{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Logistic Regression을 회기 모델로 사용하기\n",
    "\n",
    "* [Logistic Regression in Python](http://blog.yhathq.com/posts/logistic-regression-and-python.html)\n",
    "* [Regression Analysis with Python, pandas, and StatsModels.pptx](http://allendowney.blogspot.kr/2014/09/regression-with-python-pandas-and.html)\n",
    "* https://stats.oarc.ucla.edu/r/dae/logit-regression/\n",
    "\n",
    "\n",
    "* GRE (Graduate Record Examinations) - 미국 ETS(Educational Testing Service)에서 주관하는 대학원 입학 자격시험\n",
    "  * 구(old) GRE 점수 체계 (2011년 8월 이전)\n",
    "    * Verbal Reasoning (언어): 200점 ~ 800점\n",
    "    * Quantitative Reasoning (수리): 200점 ~ 800점\n",
    "    * 두 영역 모두 만점을 받으면 총 1600점\n",
    "  * 현(new) GRE 점수 체계 (2011년 8월 이후)\n",
    "    * Verbal Reasoning (언어)\t130~170점\n",
    "    * Quantitative Reasoning (수리)\t130~170점\n",
    "    * Analytical Writing (작문) 0.0~6.0점\n",
    "* GPA (Grade Point Average) - '평균 평점' 또는 '내신 성적'. 4.0 만점 스케일을 사용. (A=4.0, B=3.0, C=2.0, D=1.0, F=0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   admit  gre   gpa  rank\n",
      "0      0  380  3.61     3\n",
      "1      1  660  3.67     3\n",
      "2      1  800  4.00     1\n",
      "3      1  640  3.19     4\n",
      "4      0  520  2.93     4\n",
      "Index(['admit', 'gre', 'gpa', 'prestige'], dtype='str')\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# read the data in\n",
    "df = pd.read_csv(\"data/binary.csv\")\n",
    "\n",
    "# take a look at the dataset\n",
    "print(df.head())\n",
    "#    admit  gre   gpa  rank\n",
    "# 0      0  380  3.61     3\n",
    "# 1      1  660  3.67     3\n",
    "# 2      1  800  4.00     1\n",
    "# 3      1  640  3.19     4\n",
    "# 4      0  520  2.93     4\n",
    "\n",
    "# rename the 'rank' column because there is also a DataFrame method called 'rank'\n",
    "# prestige: 대학교 평판 수준\n",
    "df.columns = [\"admit\", \"gre\", \"gpa\", \"prestige\"]\n",
    "print(df.columns)\n",
    "# array([admit, gre, gpa, prestige], dtype=object)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            admit         gre         gpa   prestige\n",
      "count  400.000000  400.000000  400.000000  400.00000\n",
      "mean     0.317500  587.700000    3.389900    2.48500\n",
      "std      0.466087  115.516536    0.380567    0.94446\n",
      "min      0.000000  220.000000    2.260000    1.00000\n",
      "25%      0.000000  520.000000    3.130000    2.00000\n",
      "50%      0.000000  580.000000    3.395000    2.00000\n",
      "75%      1.000000  660.000000    3.670000    3.00000\n",
      "max      1.000000  800.000000    4.000000    4.00000\n",
      "admit         0.466087\n",
      "gre         115.516536\n",
      "gpa           0.380567\n",
      "prestige      0.944460\n",
      "dtype: float64\n",
      "prestige   1   2   3   4\n",
      "admit                   \n",
      "0         28  97  93  55\n",
      "1         33  54  28  12\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# summarize the data\n",
    "print(df.describe())\n",
    "#             admit         gre         gpa   prestige\n",
    "# count  400.000000  400.000000  400.000000  400.00000\n",
    "# mean     0.317500  587.700000    3.389900    2.48500\n",
    "# std      0.466087  115.516536    0.380567    0.94446\n",
    "# min      0.000000  220.000000    2.260000    1.00000\n",
    "# 25%      0.000000  520.000000    3.130000    2.00000\n",
    "# 50%      0.000000  580.000000    3.395000    2.00000\n",
    "# 75%      1.000000  660.000000    3.670000    3.00000\n",
    "# max      1.000000  800.000000    4.000000    4.00000\n",
    " \n",
    "# take a look at the standard deviation of each column\n",
    "print(df.std())\n",
    "# admit      0.466087\n",
