mnist.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting numpy\n",
      "  Using cached numpy-1.26.4-cp310-cp310-win_amd64.whl (15.8 MB)\n",
      "Installing collected packages: numpy\n",
      "Successfully installed numpy-1.26.4\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip is available: 23.0.1 -> 24.0\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting scikit-learn\n",
      "  Downloading scikit_learn-1.5.0-cp310-cp310-win_amd64.whl (11.0 MB)\n",
      "     ---------------------------------------- 0.0/11.0 MB ? eta -:--:--\n",
      "      --------------------------------------- 0.2/11.0 MB 5.0 MB/s eta 0:00:03\n",
      "     --- ------------------------------------ 1.1/11.0 MB 8.5 MB/s eta 0:00:02\n",
      "     ------ --------------------------------- 1.7/11.0 MB 9.3 MB/s eta 0:00:01\n",
      "     -------- ------------------------------- 2.3/11.0 MB 9.7 MB/s eta 0:00:01\n",
      "     ---------- ----------------------------- 2.8/11.0 MB 10.0 MB/s eta 0:00:01\n",
      "     ------------ --------------------------- 3.6/11.0 MB 10.3 MB/s eta 0:00:01\n",
      "     -------------- ------------------------- 4.0/11.0 MB 10.2 MB/s eta 0:00:01\n",
      "     ---------------- ----------------------- 4.6/11.0 MB 9.8 MB/s eta 0:00:01\n",
      "     ------------------ --------------------- 5.1/11.0 MB 9.5 MB/s eta 0:00:01\n",
      "     --------------------- ------------------ 5.8/11.0 MB 9.8 MB/s eta 0:00:01\n",
      "     ----------------------- ---------------- 6.5/11.0 MB 10.0 MB/s eta 0:00:01\n",
      "     -------------------------- ------------- 7.2/11.0 MB 9.9 MB/s eta 0:00:01\n",
      "     ---------------------------- ----------- 7.7/11.0 MB 9.9 MB/s eta 0:00:01\n",
      "     ------------------------------ --------- 8.5/11.0 MB 10.0 MB/s eta 0:00:01\n",
      "     --------------------------------- ------ 9.2/11.0 MB 10.1 MB/s eta 0:00:01\n",
      "     ------------------------------------ --- 9.9/11.0 MB 10.2 MB/s eta 0:00:01\n",
      "     ------------------------------------- - 10.5/11.0 MB 10.7 MB/s eta 0:00:01\n",
      "     --------------------------------------  11.0/11.0 MB 10.7 MB/s eta 0:00:01\n",
      "     --------------------------------------  11.0/11.0 MB 10.7 MB/s eta 0:00:01\n",
      "     --------------------------------------  11.0/11.0 MB 10.7 MB/s eta 0:00:01\n",
      "     ---------------------------------------- 11.0/11.0 MB 8.8 MB/s eta 0:00:00\n",
      "Collecting threadpoolctl>=3.1.0\n",
      "  Downloading threadpoolctl-3.5.0-py3-none-any.whl (18 kB)\n",
      "Collecting joblib>=1.2.0\n",
      "  Downloading joblib-1.4.2-py3-none-any.whl (301 kB)\n",
      "     ---------------------------------------- 0.0/301.8 kB ? eta -:--:--\n",
      "     -------------------------------------- 301.8/301.8 kB 9.1 MB/s eta 0:00:00\n",
      "Collecting scipy>=1.6.0\n",
      "  Downloading scipy-1.13.1-cp310-cp310-win_amd64.whl (46.2 MB)\n",
      "     ---------------------------------------- 0.0/46.2 MB ? eta -:--:--\n",
      "     ---------------------------------------- 0.5/46.2 MB 11.3 MB/s eta 0:00:05\n",
      "      --------------------------------------- 1.1/46.2 MB 11.3 MB/s eta 0:00:04\n",
      "     - -------------------------------------- 1.6/46.2 MB 11.5 MB/s eta 0:00:04\n",
      "     - -------------------------------------- 2.2/46.2 MB 11.5 MB/s eta 0:00:04\n",
      "     -- ------------------------------------- 2.6/46.2 MB 11.1 MB/s eta 0:00:04\n",
      "     -- ------------------------------------- 3.0/46.2 MB 11.4 MB/s eta 0:00:04\n",
      "     -- ------------------------------------- 3.5/46.2 MB 11.0 MB/s eta 0:00:04\n",
      "     --- ------------------------------------ 4.2/46.2 MB 11.2 MB/s eta 0:00:04\n",
      "     ---- ----------------------------------- 4.8/46.2 MB 11.2 MB/s eta 0:00:04\n",
      "     ---- ----------------------------------- 5.3/46.2 MB 11.3 MB/s eta 0:00:04\n",
      "     ----- ---------------------------------- 5.9/46.2 MB 11.4 MB/s eta 0:00:04\n",
      "     ----- ---------------------------------- 6.4/46.2 MB 11.4 MB/s eta 0:00:04\n",
      "     ------ --------------------------------- 6.9/46.2 MB 11.4 MB/s eta 0:00:04\n",
      "     ------ --------------------------------- 7.5/46.2 MB 11.4 MB/s eta 0:00:04\n",
      "     ------ --------------------------------- 8.0/46.2 MB 11.4 MB/s eta 0:00:04\n",
      "     ------- -------------------------------- 8.6/46.2 MB 11.5 MB/s eta 0:00:04\n",
      "     ------- -------------------------------- 9.2/46.2 MB 11.5 MB/s eta 0:00:04\n",
      "     -------- ------------------------------- 9.7/46.2 MB 11.5 MB/s eta 0:00:04\n",
      "     -------- ------------------------------ 10.3/46.2 MB 11.7 MB/s eta 0:00:04\n",
      "     --------- ----------------------------- 10.8/46.2 MB 11.5 MB/s eta 0:00:04\n",
