  0  0 45  0  0  0  0  0]\n",
      " [ 0  0  0  0 38  0  0  0  0  0]\n",
      " [ 0  0  0  0 48  0  0  0  0  0]\n",
      " [ 0  0  0  0 52  0  0  0  0  0]\n",
      " [ 0  0  0  0 48  0  0  0  0  0]\n",
      " [ 0  0  0  0 48  0  0  0  0  0]\n",
      " [ 0  0  0  0 47  0  0  0  0  0]]\n"
     ]
    }
   ],
   "source": [
    "# Look at confusion matrix\n",
    "import sklearn.metrics\n",
    "#predictions.shape\n",
    "CM = sklearn.metrics.confusion_matrix(y_test, predictions)\n",
    "print(CM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.082, 0.082, 0.082, ..., 0.   , 0.   , 0.   ],\n",
       "       [0.062, 0.062, 0.062, ..., 0.   , 0.   , 0.   ],\n",
       "       [0.092, 0.092, 0.092, ..., 0.   , 0.   , 0.   ],\n",
       "       ...,\n",
       "       [0.   , 0.   , 0.   , ..., 0.092, 0.092, 0.092],\n",
       "       [0.   , 0.   , 0.   , ..., 0.092, 0.092, 0.092],\n",
       "       [0.   , 0.   , 0.   , ..., 0.072, 0.072, 0.072]])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sp._permanences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}