Tobigs_Week4_Dimensonality_Reduction_Assignment2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 차원축소 Assignment2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MNIST data에 적용을 해보기\n",
    "\n",
    "mnist-original.mat 파일과 같은 위치에서 주피터 노트북을 열어주세요"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import numpy.linalg as lin\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sklearn.datasets import fetch_openml\n",
    "from scipy import io\n",
    "%matplotlib inline\n",
    "from mpl_toolkits.mplot3d import Axes3D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "mnist = io.loadmat('mnist-original.mat') #mnist 손글씨 데이터를 불러옵니다\n",
    "X = mnist['data'].T\n",
    "y = mnist['label'].T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 데이터 정보!\n",
    "- 7만개의 작은 숫자 이미지\n",
    "- 행 열이 반대로 되어있음 -> 전치\n",
    "- grayscale 28x28 pixel = 784 feature\n",
    "- 각 picel은 0~255의 값\n",
    "- label = 1~10 label이 총 10개인거에 주목하자"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(70000, 784) (70000, 1)\n"
     ]
    }
   ],
   "source": [
    "print(X.shape, y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pixel0</th>\n",
       "      <th>pixel1</th>\n",
       "      <th>pixel2</th>\n",
       "      <th>pixel3</th>\n",
       "      <th>pixel4</th>\n",
       "      <th>pixel5</th>\n",
       "      <th>pixel6</th>\n",
       "      <th>pixel7</th>\n",
       "      <th>pixel8</th>\n",
       "      <th>pixel9</th>\n",
       "      <th>...</th>\n",
       "      <th>pixel774</th>\n",
       "      <th>pixel775</th>\n",
       "      <th>pixel776</th>\n",
       "      <th>pixel777</th>\n",
       "      <th>pixel778</th>\n",
       "      <th>pixel779</th>\n",
       "      <th>pixel780</th>\n",
       "      <th>pixel781</th>\n",
       "      <th>pixel782</th>\n",
       "      <th>pixel783</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 784 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   pixel0  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  pixel8  \\\n",
       "0       0       0       0       0       0       0       0       0       0   \n",
       "1       0       0       0       0       0       0       0       0       0   \n",
       "2       0       0       0       0       0       0       0       0       0   \n",
       "3       0       0       0       0       0       0       0       0       0   \n",
       "4       0       0       0       0       0       0       0       0       0   \n",
       "\n",
       "   pixel9  ...  pixel774  pixel775  pixel776  pixel777  pixel778  pixel779  \\\n",
       "0       0  ...         0         0         0         0         0         0   \n",
       "1       0  ...         0         0         0         0         0         0   \n",
       "2       0  ...         0         0         0         0         0         0   \n",
       "3       0  ...         0         0         0         0         0         0   \n",
       "4       0  ...         0         0         0         0         0         0   \n",
       "\n",
       "   pixel780  pixel781  pixel782  pixel783  \n",
       "0         0         0         0         0  \n",
       "1         0         0         0         0  \n",
       "2         0         0         0         0  \n",
       "3         0         0         0         0  \n",
       "4         0         0         0         0  \n",
       "\n",
       "[5 rows x 784 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# data를 각 픽셀에 이름붙여 표현\n",
    "feat_cols = [ 'pixel'+str(i) for i in range(X.shape[1]) ]\n",
    "df = pd.DataFrame(X,columns=feat_cols)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Size of the dataframe: (70000, 785)\n"
     ]
    }
   ],
   "source": [
    "# df에 라벨 y를 붙여서 데이터프레임 생성\n",
    "df['y'] = y\n",
    "print('Size of the dataframe: {}'.format(df.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pixel0</th>\n",
       "      <th>pixel1</th>\n",
       "      <th>pixel2</th>\n",
       "      <th>pixel3</th>\n",
       "      <th>pixel4</th>\n",