      "     --------- ----------------------------- 11.4/46.2 MB 11.5 MB/s eta 0:00:04\n",
      "     ---------- ---------------------------- 12.0/46.2 MB 11.5 MB/s eta 0:00:03\n",
      "     ---------- ---------------------------- 12.5/46.2 MB 11.7 MB/s eta 0:00:03\n",
      "     ----------- --------------------------- 13.0/46.2 MB 11.7 MB/s eta 0:00:03\n",
      "     ----------- --------------------------- 13.6/46.2 MB 11.7 MB/s eta 0:00:03\n",
      "     ----------- --------------------------- 14.1/46.2 MB 11.7 MB/s eta 0:00:03\n",
      "     ------------ -------------------------- 14.7/46.2 MB 11.7 MB/s eta 0:00:03\n",
      "     ------------ -------------------------- 15.2/46.2 MB 11.7 MB/s eta 0:00:03\n",
      "     ------------- ------------------------- 15.7/46.2 MB 11.9 MB/s eta 0:00:03\n",
      "     ------------- ------------------------- 16.5/46.2 MB 11.9 MB/s eta 0:00:03\n",
      "     -------------- ------------------------ 17.0/46.2 MB 11.9 MB/s eta 0:00:03\n",
      "     -------------- ------------------------ 17.6/46.2 MB 11.9 MB/s eta 0:00:03\n",
      "     --------------- ----------------------- 18.1/46.2 MB 11.9 MB/s eta 0:00:03\n",
      "     --------------- ----------------------- 18.7/46.2 MB 11.9 MB/s eta 0:00:03\n",
      "     ---------------- ---------------------- 19.4/46.2 MB 11.9 MB/s eta 0:00:03\n",
      "     ---------------- ---------------------- 20.0/46.2 MB 11.9 MB/s eta 0:00:03\n",
      "     ----------------- --------------------- 20.4/46.2 MB 11.7 MB/s eta 0:00:03\n",
      "     ----------------- --------------------- 21.0/46.2 MB 11.7 MB/s eta 0:00:03\n",
      "     ------------------ -------------------- 21.7/46.2 MB 11.7 MB/s eta 0:00:03\n",
      "     ------------------ -------------------- 22.3/46.2 MB 11.7 MB/s eta 0:00:03\n",
      "     ------------------- ------------------- 22.8/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     ------------------- ------------------- 23.4/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     -------------------- ------------------ 24.2/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     -------------------- ------------------ 24.7/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     --------------------- ----------------- 25.2/46.2 MB 11.5 MB/s eta 0:00:02\n",
      "     --------------------- ----------------- 25.7/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     ---------------------- ---------------- 26.3/46.2 MB 11.5 MB/s eta 0:00:02\n",
      "     ---------------------- ---------------- 26.8/46.2 MB 11.5 MB/s eta 0:00:02\n",
      "     ----------------------- --------------- 27.4/46.2 MB 11.5 MB/s eta 0:00:02\n",
      "     ----------------------- --------------- 28.0/46.2 MB 11.5 MB/s eta 0:00:02\n",
      "     ------------------------ -------------- 28.5/46.2 MB 11.5 MB/s eta 0:00:02\n",
      "     ------------------------ -------------- 29.1/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     ------------------------- ------------- 29.6/46.2 MB 11.5 MB/s eta 0:00:02\n",
      "     ------------------------- ------------- 30.2/46.2 MB 11.5 MB/s eta 0:00:02\n",
      "     ------------------------- ------------- 30.8/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     -------------------------- ------------ 31.3/46.2 MB 11.9 MB/s eta 0:00:02\n",
      "     --------------------------- ----------- 32.1/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     --------------------------- ----------- 32.7/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     ---------------------------- ---------- 33.2/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     ---------------------------- ---------- 33.8/46.2 MB 11.9 MB/s eta 0:00:02\n",
      "     ---------------------------- ---------- 34.3/46.2 MB 11.7 MB/s eta 0:00:02\n",
      "     ----------------------------- --------- 34.9/46.2 MB 11.7 MB/s eta 0:00:01\n",
      "     ----------------------------- --------- 35.4/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ------------------------------ -------- 36.0/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ------------------------------ -------- 36.6/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ------------------------------- ------- 37.0/46.2 MB 11.7 MB/s eta 0:00:01\n",
