Chapter3.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['data', 'target', 'frame', 'categories', 'feature_names', 'target_names', 'DESCR', 'details', 'url'])"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.datasets import fetch_openml\n",
    "mnist = fetch_openml('mnist_784', version=1)\n",
    "mnist.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(70000, 784)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X,y = mnist[\"data\"],mnist[\"target\"]\n",
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(70000,)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        5\n",
       "1        0\n",
       "2        4\n",
       "3        1\n",
       "4        9\n",
       "        ..\n",
       "69995    2\n",
       "69996    3\n",
       "69997    4\n",
       "69998    5\n",
       "69999    6\n",
       "Name: class, Length: 70000, dtype: category\n",
       "Categories (10, object): ['0', '1', '2', '3', ..., '6', '7', '8', '9']"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## X[36000]으로 row를 뽑아내는데 에러가 나서 .loc( )를 이용하여 인덱싱하고 코드 수정함"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "### 분명 강의랑 똑같은 방식으로 가져 왔는데 some_digit=X[36000]이 아예 먹히질 않음\n",
    "### 이전에도 X[0]로 진행하였을때 에러가 없었는데 재설치해도 여전히 같은 오류 발생\n",
    "### 일단 .loc()를 이용하여 인덱싱 하여서 코드를 중간마다 약간씩 수정함\n",
    "\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "some_digit = X.loc[0]\n",
    "some_digit_image=some_digit.values.reshape(28, 28)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pixel1      0.0\n",
       "pixel2      0.0\n",
       "pixel3      0.0\n",
       "pixel4      0.0\n",
       "pixel5      0.0\n",
       "           ... \n",
       "pixel780    0.0\n",
       "pixel781    0.0\n",
       "pixel782    0.0\n",
       "pixel783    0.0\n",
       "pixel784    0.0\n",
       "Name: 0, Length: 784, dtype: float64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "some_digit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(some_digit_image,cmap = mpl.cm.binary, interpolation=\"nearest\")\n",
    "plt.axis(\"off\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'5'"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = y.astype(np.uint8)\n",
    "#분류시 데이터의 라벨값은 string형보다는 int형 선호\n",
    "#str -> int형 변환"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = X[:60000], X[60000:], y[:60000], y[60000:]\n",
    "#mnistdata는 이미 60000 이전,이후로 data가 split되어있음"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_5 = (y_train == 5)\n",
    "y_test_5 = (y_test == 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SGDClassifier(random_state=42)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import SGDClassifier\n",
    "\n",
    "sgd_clf = SGDClassifier(random_state=42)\n",
    "sgd_clf.fit(X_train, y_train_5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>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>pixel10</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>pixel784</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</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>59995</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59996</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59997</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59998</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59999</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>60000 rows × 784 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  pixel8  pixel9  \\\n",
       "0         0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "1         0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "2         0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "3         0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "4         0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "...       ...     ...     ...     ...     ...     ...     ...     ...     ...   \n",