       "      <th>pixel5</th>\n",
       "      <th>pixel6</th>\n",
       "      <th>pixel7</th>\n",
       "      <th>pixel8</th>\n",
       "      <th>pixel9</th>\n",
       "      <th>...</th>\n",
       "      <th>pixel775</th>\n",
       "      <th>pixel776</th>\n",
       "      <th>pixel777</th>\n",
       "      <th>pixel778</th>\n",
       "      <th>pixel779</th>\n",
       "      <th>pixel780</th>\n",
       "      <th>pixel781</th>\n",
       "      <th>pixel782</th>\n",
       "      <th>pixel783</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69995</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69996</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69997</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69998</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69999</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>70000 rows × 785 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       pixel0  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  pixel8  \\\n",
       "0           0       0       0       0       0       0       0       0       0   \n",
       "1           0       0       0       0       0       0       0       0       0   \n",
       "2           0       0       0       0       0       0       0       0       0   \n",
       "3           0       0       0       0       0       0       0       0       0   \n",
       "4           0       0       0       0       0       0       0       0       0   \n",
       "...       ...     ...     ...     ...     ...     ...     ...     ...     ...   \n",
       "69995       0       0       0       0       0       0       0       0       0   \n",
       "69996       0       0       0       0       0       0       0       0       0   \n",
       "69997       0       0       0       0       0       0       0       0       0   \n",
       "69998       0       0       0       0       0       0       0       0       0   \n",
       "69999       0       0       0       0       0       0       0       0       0   \n",
       "\n",
       "       pixel9  ...  pixel775  pixel776  pixel777  pixel778  pixel779  \\\n",
       "0           0  ...         0         0         0         0         0   \n",
       "1           0  ...         0         0         0         0         0   \n",
       "2           0  ...         0         0         0         0         0   \n",
       "3           0  ...         0         0         0         0         0   \n",
       "4           0  ...         0         0         0         0         0   \n",
       "...       ...  ...       ...       ...       ...       ...       ...   \n",
       "69995       0  ...         0         0         0         0         0   \n",
       "69996       0  ...         0         0         0         0         0   \n",
       "69997       0  ...         0         0         0         0         0   \n",
       "69998       0  ...         0         0         0         0         0   \n",
       "69999       0  ...         0         0         0         0         0   \n",
       "\n",
       "       pixel780  pixel781  pixel782  pixel783    y  \n",
       "0             0         0         0         0  0.0  \n",
       "1             0         0         0         0  0.0  \n",
       "2             0         0         0         0  0.0  \n",
       "3             0         0         0         0  0.0  \n",
       "4             0         0         0         0  0.0  \n",
       "...         ...       ...       ...       ...  ...  \n",
       "69995         0         0         0         0  9.0  \n",
       "69996         0         0         0         0  9.0  \n",
       "69997         0         0         0         0  9.0  \n",
       "69998         0         0         0         0  9.0  \n",
       "69999         0         0         0         0  9.0  \n",
       "\n",
       "[70000 rows x 785 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x700 with 15 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 데이터 형태 시각화\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "rndperm = np.random.permutation(df.shape[0])\n",