    "# gre      115.516536\n",
    "# gpa        0.380567\n",
    "# prestige   0.944460\n",
    " \n",
    "# frequency table cutting presitge and whether or not someone was admitted\n",
    "print(pd.crosstab(df['admit'], df['prestige'], rownames=['admit']))\n",
    "# prestige   1   2   3   4\n",
    "# admit                   \n",
    "# 0         28  97  93  55\n",
    "# 1         33  54  28  12\n",
    " \n",
    "# plot all of the columns\n",
    "df.hist();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "get_dummies() 메서드는 주어진 열 값에 따라 이진화 열과 값을 생성해낸다. 아래 예는 prestige 열의 값을 dummify한 것이다.\n",
    "이 table을 기존의 table과 합친다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   prestige_1  prestige_2  prestige_3  prestige_4\n",
      "0           0           0           1           0\n",
      "1           0           0           1           0\n",
      "2           1           0           0           0\n",
      "3           0           0           0           1\n",
      "4           0           0           0           1\n",
      "   admit  gre   gpa  prestige_2  prestige_3  prestige_4\n",
      "0      0  380  3.61           0           1           0\n",
      "1      1  660  3.67           0           1           0\n",
      "2      1  800  4.00           0           0           0\n",
      "3      1  640  3.19           0           0           1\n",
      "4      0  520  2.93           0           0           1\n"
     ]
    }
   ],
   "source": [
    "# dummify rank\n",
    "dummy_ranks = pd.get_dummies(df['prestige'], prefix='prestige', dtype=int)\n",
    "print(dummy_ranks.head())\n",
    "#    prestige_1  prestige_2  prestige_3  prestige_4\n",
    "# 0           0           0           1           0\n",
    "# 1           0           0           1           0\n",
    "# 2           1           0           0           0\n",
    "# 3           0           0           0           1\n",
    "# 4           0           0           0           1\n",
    " \n",
    "# create a clean data frame for the regression\n",
    "cols_to_keep = ['admit', 'gre', 'gpa']\n",
    "data = df[cols_to_keep].join(dummy_ranks.loc[:, 'prestige_2':])\n",
    "print(data.head())\n",
    "#    admit  gre   gpa  prestige_2  prestige_3  prestige_4\n",
    "# 0      0  380  3.61           0           1           0\n",
    "# 1      1  660  3.67           0           1           0\n",
    "# 2      1  800  4.00           0           0           0\n",
    "# 3      1  640  3.19           0           0           1\n",
    "# 4      0  520  2.93           0           0           1\n",
    " \n",
    "# manually add the intercept\n",
    "data['intercept'] = 1.0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 로짓 변환\n",
    "\n",
    "gre, gpa, prestige_2, prestige_3, prestige_4를 이용하여 admit을 예측해보자.\n",
    "statsmodels.api 모듈의 Logit 클래스를 이용한다."
   ]
  },
  {
   "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>admit</th>\n",
       "      <th>gre</th>\n",
       "      <th>gpa</th>\n",
       "      <th>prestige_2</th>\n",
       "      <th>prestige_3</th>\n",
       "      <th>prestige_4</th>\n",
       "      <th>intercept</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>380</td>\n",
       "      <td>3.61</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>660</td>\n",
       "      <td>3.67</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>800</td>\n",
       "      <td>4.00</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>640</td>\n",
       "      <td>3.19</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>520</td>\n",
       "      <td>2.93</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   admit  gre   gpa  prestige_2  prestige_3  prestige_4  intercept\n",