       "59995     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "59996     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "59997     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "59998     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "59999     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "\n",
       "       pixel10  ...  pixel775  pixel776  pixel777  pixel778  pixel779  \\\n",
       "0          0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "1          0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "2          0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "3          0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "4          0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "...        ...  ...       ...       ...       ...       ...       ...   \n",
       "59995      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "59996      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "59997      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "59998      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "59999      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "\n",
       "       pixel780  pixel781  pixel782  pixel783  pixel784  \n",
       "0           0.0       0.0       0.0       0.0       0.0  \n",
       "1           0.0       0.0       0.0       0.0       0.0  \n",
       "2           0.0       0.0       0.0       0.0       0.0  \n",
       "3           0.0       0.0       0.0       0.0       0.0  \n",
       "4           0.0       0.0       0.0       0.0       0.0  \n",
       "...         ...       ...       ...       ...       ...  \n",
       "59995       0.0       0.0       0.0       0.0       0.0  \n",
       "59996       0.0       0.0       0.0       0.0       0.0  \n",
       "59997       0.0       0.0       0.0       0.0       0.0  \n",
       "59998       0.0       0.0       0.0       0.0       0.0  \n",
       "59999       0.0       0.0       0.0       0.0       0.0  \n",
       "\n",
       "[60000 rows x 784 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "### X_train이 int형 index가 아닌 다른값으로 반환된는것 같아 수정 필요\n",
    "\n",
    "# from sklearn.model_selection import StratifiedKFold\n",
    "# from sklearn.base import clone\n",
    "\n",
    "# skfolds = StratifiedKFold(n_splits=3, random_state=42, shuffle=True)\n",
    "\n",
    "# for train_index, test_index in skfolds.split(X_train, y_train_5):\n",
    "#     clone_clf = clone(sgd_clf)\n",
    "#     X_train_folds = X_train[train_index]\n",
    "#     y_train_folds = y_train_5[train_index]\n",
    "#     X_test_fold = X_train[test_index]\n",
    "#     y_test_fold = y_train_5[test_index]\n",
    "\n",
    "#     clone_clf.fit(X_train_folds, y_train_folds)\n",
    "#     y_pred = clone_clf.predict(X_test_fold)\n",
    "#     n_correct = sum(y_pred == y_test_fold)\n",
    "#     print(n_correct / len(y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.95035, 0.96035, 0.9604 ])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "cross_val_score(sgd_clf, X_train, y_train_5, cv=3, scoring=\"accuracy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.base import BaseEstimator\n",
    "\n",
    "class Never5Classifier(BaseEstimator):\n",
    "    def fit(self, X, y=None):\n",
    "        pass\n",
    "    def predict(self, X):\n",
    "        return np.zeros((len(X), 1), dtype=bool)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.91125, 0.90855, 0.90915])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "never_5_clf = Never5Classifier()\n",
    "cross_val_score(never_5_clf, X_train, y_train_5, cv=3, scoring=\"accuracy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_predict\n",