    "\n",
    "# Plot the graph\n",
    "plt.gray()\n",
    "fig = plt.figure( figsize=(16,7) )\n",
    "for i in range(0,15):\n",
    "    ax = fig.add_subplot(3,5,i+1, title=\"Digit: {}\".format(str(df.loc[rndperm[i],'y'])) )\n",
    "    ax.matshow(df.loc[rndperm[i],feat_cols].values.reshape((28,28)).astype(float))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1) train_test_split을 통해 데이터를 0.8 0.2의 비율로 분할 해 주세요"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,y, test_size=0.2, random_state=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# PCA를 진행하기 위해 train data scaling 먼저 진행\n",
    "# train set은 fit 진행 후 transform\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "scaler = StandardScaler()\n",
    "scaler.fit(X_train)\n",
    "X_train_std = scaler.transform(X_train)\n",
    "\n",
    "# test set은 transfrom만 진행\n",
    "X_test_std = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 여기서부터는 원본데이터 & PCA 축소 데이터 & LDA 축소 데이터 비교해주세요"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2) 차원축소 기법(PCA와 LDA)을 이용하여 mnist data를 축소시켜주세요\n",
    "\n",
    "pca를 이용할 때는, 주성분 개수를 정하는 과정에 대해 잘 서술해주시면 좋겠죠!<br>\n",
    "강의에서 배웠던 3가지 중 어떤 걸 고려해서 갯수를 정했는지요!!!!<br>\n",
    "scree plot같은거는 직접 그려서 확인해주면 좋겠죠???"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### `주성분 개수 결정`\n",
    "\n",
    "1. `Elbow piont : 곡선의 기울기가 급격히 감소하는 지점`\n",
    "2. Kaiser's Rule : 고유값 1이상의 주성분들\n",
    "3. `누적 설명률이 70~80 이상인 지점`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "145"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "\n",
    "# 누적 설명률 판단\n",
    "\n",
    "pca = PCA(n_components= 0.8)\n",
    "pca.fit(X_train_std)\n",
    "pca.n_components_\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "118"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca = PCA(n_components=0.75)\n",
    "pca.fit(X_train_std)\n",
    "pca.n_components_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`스케일링을 한 데이터의 주성분은 118개가 적당하고 판단`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 엘보우 포인트 판단\n",
    "\n",
    "pca = PCA()\n",
    "X_pca = pca.fit_transform(X_train_std)\n",
    "\n",
    "exp_var_ratio = pca.explained_variance_ratio_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Scree plot 그리기\n",
    "\n",
    "plt.plot(range(len(exp_var_ratio)), exp_var_ratio, marker = '.')\n",
    "plt.title(\"Scree plot\")\n",
    "plt.xlabel(\"Number\")\n",
    "plt.ylabel(\"Eigenvalues\")\n",
    "\n",
    "#kneelocater 이용 -> elbow point 찾기\n",
    "from kneed import KneeLocator\n",
    "\n",
    "kn = KneeLocator(range(len(exp_var_ratio)), exp_var_ratio, curve = 'convex', direction = 'decreasing')\n",
    "plt.vlines(kn.knee, plt.ylim()[0], plt.ylim()[1], linestyles= 'dashdot', colors='r')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Elbow Point : 42\n"
     ]
    }
   ],
   "source": [
    "print(\"Elbow Point :\", kn.knee) # 엘보우 포인트 확인"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# LDA 이용\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "\n",
    "# LDA - 지도학습(분류)이므로 fit 할 때 y_train도 같이 학습\n",
    "# 따라서 y_train 학습 시 1차원 배열 형태로 학습 시킴 (np.ravel 이용)\n",
    "lda = LinearDiscriminantAnalysis()\n",
    "X_train_lda = lda.fit_transform(X_train_std, np.ravel(y_train))\n",
    "X_test_lda = lda.transform(X_test)\n",
    "\n",
    "# 분산 비율\n",
    "var_ratio = lda.explained_variance_ratio_\n",
    "\n",
    "# SCree plot 그리기\n",
    "plt.plot(np.cumsum(var_ratio))\n",
    "plt.xlabel(\"Number\")\n",
    "plt.ylabel(\"Cumlative explained variance\")\n",
    "plt.title(\"Scree Plot\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3) 지금까지 배웠던 머신러닝 기법을 이용하여 학습해주세요 (2개이상 사용)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  이때 time stamp를 찍어서 training 시간을 비교하고, test accuracy도 비교해주세요\n",