      "     ------------------------------- ------- 37.8/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     -------------------------------- ------ 38.4/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     -------------------------------- ------ 39.0/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------- ----- 39.5/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------- ----- 40.0/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ---------------------------------- ---- 40.8/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ---------------------------------- ---- 41.4/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ----------------------------------- --- 41.9/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ----------------------------------- --- 42.5/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ------------------------------------ -- 43.0/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ------------------------------------ -- 43.6/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ------------------------------------- - 44.1/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ------------------------------------- - 44.7/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  45.2/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  45.8/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  46.2/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  46.2/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  46.2/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  46.2/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  46.2/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  46.2/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  46.2/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  46.2/46.2 MB 11.9 MB/s eta 0:00:01\n",
      "     ---------------------------------------- 46.2/46.2 MB 7.4 MB/s eta 0:00:00\n",
      "Requirement already satisfied: numpy>=1.19.5 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from scikit-learn) (1.26.4)\n",
      "Installing collected packages: threadpoolctl, scipy, joblib, scikit-learn\n",
      "Successfully installed joblib-1.4.2 scikit-learn-1.5.0 scipy-1.13.1 threadpoolctl-3.5.0\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip is available: 23.0.1 -> 24.0\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting matplotlib\n",
      "  Downloading matplotlib-3.9.0-cp310-cp310-win_amd64.whl (8.0 MB)\n",
      "     ---------------------------------------- 0.0/8.0 MB ? eta -:--:--\n",
      "     - -------------------------------------- 0.3/8.0 MB 7.1 MB/s eta 0:00:02\n",
      "     ---- ----------------------------------- 0.8/8.0 MB 9.0 MB/s eta 0:00:01\n",
      "     ------- -------------------------------- 1.4/8.0 MB 9.9 MB/s eta 0:00:01\n",
      "     --------- ------------------------------ 1.9/8.0 MB 10.1 MB/s eta 0:00:01\n",
      "     ------------- -------------------------- 2.7/8.0 MB 10.9 MB/s eta 0:00:01\n",
      "     ----------------- ---------------------- 3.5/8.0 MB 10.8 MB/s eta 0:00:01\n",
      "     --------------------- ------------------ 4.2/8.0 MB 11.3 MB/s eta 0:00:01\n",
      "     ------------------------- -------------- 5.0/8.0 MB 11.0 MB/s eta 0:00:01\n",
      "     ---------------------------- ----------- 5.6/8.0 MB 11.1 MB/s eta 0:00:01\n",
      "     ------------------------------ --------- 6.0/8.0 MB 10.9 MB/s eta 0:00:01\n",
      "     ---------------------------------- ----- 6.8/8.0 MB 11.2 MB/s eta 0:00:01\n",
      "     ------------------------------------- -- 7.5/8.0 MB 10.9 MB/s eta 0:00:01\n",
      "     ---------------------------------------  7.9/8.0 MB 10.2 MB/s eta 0:00:01\n",
      "     ---------------------------------------- 8.0/8.0 MB 7.8 MB/s eta 0:00:00\n",
      "Collecting pillow>=8\n",
      "  Downloading pillow-10.3.0-cp310-cp310-win_amd64.whl (2.5 MB)\n",
      "     ---------------------------------------- 0.0/2.5 MB ? eta -:--:--\n",
      "     -------- ------------------------------- 0.6/2.5 MB 17.5 MB/s eta 0:00:01\n",
      "     ---------------------- ----------------- 1.4/2.5 MB 12.9 MB/s eta 0:00:01\n",
      "     -------------------------------- ------- 2.0/2.5 MB 12.9 MB/s eta 0:00:01\n",
      "     ---------------------------------------  2.5/2.5 MB 13.3 MB/s eta 0:00:01\n",
      "     ---------------------------------------- 2.5/2.5 MB 10.1 MB/s eta 0:00:00\n",
      "Requirement already satisfied: numpy>=1.23 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib) (1.26.4)\n",
      "Collecting pyparsing>=2.3.1\n",
      "  Downloading pyparsing-3.1.2-py3-none-any.whl (103 kB)\n",
      "     ---------------------------------------- 0.0/103.2 kB ? eta -:--:--\n",
      "     -------------------------------------- 103.2/103.2 kB 3.0 MB/s eta 0:00:00\n",
      "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib) (2.9.0.post0)\n",
      "Collecting kiwisolver>=1.3.1\n",
      "  Downloading kiwisolver-1.4.5-cp310-cp310-win_amd64.whl (56 kB)\n",
      "     ---------------------------------------- 0.0/56.1 kB ? eta -:--:--\n",
      "     ---------------------------------------- 56.1/56.1 kB 2.9 MB/s eta 0:00:00\n",
      "Requirement already satisfied: packaging>=20.0 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib) (24.0)\n",