       "0      0  380  3.61           0           1           0        1.0\n",
       "1      1  660  3.67           0           1           0        1.0\n",
       "2      1  800  4.00           0           0           0        1.0\n",
       "3      1  640  3.19           0           0           1        1.0\n",
       "4      0  520  2.93           0           0           1        1.0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "회귀계수: [[ 0.00226181  0.80589736 -0.6765609  -1.34035637 -1.55277503 -1.99705521]]\n",
      "절편: [-1.99705521]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "train_cols = data.columns[1:]\n",
    "X = data[train_cols]\n",
    "y = data['admit']\n",
    "\n",
    "# 1. C=float('inf')로 규제화를 해제하고, max_iter를 늘려 수렴 오류를 방지합니다.\n",
    "model = LogisticRegression(C=float('inf'), max_iter=1000)\n",
    "result = model.fit(X, y)\n",
    "\n",
    "print(\"회귀계수:\", result.coef_)\n",
    "print(\"절편:\", result.intercept_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gre           0.002262\n",
      "gpa           0.805897\n",
      "prestige_2   -0.676561\n",
      "prestige_3   -1.340356\n",
      "prestige_4   -1.552775\n",
      "intercept    -1.997055\n",
      "intercept    -1.997055\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# 변수 이름과 회귀계수를 각각 리스트로 모읍니다.\n",
    "feature_names = list(train_cols) + ['intercept']\n",
    "coefficients = list(model.coef_[0]) + [model.intercept_[0]]\n",
    "\n",
    "# 판다스 시리즈로 변환하여 출력합니다.\n",
    "result_series = pd.Series(coefficients, index=feature_names)\n",
    "print(result_series)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "계수에 지수함수를 취하면 odds ratio를 얻는다. 이 값은 변수의 1단위가 증감할 때 합격 odds에 얼마나 영향을 미치는지 말해준다. 예를 들어, prestige_2가 감소할 때 50%의 합격 오즈가 감소한다는 것을 알 수 있다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gre           1.002264\n",
      "gpa           2.238705\n",
      "prestige_2    0.508362\n",
      "prestige_3    0.261752\n",
      "prestige_4    0.211660\n",
      "intercept     0.135734\n",
      "intercept     0.135734\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# odds ratio 출력해보기\n",
    "# 회귀계수에 지수 함수(np.exp)를 적용하여 오즈비를 계산합니다.\n",
    "# 판다스 시리즈 형태로 보기 좋게 출력합니다.\n",
    "odds_series = pd.Series(np.exp(coefficients), index=feature_names)\n",
    "print(odds_series)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "신뢰 구간으로 표시하면 불확실한 변수에 대해 좀 더 확실히 알 수 있다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            Odds Ratio  Lower CI (2.5%)  Upper CI (97.5%)\n",
      "intercept     0.135734         0.044410          0.414859\n",
      "gre           1.002264         1.000118          1.004416\n",
      "gpa           2.238705         1.168202          4.290181\n",
      "prestige_2    0.508362         0.273374          0.945342\n",
      "prestige_3    0.261752         0.133036          0.515007\n",
      "prestige_4    0.211660         0.093309          0.480123\n",
      "intercept     0.135734         0.044410          0.414859\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from scipy import stats\n",
    "\n",
    "# 1. 예측 확률 계산 및 디자인 행렬(상수항 포함) 생성\n",
    "pred_probs = model.predict_proba(X)[:, 1]\n",
    "# X가 DataFrame인 경우 내부 연산을 위해 .values를 활용할 수 있습니다.\n",
    "X_values = X.values if isinstance(X, pd.DataFrame) else X\n",
    "X_design = np.hstack([np.ones((X_values.shape[0], 1)), X_values])\n",
    "\n",
    "# 2. 공분산 행렬을 통한 표준오차(SE, Standard Error) 계산\n",
    "W = np.diag(pred_probs * (1 - pred_probs))\n",
    "# inv 대신 pinv(의사역행렬)를 사용하여 Singular Matrix 에러를 방지합니다.\n",
    "cov_matrix = np.linalg.pinv(X_design.T @ W @ X_design)\n",
    "standard_errors = np.sqrt(np.diag(cov_matrix))\n",
    "\n",
    "# 3. 회귀계수와 변수 이름 정리 (절편을 가장 앞으로 배치)\n",
    "coefficients = np.hstack([model.intercept_, model.coef_[0]])\n",