    "\n",
    "y_train_pred = cross_val_predict(sgd_clf, X_train, y_train_5, cv=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[53892,   687],\n",
       "       [ 1891,  3530]], dtype=int64)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "confusion_matrix(y_train_5, y_train_pred)\n",
    "#오차행렬에서 행은 실제 클래스, 열은 예측한 클래스임\n",
    "# ex) 53892개의 train_data가 5가아니며 예측결과 5가 아니고 687개는 실제로 5가 아니지만 예측시 5로 분류함"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[54579,     0],\n",
       "       [    0,  5421]], dtype=int64)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_perfect_predictions = y_train_5\n",
    "confusion_matrix(y_train_5, y_train_perfect_predictions)\n",
    "#예측결과 100% 정확한 예"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8370879772350012"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score, recall_score\n",
    "precision_score(y_train_5, y_train_pred)\n",
    "#687/ (687 + 3530)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6511713705958311"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(y_train_5, y_train_pred)\n",
    "#1891 / (1891 + 3530)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7325171197343846"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import f1_score\n",
    "f1_score(y_train_5, y_train_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2164.22030239])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_scores = sgd_clf.decision_function([some_digit])\n",
    "y_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True])"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "threshold=0\n",
    "y_some_digit_pred = (y_scores > threshold)\n",
    "y_some_digit_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "threshold=2164\n",
    "y_some_digit_pred = (y_scores > threshold)\n",
    "y_some_digit_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_scores = cross_val_predict(sgd_clf, X_train, y_train_5, cv=3, method=\"decision_function\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import precision_recall_curve\n",
    "\n",
    "precisions, recalls, thresholds = precision_recall_curve(y_train_5, y_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_precision_recall_vs_threshold(precisions, reacalls, thresholds):\n",
    "    plt.plot(thresholds, precisions[:-1], \"b--\", label=\"Precision\")\n",
    "    plt.plot(thresholds, recalls[:-1], \"g--\", label=\"Recall\")\n",
    "    plt.xlabel(\"Thresholds\")\n",
    "    plt.legend(loc=\"center left\")\n",
    "    plt.ylim([0,1])\n",
    "    \n",
    "plot_precision_recall_vs_threshold(precisions, recalls, thresholds)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "threshold_90_precision = thresholds[np.argmax(precisions >= 0.90)]\n",
    "y_train_pred_90 = (y_scores >= threshold_90_precision)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9000345901072293"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision_score(y_train_5, y_train_pred_90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4799852425751706"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(y_train_5, y_train_pred_90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "\n",
    "fpr, tpr, thresholds = roc_curve(y_train_5, y_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_roc_curve(fpr, tpr, label=None):\n",
    "    plt.plot(fpr, tpr, linewidth=2, label=label)\n",
    "    plt.plot([0,1], [0,1], \"k--\")\n",
    "    plt.xlabel(\"FPR\")\n",
    "    plt.ylabel(\"TPR\")\n",
    "plot_roc_curve(fpr, tpr)\n",
    "plt.show()\n",
    "#거짓 양성 높을수록 진짜 양성 높음\n",
    "# -> 재현율과 거짓양성은 비례관계"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9604938554008616"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "roc_auc_score(y_train_5, y_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "forest_clf = RandomForestClassifier(random_state=42)\n",