      "Collecting fonttools>=4.22.0\n",
      "  Downloading fonttools-4.53.0-cp310-cp310-win_amd64.whl (2.2 MB)\n",
      "     ---------------------------------------- 0.0/2.2 MB ? eta -:--:--\n",
      "     --- ------------------------------------ 0.2/2.2 MB 10.2 MB/s eta 0:00:01\n",
      "     --- ------------------------------------ 0.2/2.2 MB 10.2 MB/s eta 0:00:01\n",
      "     ---------- ----------------------------- 0.6/2.2 MB 4.1 MB/s eta 0:00:01\n",
      "     ----------- ---------------------------- 0.6/2.2 MB 4.3 MB/s eta 0:00:01\n",
      "     ----------- ---------------------------- 0.6/2.2 MB 4.3 MB/s eta 0:00:01\n",
      "     -------------- ------------------------- 0.8/2.2 MB 2.8 MB/s eta 0:00:01\n",
      "     -------------------- ------------------- 1.1/2.2 MB 3.8 MB/s eta 0:00:01\n",
      "     ---------------------------- ----------- 1.5/2.2 MB 3.9 MB/s eta 0:00:01\n",
      "     ------------------------------- -------- 1.7/2.2 MB 4.2 MB/s eta 0:00:01\n",
      "     ------------------------------- -------- 1.7/2.2 MB 4.2 MB/s eta 0:00:01\n",
      "     -------------------------------- ------- 1.8/2.2 MB 3.4 MB/s eta 0:00:01\n",
      "     ---------------------------------------  2.2/2.2 MB 3.9 MB/s eta 0:00:01\n",
      "     ---------------------------------------- 2.2/2.2 MB 3.6 MB/s eta 0:00:00\n",
      "Collecting contourpy>=1.0.1\n",
      "  Downloading contourpy-1.2.1-cp310-cp310-win_amd64.whl (187 kB)\n",
      "     ---------------------------------------- 0.0/187.5 kB ? eta -:--:--\n",
      "     ------------------- -------------------- 92.2/187.5 kB ? eta -:--:--\n",
      "     ------------------- -------------------- 92.2/187.5 kB ? eta -:--:--\n",
      "     -------------------------------------  184.3/187.5 kB 1.4 MB/s eta 0:00:01\n",
      "     -------------------------------------- 187.5/187.5 kB 1.3 MB/s eta 0:00:00\n",
      "Collecting cycler>=0.10\n",
      "  Downloading cycler-0.12.1-py3-none-any.whl (8.3 kB)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from python-dateutil>=2.7->matplotlib) (1.16.0)\n",
      "Installing collected packages: pyparsing, pillow, kiwisolver, fonttools, cycler, contourpy, matplotlib\n",
      "Successfully installed contourpy-1.2.1 cycler-0.12.1 fonttools-4.53.0 kiwisolver-1.4.5 matplotlib-3.9.0 pillow-10.3.0 pyparsing-3.1.2\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip is available: 23.0.1 -> 24.0\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting pandasNote: you may need to restart the kernel to use updated packages.\n",
      "\n",
      "  Downloading pandas-2.2.2-cp310-cp310-win_amd64.whl (11.6 MB)\n",
      "     ---------------------------------------- 0.0/11.6 MB ? eta -:--:--\n",
      "     - -------------------------------------- 0.4/11.6 MB 8.1 MB/s eta 0:00:02\n",
      "     --- ------------------------------------ 0.9/11.6 MB 9.9 MB/s eta 0:00:02\n",
      "     ----- ---------------------------------- 1.5/11.6 MB 10.5 MB/s eta 0:00:01\n",
      "     ------- -------------------------------- 2.1/11.6 MB 11.1 MB/s eta 0:00:01\n",
      "     -------- ------------------------------- 2.6/11.6 MB 11.1 MB/s eta 0:00:01\n",
      "     ---------- ----------------------------- 3.2/11.6 MB 11.3 MB/s eta 0:00:01\n",
      "     ------------ --------------------------- 3.7/11.6 MB 11.3 MB/s eta 0:00:01\n",
      "     -------------- ------------------------- 4.3/11.6 MB 11.4 MB/s eta 0:00:01\n",
      "     ---------------- ----------------------- 4.9/11.6 MB 11.5 MB/s eta 0:00:01\n",
      "     ------------------ --------------------- 5.4/11.6 MB 11.6 MB/s eta 0:00:01\n",
      "     -------------------- ------------------- 6.0/11.6 MB 11.6 MB/s eta 0:00:01\n",
      "     ---------------------- ----------------- 6.5/11.6 MB 11.6 MB/s eta 0:00:01\n",
      "     ------------------------ --------------- 7.1/11.6 MB 11.6 MB/s eta 0:00:01\n",
      "     -------------------------- ------------- 7.6/11.6 MB 11.6 MB/s eta 0:00:01\n",
      "     ---------------------------- ----------- 8.2/11.6 MB 11.6 MB/s eta 0:00:01\n",
      "     ------------------------------ --------- 8.7/11.6 MB 11.6 MB/s eta 0:00:01\n",
      "     -------------------------------- ------- 9.3/11.6 MB 11.7 MB/s eta 0:00:01\n",
      "     --------------------------------- ------ 9.8/11.6 MB 11.6 MB/s eta 0:00:01\n",
      "     ---------------------------------- ---- 10.4/11.6 MB 11.9 MB/s eta 0:00:01\n",
      "     ------------------------------------ -- 10.9/11.6 MB 11.7 MB/s eta 0:00:01\n",
      "     --------------------------------------  11.5/11.6 MB 11.9 MB/s eta 0:00:01\n",
      "     --------------------------------------  11.6/11.6 MB 11.7 MB/s eta 0:00:01\n",
      "     --------------------------------------  11.6/11.6 MB 11.7 MB/s eta 0:00:01\n",