    "#### (원본 데이터 & PCA 축소 데이터 & LDA 축소 데이터 비교)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 로지스틱회귀분석"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\gjm10\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\sklearn\\utils\\validation.py:1184: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "정확도 평가 : 0.9244285714285714\n",
      "소요시간 : 8.220218420028687\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\gjm10\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\sklearn\\linear_model\\_logistic.py:460: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    }
   ],
   "source": [
    "# Logistic Regression 이용\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "import time\n",
    "\n",
    "lr = LogisticRegression(random_state=0)\n",
    "\n",
    "start = time.time()\n",
    "lr.fit(X_train, y_train)\n",
    "end = time.time()\n",
    "pred = lr.predict(X_test)\n",
    "\n",
    "print(f\"정확도 평가 : {accuracy_score(pred, y_test)}\")\n",
    "print(f\"소요시간 : {end-start}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "원본 데이터로 로지스틱회귀분석 진행"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\gjm10\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\sklearn\\utils\\validation.py:1184: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "정확도 평가 : 0.7777142857142857\n",
      "소요시간 : 8.527161121368408\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\gjm10\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\sklearn\\linear_model\\_logistic.py:460: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    }
   ],
   "source": [
    "# pca에 대한 데이터로 학습 진행\n",
    "lr = LogisticRegression(random_state=0)\n",
    "\n",
    "start = time.time()\n",
    "lr.fit(X_train, y_train)\n",
    "end = time.time()\n",
    "pred = lr.predict(X_test_std)\n",
    "\n",
    "print(f\"정확도 평가 : {accuracy_score(pred, y_test)}\")\n",
    "print(f\"소요시간 : {end-start}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* 원본 데이터로 로지스틱회귀분석을 진행했을 때, 정확도 측면에서 더 나은 결과를 확인할 수 있다.\n",
    "* 시간적인 부분에서는 크게 차이나지 않는다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## knn 진행"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\gjm10\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\sklearn\\neighbors\\_classification.py:228: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  return self._fit(X, y)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "정확도 평가 : 0.9714285714285714\n",
      "소요시간 : 0.0\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "kn = KNeighborsClassifier(n_neighbors=5)\n",
    "\n",
    "start = time.time()\n",
    "kn.fit(X_train, y_train)\n",
    "end = time.time()\n",
    "pred = kn.predict(X_test)\n",
    "\n",
    "print(f\"정확도 평가 : {accuracy_score(pred, y_test)}\")\n",
    "print(f\"소요시간 : {end-start}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "원본 데이터로 knn 진행했을 때, 소요시간은 약 22~23초 사이를 나타낸다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\gjm10\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\sklearn\\neighbors\\_classification.py:228: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  return self._fit(X, y)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "정확도 평가 : 0.11278571428571428\n",
      "소요시간 : 0.0037059783935546875\n"
     ]
    }
   ],
   "source": [
    "# pca에 대한 데이터로 학습 진행\n",
    "kn = KNeighborsClassifier(n_neighbors=5)\n",
    "\n",
    "start = time.time()\n",
    "kn.fit(X_train, y_train)\n",
    "end = time.time()\n",
    "pred = kn.predict(X_test_std)\n",
    "\n",
    "print(f\"정확도 평가 : {accuracy_score(pred, y_test)}\")\n",
    "print(f\"소요시간 : {end-start}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PCA 데이터로 knn 진행했을 때, 소요시간은 약 23~24초 사이를 나타낸다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* 원본 데이터로 KNN 진행했을 때와 pca 데이터로 KNN 진행했을 때 정확도 측면에서 큰 차이를 나타낸다. (원본데이터 : 0.971, pca데이터 : 0.112)\n",
    "* 시간적인 부분은 큰 차이를 보이지 않는 것을 확인할 수 있다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "``결론적으로 각 머신러닝 기법을 대입시켜서 원본데이터와 pca 데이터의 정확도를 비교했을 때, 원본데이터에 대해서 더 나은 정확도를 나타냈다.``"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "특정한 틀 없이 자유롭게 해주시면 됩니다!"
   ]
  }
 ],
 "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}