    "y_probas_forest = cross_val_predict(forest_clf, X_train, y_train_5, cv=3, method=\"predict_proba\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_scores_forest = y_probas_forest[:,1]\n",
    "fpr_forest, tpr_forest, thresholds_forest = roc_curve(y_train_5, y_scores_forest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(fpr, tpr, \"b:\", label=\"SGD\")\n",
    "plot_roc_curve(fpr_forest, tpr_forest, \"Random Forest\")\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9983436731328145"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(y_train_5, y_scores_forest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([3], dtype=uint8)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf.fit(X_train, y_train)\n",
    "sgd_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-31893.03095419, -34419.69069632,  -9530.63950739,\n",
       "          1823.73154031, -22320.14822878,  -1385.80478895,\n",
       "        -26188.91070951, -16147.51323997,  -4604.35491274,\n",
       "        -12050.767298  ]])"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "some_digit_scores = sgd_clf.decision_function([some_digit])\n",
    "some_digit_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.argmax(some_digit_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=uint8)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf.classes_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sgd_clf.classes_[3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5], dtype=uint8)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.multiclass import OneVsOneClassifier\n",
    "ovo_clf = OneVsOneClassifier(SGDClassifier(random_state=42))\n",
    "ovo_clf.fit(X_train, y_train)\n",
    "ovo_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "45"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(ovo_clf.estimators_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5], dtype=uint8)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest_clf.fit(X_train, y_train)\n",
    "forest_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.  , 0.  , 0.01, 0.08, 0.  , 0.9 , 0.  , 0.  , 0.  , 0.01]])"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest_clf.predict_proba([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.87365, 0.85835, 0.8689 ])"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(sgd_clf, X_train, y_train, cv=3, scoring=\"accuracy\",n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.8983, 0.891 , 0.9018])"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "X_train_scaled = scaler.fit_transform(X_train.astype(np.float64))\n",
    "cross_val_score(sgd_clf, X_train_scaled, y_train, cv=3, scoring=\"accuracy\", n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "############################# 원래 cv=3 이지만 평가 굉장히 오래걸려 변경\n",
    "y_train_pred = cross_val_predict(sgd_clf, X_train_scaled, y_train, cv=2, n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[5601,    0,   26,    7,   11,   53,   39,    6,  179,    1],\n",
       "       [   0, 6427,   45,   21,    4,   48,    4,   11,  171,   11],\n",
       "       [  31,   32, 5279,   90,   65,   25,   74,   46,  303,   13],\n",
       "       [  29,   24,  125, 5238,    2,  224,   26,   45,  337,   81],\n",
       "       [  15,   14,   52,   12, 5246,   10,   39,   24,  262,  168],\n",
       "       [  33,   17,   32,  173,   50, 4486,   79,   23,  452,   76],\n",
       "       [  29,   19,   46,    2,   48,   95, 5560,    5,  113,    1],\n",
       "       [  18,   12,   58,   27,   53,    9,    4, 5731,  130,  223],\n",
       "       [  19,   67,   57,   82,    4,  122,   31,   10, 5395,   64],\n",
       "       [  22,   17,   36,   62,  133,   28,    1,  182,  287, 5181]],\n",