      "     --------------------------------------- 11.6/11.6 MB 10.4 MB/s eta 0:00:00\n",
      "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from pandas) (2.9.0.post0)\n",
      "Requirement already satisfied: numpy>=1.22.4 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from pandas) (1.26.4)\n",
      "Collecting tzdata>=2022.7\n",
      "  Downloading tzdata-2024.1-py2.py3-none-any.whl (345 kB)\n",
      "     ---------------------------------------- 0.0/345.4 kB ? eta -:--:--\n",
      "     ------------------------------------- 345.4/345.4 kB 10.8 MB/s eta 0:00:00\n",
      "Collecting pytz>=2020.1\n",
      "  Downloading pytz-2024.1-py2.py3-none-any.whl (505 kB)\n",
      "     ---------------------------------------- 0.0/505.5 kB ? eta -:--:--\n",
      "     ------------------------------------  501.8/505.5 kB 15.9 MB/s eta 0:00:01\n",
      "     ------------------------------------- 505.5/505.5 kB 10.5 MB/s eta 0:00:00\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from python-dateutil>=2.8.2->pandas) (1.16.0)\n",
      "Installing collected packages: pytz, tzdata, pandas\n",
      "Successfully installed pandas-2.2.2 pytz-2024.1 tzdata-2024.1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip is available: 23.0.1 -> 24.0\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting seaborn[stats]\n",
      "  Downloading seaborn-0.13.2-py3-none-any.whl (294 kB)\n",
      "     ---------------------------------------- 0.0/294.9 kB ? eta -:--:--\n",
      "     -------------------------------------- 294.9/294.9 kB 8.9 MB/s eta 0:00:00\n",
      "Requirement already satisfied: pandas>=1.2 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from seaborn[stats]) (2.2.2)\n",
      "Requirement already satisfied: matplotlib!=3.6.1,>=3.4 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from seaborn[stats]) (3.9.0)\n",
      "Requirement already satisfied: numpy!=1.24.0,>=1.20 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from seaborn[stats]) (1.26.4)\n",
      "Requirement already satisfied: scipy>=1.7 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from seaborn[stats]) (1.13.1)\n",
      "Collecting statsmodels>=0.12\n",
      "  Downloading statsmodels-0.14.2-cp310-cp310-win_amd64.whl (9.8 MB)\n",
      "     ---------------------------------------- 0.0/9.8 MB ? eta -:--:--\n",
      "     -- ------------------------------------- 0.5/9.8 MB 15.9 MB/s eta 0:00:01\n",
      "     ---- ----------------------------------- 1.0/9.8 MB 12.6 MB/s eta 0:00:01\n",
      "     ------ --------------------------------- 1.5/9.8 MB 12.0 MB/s eta 0:00:01\n",
      "     -------- ------------------------------- 2.1/9.8 MB 12.1 MB/s eta 0:00:01\n",
      "     -------- ------------------------------- 2.1/9.8 MB 12.3 MB/s eta 0:00:01\n",
      "     ----------- ---------------------------- 2.8/9.8 MB 10.6 MB/s eta 0:00:01\n",
      "     ------------- -------------------------- 3.4/9.8 MB 10.8 MB/s eta 0:00:01\n",
      "     --------------- ------------------------ 3.9/9.8 MB 10.9 MB/s eta 0:00:01\n",
      "     ------------------ --------------------- 4.5/9.8 MB 11.1 MB/s eta 0:00:01\n",
      "     -------------------- ------------------- 5.1/9.8 MB 11.1 MB/s eta 0:00:01\n",
      "     ---------------------- ----------------- 5.6/9.8 MB 11.2 MB/s eta 0:00:01\n",
      "     ------------------------ --------------- 6.2/9.8 MB 11.3 MB/s eta 0:00:01\n",
      "     --------------------------- ------------ 6.7/9.8 MB 11.2 MB/s eta 0:00:01\n",
      "     ----------------------------- ---------- 7.2/9.8 MB 11.3 MB/s eta 0:00:01\n",
      "     ------------------------------- -------- 7.8/9.8 MB 11.4 MB/s eta 0:00:01\n",
      "     ---------------------------------- ----- 8.4/9.8 MB 11.4 MB/s eta 0:00:01\n",
      "     ------------------------------------ --- 8.9/9.8 MB 11.4 MB/s eta 0:00:01\n",
      "     -------------------------------------- - 9.5/9.8 MB 11.5 MB/s eta 0:00:01\n",
      "     ---------------------------------------  9.8/9.8 MB 11.4 MB/s eta 0:00:01\n",
      "     ---------------------------------------- 9.8/9.8 MB 10.7 MB/s eta 0:00:00\n",
      "Requirement already satisfied: packaging>=20.0 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn[stats]) (24.0)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn[stats]) (2.9.0.post0)\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn[stats]) (0.12.1)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn[stats]) (4.53.0)\n",
      "Requirement already satisfied: contourpy>=1.0.1 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn[stats]) (1.2.1)\n",
      "Requirement already satisfied: pyparsing>=2.3.1 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn[stats]) (3.1.2)\n",
      "Requirement already satisfied: pillow>=8 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn[stats]) (10.3.0)\n",
      "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn[stats]) (1.4.5)\n",
      "Requirement already satisfied: tzdata>=2022.7 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from pandas>=1.2->seaborn[stats]) (2024.1)\n",
      "Requirement already satisfied: pytz>=2020.1 in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from pandas>=1.2->seaborn[stats]) (2024.1)\n",
      "Collecting patsy>=0.5.6\n",
      "  Downloading patsy-0.5.6-py2.py3-none-any.whl (233 kB)\n",
      "     ---------------------------------------- 0.0/233.9 kB ? eta -:--:--\n",
      "     ------------------------------------- 233.9/233.9 kB 14.9 MB/s eta 0:00:00\n",
      "Requirement already satisfied: six in c:\\users\\ppsha\\documents\\github\\algorithms\\env\\lib\\site-packages (from patsy>=0.5.6->statsmodels>=0.12->seaborn[stats]) (1.16.0)\n",
      "Installing collected packages: patsy, statsmodels, seaborn\n",
      "Successfully installed patsy-0.5.6 seaborn-0.13.2 statsmodels-0.14.2\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "[notice] A new release of pip is available: 23.0.1 -> 24.0\n",
      "[notice] To update, run: python.exe -m pip install --upgrade pip\n"
     ]
    }
   ],
   "source": [
    "%pip install numpy\n",
    "%pip install scikit-learn\n",
    "%pip install matplotlib\n",
    "%pip install pandas\n",
    "%pip install seaborn[stats]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The goal here is to create an efficient algorithmic way to determine what number an image represents."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Displaying number 5\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x2793c061c90>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.datasets import fetch_openml\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "mnist_data = fetch_openml('mnist_784', version=1, return_X_y=True)\n",
    "images, targets = mnist_data\n",
    "images = images.to_numpy()\n",
    "targets = targets.astype(int)\n",
    "\n",
    "\n",
    "# Visualizing an iamge\n",
    "image_idx = 0\n",
    "an_image = images[image_idx][:].reshape(28, 28)\n",
    "print(\"Displaying number\", targets[image_idx])\n",
    "plt.imshow(an_image) # Should display a 5 that looks like an S."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's display the images we will try to learn from."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 :  1\n",
      "1 :  3\n",
      "2 :  5\n",
      "3 :  7\n",
      "4 :  2\n",
      "5 :  0\n",
      "6 :  13\n",
      "7 :  15\n",
      "8 :  17\n",
      "9 :  4\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Let's get the first index of each number\n",
    "number_idx = {}\n",
    "for idx, number in enumerate(targets):\n",
    "  if number not in number_idx:\n",
    "    number_idx[number] = idx\n",
    "\n",
    "for number in range(10):\n",
    "  print(number, \": \", number_idx[number])\n",
    "\n",
    "# Modified from 3.5 AI code\n",
    "fig, axes = plt.subplots(2, 5, figsize=(12, 6))\n",
    "for number in range(10):\n",
    "  ax = axes[number // 5, number % 5]\n",
    "  ax.imshow(images[number_idx[number]][:].reshape(28, 28), cmap='gray')\n",
    "  ax.set_title(number)\n",
    "  ax.axis('off')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Error rate of 48.16714285714285 percent\n",
      "27969 : [0.1233  0.0837  0.1079  0.1221  0.1283  0.0896  0.0891  0.1017  0.07574\n",
      " 0.0786 ]\n",
      "44285 : [0.122   0.0915  0.10535 0.1192  0.11426 0.1259  0.1124  0.04813 0.07947\n",
      " 0.0817 ]\n",
      "864 : [0.12067 0.07697 0.10925 0.1416  0.115   0.0883  0.10614 0.08545 0.07605\n",
      " 0.0805 ]\n",
      "44272 : [0.1287  0.08044 0.1072  0.0991  0.11346 0.0862  0.1017  0.1103  0.0854\n",
      " 0.0876 ]\n",
      "66650 : [0.12054 0.08887 0.0997  0.129   0.10077 0.1288  0.109   0.0512  0.09625\n",
      " 0.07587]\n",
      "44999 : [0.0845  0.09064 0.10077 0.1056  0.1312  0.08405 0.124   0.1111  0.06995\n",
      " 0.09814]\n",
      "36878 : [0.09064 0.1096  0.08154 0.06134 0.1171  0.10425 0.1122  0.1255  0.08673\n",
      " 0.1112 ]\n",
      "30566 : [0.08606 0.10956 0.0953  0.10944 0.1477  0.1091  0.0942  0.0876  0.0751\n",
      " 0.08594]\n",
      "3967 : [0.12305 0.1149  0.09515 0.1114  0.1247  0.10736 0.09955 0.06067 0.0905\n",
      " 0.0727 ]\n",
      "2066 : [0.09534 0.1054  0.09656 0.09955 0.10736 0.0985  0.0825  0.10815 0.10486\n",
      " 0.10175]\n",
      "3435 : [0.1224  0.0732  0.11975 0.07477 0.116   0.0706  0.1224  0.1119  0.09705\n",
      " 0.092  ]\n",
      "59026 : [0.1361  0.08527 0.0952  0.125   0.11694 0.1125  0.08716 0.0995  0.07526\n",
      " 0.0671 ]\n",