       "      dtype=int64)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "conf_mx = confusion_matrix(y_train, y_train_pred)\n",
    "conf_mx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPoAAAECCAYAAADXWsr9AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAALMElEQVR4nO3dz4vc9R3H8ddrf8S4SUVjezErTSLFVgJ1ZS1qwIPx0FZRhB4sRqiXvbQaRRDtxX9ARA9FWGK9GPQQA4oUa0E99BK6JpGYrBVRm0QjpoFGUdjN7rx7mJEmu1vnO+73s98Z388HCGadfHyzyTPfmc133uuIEIDvt6GmBwBQHqEDCRA6kAChAwkQOpAAoQMJNBa67V/a/qftD2w/2tQcVdm+0vabtmdtH7W9u+mZqrA9bPuQ7VebnqUK25fa3mf7vc7n+samZ+rG9kOd3xPv2n7B9vqmZ1qqkdBtD0v6k6RfSbpG0m9tX9PELD1YkPRwRPxM0g2Sfj8AM0vSbkmzTQ/Rg6clvRYRP5X0c/X57LY3S3pA0mREbJc0LOnuZqdarqkr+i8kfRARH0bEvKQXJd3Z0CyVRMSpiDjY+fcv1f4NuLnZqb6d7XFJt0na0/QsVdi+RNLNkp6VpIiYj4j/NDpUNSOSLrY9ImlM0qcNz7NMU6FvlnTivB+fVJ9Hcz7bWyRNSDrQ8CjdPCXpEUmthueoapuk05Ke67zc2GN7Q9NDfZuI+ETSE5KOSzol6WxEvN7sVMs1FbpX+NhA3Itre6OklyQ9GBFfND3P/2P7dkmfR8TbTc/SgxFJ10l6JiImJH0lqa+/fmP7MrWfjW6VdIWkDbZ3NTvVck2FflLSlef9eFx9+HRnKdujake+NyL2Nz1PFzsk3WH7Y7VfGt1i+/lmR+rqpKSTEfHNM6V9aoffz26V9FFEnI6Ic5L2S7qp4ZmWaSr0f0j6ie2tttep/cWLVxqapRLbVvu142xEPNn0PN1ExGMRMR4RW9T+/L4REX13pTlfRHwm6YTtqzsf2inpWIMjVXFc0g22xzq/R3aqD7+AONLE/zQiFmz/QdJf1f4q5Z8j4mgTs/Rgh6R7JR2xfbjzsT9GxF+aG+l76X5JezsXgA8l3dfwPN8qIg7Y3ifpoNp/M3NI0nSzUy1n3qYKfP9xZxyQAKEDCRA6kAChAwkQOpBA46Hbnmp6hl4M2rwSM6+Ffp+38dAl9fUnaAWDNq/EzGuhr+fth9ABFFbkhplNmzbF+Ph4pceeOXNGl19+eaXHHjlyZDVjAT1r39XaXURUfuz5P6eEiFg2SJFbYMfHx/XKK/Xfur5169baz/xGr79IVQ3inYdDQ2We6JX8XJQ6e926dUXOlaS5ubliZy/FU3cgAUIHEiB0IAFCBxIgdCCBSqEP2g52ABfqGvqA7mAHcJ4qV/SB28EO4EJVQh/oHewAqoVeaQe77SnbM7Znzpw5s/rJANSmSuiVdrBHxHRETEbEZNV71wGsjSqhD9wOdgAX6vqmlgHdwQ7gPJXevdb5JgV8owJgQHFnHJAAoQMJEDqQAKEDCRA6kECR5ZC2iyzwKrlzbBD3pJXC/rz/Kbkzbn5+vsi5Ky2H5IoOJEDoQAKEDiRA6EAChA4kQOhAAoQOJEDoQAKEDiRA6EAChA4kQOhAAoQOJEDoQAKEDiRA6EAChA4kQOhAAoQOJEDoQAKEDiRA6EAClb7J4ndRYmVwqZXMkvTOO+8UOXdiYqLIuVK59cmtVqvIuSV//Up9LgZx3fNKuKIDCRA6kAChAwkQOpAAoQMJEDqQAKEDCXQN3faVtt+0PWv7qO3dazEYgPpUuWFmQdLDEXHQ9g8kvW37bxFxrPBsAGrS9YoeEaci4mDn37+UNCtpc+nBANSnp9fotrdImpB0oMg0AIqofK+77Y2SXpL0YER8scJ/n5I0VeNsAGpSKXTbo2pHvjci9q/0mIiYljTdeXyZdxgA+E6qfNXdkp6VNBsRT5YfCUDdqrxG3yHpXkm32D7c+efXhecCUKOuT90j4u+S6n9zOYA1w51xQAKEDiRA6EAChA4kQOhAAi6xPdN2lNgCW2rTpySNjJRZiDszM1PkXEm69tpri5y7fv36IufOzc0VObekjRs3Fjv766+/rv3MxcVFRcSy+LiiAwkQOpAAoQMJEDqQAKEDCRA6kAChAwkQOpAAoQMJEDqQAKEDCRA6kAChAwkQOpAAoQMJEDqQAKEDCRA6kAChAwkQOpAAoQMJEDqQQLF1z7UfWtjQUJk/81qtVpFzJenw4cNFzi21RrrECvBvlFoFPjY2VuRcSVpYWKj9zHPnzqnVarHuGciI0IEECB1IgNCBBAgdSIDQgQQIHUigcui2h20fsv1qyYEA1K+XK/puSbOlBgFQTqXQbY9Luk3SnrLjACih6hX9KUmPSCp3PyeAYrqGbvt2SZ9HxNtdHjdle8b2TG3TAahFlSv6Dkl32P5Y0ouSbrH9/NIHRcR0RExGxGTNMwJYpa6hR8RjETEeEVsk3S3pjYjYVXwyALXh79GBBEZ6eXBEvCXprSKTACiGKzqQAKEDCRA6kAChAwkQOpBAsS2wJTZ+ltr0KZXbUDo6OlrkXKm98bOEl19+uci5d911V5FzJWlxcbHIuZs2bSpyriSdPXu29jMXFxcVEWyBBTIidCABQgcSIHQgAUIHEiB0IAFCBxIgdCABQgcSIHQgAUIHEiB0IAFCBxIgdCABQgcSIHQgAUIHEiB0IAFCBxIgdCABQgcSYAtsR6ktsIM489BQmT//33///SLnStJVV11V5NyRkZ6+PWFPFhYWipzLFlggKUIHEiB0IAFCBxIgdCABQgcSIHQggUqh277U9j7b79metX1j6cEA1Kfq3QBPS3otIn5je52ksYIzAahZ19BtXyLpZkm/k6SImJc0X3YsAHWq8tR9m6TTkp6zfcj2HtsbCs8FoEZVQh+RdJ2kZyJiQtJXkh5d+iDbU7ZnbM/UPCOAVaoS+klJJyPiQOfH+9QO/wIRMR0RkxExWeeAAFava+gR8ZmkE7av7nxop6RjRacCUKuqX3W/X9LezlfcP5R0X7mRANStUugRcVgST8mBAcWdcUAChA4kQOhAAoQOJEDoQAKEDiRQbN1z7YcWVmrF8SCue261WkXOLenEiRNFzt22bVuRcyVpeHi49jPn5ubUarVY9wxkROhAAoQOJEDoQAKEDiRA6EAChA4kQOhAAoQOJEDoQAKEDiRA6EAChA4kQOhAAoQOJEDoQAKEDiRA6EAChA4kQOhAAoQOJDBQW2BLbWotqdSmVklaXFwscu7ISNVvstubUvNK5bbtHjtW7juEb9++vfYzW62WIoItsEBGhA4kQOhAAoQOJEDoQAKEDiRA6EAClUK3/ZDto7bftf2C7fWlBwNQn66h294s6QFJkxGxXdKwpLtLDwagPlWfuo9Iutj2iKQxSZ+WGwlA3bqGHhGfSHpC0nFJpySdjYjXSw8GoD5VnrpfJulOSVslXSFpg+1dKzxuyvaM7Zn6xwSwGlWeut8q6aOIOB0R5yTtl3TT0gdFxHRETEbEZN1DAlidKqEfl3SD7TG334q1U9Js2bEA1KnKa/QDkvZJOijpSOfnTBeeC0CNKr3xOCIel/R44VkAFMKdcUAChA4kQOhAAoQOJEDoQAKEDiRQZq9vIa1Wq9jZJdcylzI6Olrk3IWFhSLnllrJLEkXXXRRkXOvv/76IudK0sxM/XeL33PPPSt+nCs6kAChAwkQOpAAoQMJEDqQAKEDCRA6kAChAwkQOpAAoQMJEDqQAKEDCRA6kAChAwkQOpAAoQMJEDqQAKEDCRA6kAChAwkQOpCAS2zmtH1a0r8qPvyHkv5d+xDlDNq8EjOvhX6Z98cR8aOlHywSei9sz0TEZKND9GDQ5pWYeS30+7w8dQcSIHQggX4IfbrpAXo0aPNKzLwW+nrexl+jAyivH67oAAojdCABQgcSIHQgAUIHEvgvpdm/fJGrbWgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.matshow(conf_mx, cmap=plt.cm.gray)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "row_sums = conf_mx.sum(axis=1, keepdims=True)\n",