      "49190 : [0.0957  0.0924  0.10913 0.09576 0.09985 0.0997  0.09894 0.10376 0.1017\n",
      " 0.1031 ]\n",
      "25361 : [0.1326  0.1003  0.0966  0.0881  0.09766 0.1157  0.10175 0.0928  0.11127\n",
      " 0.0633 ]\n",
      "23930 : [0.1317  0.0864  0.1118  0.1108  0.1022  0.08734 0.0984  0.10297 0.0926\n",
      " 0.0757 ]\n",
      "57324 : [0.10724 0.0969  0.0845  0.08044 0.1057  0.1106  0.1101  0.1118  0.10223\n",
      " 0.09045]\n",
      "42454 : [0.1339  0.0791  0.09564 0.1164  0.1015  0.0968  0.09406 0.0982  0.089\n",
      " 0.0954 ]\n",
      "34230 : [0.1256  0.1031  0.0952  0.09955 0.0883  0.088   0.1127  0.09564 0.10876\n",
      " 0.083  ]\n",
      "23055 : [0.0965  0.127   0.09106 0.1376  0.1008  0.11035 0.0733  0.07605 0.10626\n",
      " 0.08124]\n",
      "12486 : [0.1227  0.07715 0.1094  0.1294  0.1022  0.0994  0.10315 0.0989  0.0877\n",
      " 0.07007]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 20 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def mean_squared_error(y_true, y_pred):\n",
    "  return np.mean((y_true - y_pred) ** 2)\n",
    "\n",
    "def min_probe(index):\n",
    "  global number_idx\n",
    "  global images\n",
    "  probe_list = []\n",
    "  for number in range(10):\n",
    "    probe_list.append(mean_squared_error(images[number_idx[number]], images[index]))\n",
    "  min = np.Infinity\n",
    "  min_idx = None\n",
    "  for idx, score in enumerate(probe_list):\n",
    "    if score < min:\n",
    "      min = score\n",
    "      min_idx = idx\n",
    "  return min_idx, probe_list\n",
    "\n",
    "correct = 0\n",
    "incorrect = []\n",
    "for number in range(70000):\n",
    "  prediction, approximations = min_probe(number)\n",
    "  actual = targets[number]\n",
    "  if prediction == actual:\n",
    "    correct += 1\n",
    "  else:\n",
    "    incorrect.append((number, prediction, actual, approximations))\n",
    "  \n",
    "print(\"Error rate of\", 100*(1-correct/70000), \"%\")\n",
    "\n",
    "import random\n",
    "\n",
    "random.seed(0)\n",
    "shuffled_incorrects = incorrect[:]\n",
    "random.shuffle(shuffled_incorrects)\n",
    "\n",
    "amount_to_show = 20\n",
    "fig, axes = plt.subplots(4, 5, figsize=(12, 6))\n",
    "\n",
    "for idx, (image_idx, prediction, actual, approximations) in enumerate(shuffled_incorrects[:amount_to_show]):\n",
    "  ax = axes[idx // 5, idx % 5]\n",
    "  ax.imshow(images[image_idx][:].reshape(28, 28), cmap='gray')\n",
    "  ax.set_title(f\"Guessed {prediction}, Actual {actual}\")\n",
    "  ax.set_xlabel(f\"{image_idx}\")\n",
    "  print(image_idx, \":\", (approximations / np.sum(approximations)).astype('float16'))\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "An error rate of 48.167142857142853%. A good error rate is around 0.25%. Still, it was on only trained on 10 images, and it ran in approximately 10 seconds on CPU.\n",
    "\n",
    "The next goal is to take an already working implementation that runs fast, and convert it to use a data structure-centered approach to improve efficiency."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Regression : 6.453782320022583 seconds\n",
      "Error Rate:  5.44 %\n",
      "Logistic Regression : 56.95826554298401 seconds\n",
      "Error Rate:  10.119999999999996 %\n",
      "Support Vector Machine Classification : 34.233911752700806 seconds\n",
      "Error Rate:  3.8100000000000023 %\n",
      "Gradient Boosting Machines Classification : 382.7396128177643 seconds\n",
      "Error Rate:  7.5699999999999985 %\n",
      "K Neighbord Classifiers : 3.8780148029327393 seconds\n",
      "Error Rate:  5.479999999999996 %\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report \n",
    "import time\n",
    "sizes = 10000\n",
    "\n",
    "\n",
    "start_time = time.time()\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "randf_model = RandomForestClassifier(n_estimators=100, random_state=42)\n",
    "randf_model.fit(images[:sizes], targets[:sizes])\n",
    "randf_pred = randf_model.predict(images[sizes:2*sizes])\n",
    "end_time = time.time()\n",
    "print(\"Random Forest Regression\", \":\", end_time - start_time, \"seconds\")\n",
    "randf_report = classification_report(randf_pred, targets[sizes:2*sizes], output_dict=True, digits=6)\n",
    "print(\"Error Rate: \", 100*(1 - randf_report['accuracy']), \"%\")\n",
    "\n",
    "\n",
    "start_time = time.time()\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "logistic_model = LogisticRegression(C=50.0, tol=0.01, \n",
    "                           penalty=\"l1\", \n",
    "                           solver=\"saga\") \n",
    "logistic_model.fit(images[:sizes], targets[:sizes]) \n",
    "logistic_pred = logistic_model.predict(images[sizes:2*sizes])\n",
    "end_time = time.time()\n",
    "print(\"Logistic Regression\", \":\", end_time - start_time, \"seconds\")\n",
    "logistic_report = classification_report(logistic_pred, targets[sizes:2*sizes], output_dict=True, digits=6)\n",
    "print(\"Error Rate: \", 100*(1 - logistic_report['accuracy']), \"%\")\n",
    "\n",
    "\n",
    "start_time = time.time()\n",
    "from sklearn.svm import SVC\n",
    "svm_model = SVC(kernel='rbf', C=1.0, gamma='scale')\n",
    "svm_model.fit(images[:sizes], targets[:sizes])\n",
    "svm_pred = svm_model.predict(images[sizes:2*sizes])\n",
    "end_time = time.time()\n",
    "print(\"Support Vector Machine Classification\", \":\", end_time - start_time, \"seconds\")\n",
    "svm_report = classification_report(svm_pred, targets[sizes:2*sizes], output_dict=True, digits=6)\n",
    "print(\"Error Rate: \", 100*(1 - svm_report['accuracy']), \"%\")\n",
    "\n",
    "\n",
    "start_time = time.time()\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "gbm_model = GradientBoostingClassifier(n_estimators=100, learning_rate=0.1, random_state=42)\n",
    "gbm_model.fit(images[:sizes], targets[:sizes])\n",
    "gbm_pred = gbm_model.predict(images[sizes:2*sizes])\n",
    "end_time = time.time()\n",
    "print(\"Gradient Boosting Machines Classification\", \":\", end_time - start_time, \"seconds\")\n",
    "gbm_report = classification_report(gbm_pred, targets[sizes:2*sizes], output_dict=True, digits=6)\n",
    "print(\"Error Rate: \", 100*(1 - gbm_report['accuracy']), \"%\")\n",
    "\n",
    "\n",
    "start_time = time.time()\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "kn_model = KNeighborsClassifier(n_neighbors=5)\n",
    "kn_model.fit(images[:sizes], targets[:sizes])\n",
    "kn_pred = kn_model.predict(images[sizes:2*sizes])\n",
    "end_time = time.time()\n",
    "print(\"K Neighbors Classifiers\", \":\", end_time - start_time, \"seconds\")\n",
    "kn_report = classification_report(kn_pred, targets[sizes:2*sizes], output_dict=True, digits=6)\n",
    "print(\"Error Rate: \", 100*(1 - kn_report['accuracy']), \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The one that performed the best in my mind is K neighbors classifier. It's speed was the fastest and the accuracy is still fairly high."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "K Neighbors Classifiers : 67.12730121612549 seconds\n",
      "Error Rate:  4.4 %\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import time\n",
    "\n",
    "class KNN_Classifier:\n",
    "  def __init__(self, x_train, x_true, k):\n",
    "    self.x_train = x_train\n",
    "    self.x_true = x_true\n",
    "    self.k = k\n",
    "\n",
    "  def predict(self, vector):\n",
    "    distances = np.linalg.norm(self.x_train - vector[None, :], axis=-1)\n",
    "    kth_smallest = np.argpartition(distances, self.k - 1)[:self.k]\n",
    "    count = [0]*10\n",
    "    for idx in kth_smallest:\n",
    "      count[self.x_true[idx]] += 1\n",
    "    return np.argmax(count)\n",
    "\n",
    "\n",
    "sizes = 10000\n",
    "test_size = 1000\n",
    "correct = 0\n",
    "\n",
    "start_time = time.time()\n",
    "kn_model = KNN_Classifier(images[:sizes], targets[:sizes], 5)\n",
    "for image_idx in range(10000, 10000 + test_size):\n",
    "  kn_pred = kn_model.predict(images[image_idx])\n",
    "  # print(image_idx, end=\"\")\n",
    "  if kn_pred == targets[image_idx]:\n",
    "    correct += 1\n",
    "end_time = time.time()\n",
    "print(\"K Neighbors Classifiers\", \":\", end_time - start_time, \"seconds\")\n",
    "print(\"Error Rate: \", 100 * (test_size - correct) / test_size, \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- No Optimization 1\n",
    "\n",
    "K Neighbors Classifiers : 17.801165103912354 seconds\n",
    "\n",
    "Error Rate:  4.0 %\n",
    "\n",
    "- With Preallocation and views\n",
    "\n",
    "K Neighbors Classifiers : 9.664044618606567 seconds\n",
    "\n",
    "Error Rate:  4.0 %\n",
    "\n",
    "- With in-place calculation and views\n",
    "\n",
    "```distances = np.linalg.norm(self.x_train - vector[None, :], axis=-1)```\n",
    "\n",
    "K Neighbors Classifiers : 6.568434715270996 seconds\n",
    "\n",
    "Error Rate:  4.0 %\n",
    "\n",
    "For 1000 images with the same algorithm as above:\n",
    "\n",
    "K Neighbors Classifiers : 67.12730121612549 seconds\n",
    "\n",
    "Error Rate:  4.4 %\n",
    "\n",
    "- These are all slower than the original significantly for 10000 images:\n",
    "\n",
    "K Neighbord Classifiers : 3.8780148029327393 seconds\n",
    "\n",
    "Error Rate:  5.479999999999996 %"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((784,), (70000, 784))"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "images[0].shape, images.shape"
   ]
  }
 ],
 "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": "3.10.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}