    "norm_conf_mx = conf_mx / row_sums"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "np.fill_diagonal(norm_conf_mx, 0)\n",
    "plt.matshow(norm_conf_mx, cmap=plt.cm.gray)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "### 코드를 건드린 적이 없는데 DF for문이 col을 출력하여\n",
    "### instances.values로 수정함 \n",
    "\n",
    "import matplotlib\n",
    "def plot_digits(instances, images_per_row=10, **options):\n",
    "    size = 28\n",
    "    images_per_row = min(len(instances), images_per_row)\n",
    "    images = [instance.reshape(size,size) for instance in instances.values]\n",
    "    n_rows = (len(instances) - 1) // images_per_row + 1\n",
    "    row_images = []\n",
    "    n_empty = n_rows * images_per_row - len(instances)\n",
    "    images.append(np.zeros((size, size * n_empty)))\n",
    "    for row in range(n_rows):\n",
    "        rimages = images[row * images_per_row : (row + 1) * images_per_row]\n",
    "        row_images.append(np.concatenate(rimages, axis=1))\n",
    "    image = np.concatenate(row_images, axis=0)\n",
    "    plt.imshow(image, cmap = matplotlib.cm.binary, **options)\n",
    "    plt.axis(\"off\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cl_a, cl_b = 3, 5\n",
    "X_aa = X_train[(y_train == cl_a) & (y_train_pred == cl_a)]\n",
    "X_ab = X_train[(y_train == cl_a) & (y_train_pred == cl_b)]\n",
    "X_ba = X_train[(y_train == cl_b) & (y_train_pred == cl_a)]\n",
    "X_bb = X_train[(y_train == cl_b) & (y_train_pred == cl_b)]\n",
    "\n",
    "plt.figure(figsize=(8,8))\n",
    "plt.subplot(221); plot_digits(X_aa[:25], images_per_row=5)\n",
    "plt.subplot(222); plot_digits(X_ab[:25], images_per_row=5)\n",
    "plt.subplot(223); plot_digits(X_ba[:25], images_per_row=5)\n",
    "plt.subplot(224); plot_digits(X_bb[:25], images_per_row=5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier()"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "y_train_large = (y_train >=7)\n",
    "y_train_odd = (y_train %2 ==1)\n",
    "y_multilabel = np.c_[y_train_large, y_train_odd]\n",
    "\n",
    "knn_clf = KNeighborsClassifier()\n",
    "knn_clf.fit(X_train, y_multilabel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[False,  True]])"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_clf.predict([some_digit])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9735086054032637"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_knn_pred = cross_val_predict(knn_clf, X_train, y_multilabel, cv=2, n_jobs=-1)\n",
    "f1_score(y_multilabel, y_train_knn_pred, average=\"macro\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "noise = np.random.randint(0, 100, (len(X_train), 784))\n",
    "X_train_mod = X_train + noise\n",
    "noise = np.random.randint(0, 100, (len(X_test), 784))\n",
    "X_test_mod = X_test + noise\n",
    "y_train_mod = X_train\n",
    "y_test_mode = X_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "### some_digit을 loc()로 받아오면서 꼬여버렸다...\n",
    "\n",
    "\n",
    "# knn_clf.fit(X_train_mod, y_train_mod)\n",
    "# clean_digit = knn_clf.predict([X_test_mod[some_digit_image.reshape(1,784)]])\n",
    "# plot_digit(clean_digit)"
   ]
  },
  {
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}