    "feature_names = ['intercept'] + list(train_cols)\n",
    "\n",
    "# 4. 회귀계수의 95% 신뢰 구간 계산\n",
    "z_critical = stats.norm.ppf(0.975)  # 약 1.96\n",
    "coef_lower = coefficients - z_critical * standard_errors\n",
    "coef_upper = coefficients + z_critical * standard_errors\n",
    "\n",
    "# 5. 오즈비(Odds Ratio) 및 오즈비의 신뢰 구간 계산\n",
    "odds_ratios = np.exp(coefficients)\n",
    "or_lower = np.exp(coef_lower)\n",
    "or_upper = np.exp(coef_upper)\n",
    "\n",
    "# 6. 데이터프레임 형태로 모아서 출력\n",
    "result_df = pd.DataFrame({\n",
    "    'Odds Ratio': odds_ratios,\n",
    "    'Lower CI (2.5%)': or_lower,\n",
    "    'Upper CI (97.5%)': or_upper\n",
    "}, index=feature_names)\n",
    "\n",
    "print(result_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   admit  gre   gpa  prestige_2  prestige_3  prestige_4  intercept\n",
      "0      0  380  3.61           0           1           0        1.0\n",
      "1      1  660  3.67           0           1           0        1.0\n",
      "2      1  800  4.00           0           0           0        1.0\n",
      "3      1  640  3.19           0           0           1        1.0\n",
      "4      0  520  2.93           0           0           1        1.0\n"
     ]
    }
   ],
   "source": [
    "print(data.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "GRE, GPA 점수에 따른 합격률을 계산해보자"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import itertools\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "# =========================================================================\n",
    "# 사전 작업: 이미 존재하는 data를 기반으로 가상 데이터셋(combos) 생성\n",
    "# =========================================================================\n",
    "\n",
    "# 1. 원본 데이터의 범위를 기준으로 gre와 gpa의 촘촘한 격자 데이터를 정의합니다.\n",
    "gre_range = np.linspace(data['gre'].min(), data['gre'].max(), 20)\n",
    "gpa_range = np.linspace(data['gpa'].min(), data['gpa'].max(), 20)\n",
    "\n",
    "# 2. prestige의 4가지 상태를 원-핫 인코딩 형태로 직접 정의합니다.\n",
    "# 순서대로 prestige 1(모두 0), 2, 3, 4를 의미합니다.\n",
    "prestige_dummies = [\n",
    "    (0, 0, 0),  # prestige 1\n",
    "    (1, 0, 0),  # prestige 2\n",
    "    (0, 1, 0),  # prestige 3\n",
    "    (0, 0, 1)   # prestige 4\n",
    "]\n",
    "\n",
    "# 3. 모든 변수의 가능한 조합을 만들어 데이터프레임으로 변환합니다.\n",
    "combinations = list(itertools.product(gre_range, gpa_range, prestige_dummies))\n",
    "\n",
    "# 4. 조합된 데이터를 학습용(X_combos)과 시각화용(combos)으로 각각 정렬합니다.\n",
    "rows_X = []\n",
    "rows_combos = []\n",
    "\n",
    "for gre, gpa, (p2, p3, p4) in combinations:\n",
    "    # 모델 입력용 데이터 구조 (train_cols 순서와 일치: gre, gpa, prestige_2, prestige_3, prestige_4, intercept)\n",
    "    rows_X.append([gre, gpa, p2, p3, p4, 1.0])\n",
    "    \n",
    "    # 시각화 함수(pivot_table) 분석용 데이터 구조 (원본 prestige 번호 복원)\n",
    "    p_num = 1 if (p2, p3, p4) == (0, 0, 0) else (2 if p2 == 1 else (3 if p3 == 1 else 4))\n",
    "    rows_combos.append([gre, gpa, p_num])\n",
    "\n",
    "# 데이터프레임 생성\n",
    "X_combos = pd.DataFrame(rows_X, columns=list(train_cols))\n",
    "combos = pd.DataFrame(rows_combos, columns=['gre', 'gpa', 'prestige'])\n",
    "\n",
    "# 5. 기존 학습된 사이킷런 model로 합격 확률을 예측하여 combos에 추가합니다.\n",
    "combos['admit_pred'] = model.predict_proba(X_combos)[:, 1]\n",
    "\n",
    "\n",
    "# =========================================================================\n",
    "# 시각화 함수 정의 및 실행\n",
    "# =========================================================================\n",
    "def isolate_and_plot(variable):\n",
    "    # 피벗 테이블을 생성하여 지정된 변수와 prestige별 합격 확률 평균을 구합니다.\n",
    "    grouped = pd.pivot_table(\n",
    "        combos, \n",
    "        values=['admit_pred'], \n",
    "        index=[variable, 'prestige'],\n",
    "        aggfunc='mean'\n",
    "    )\n",
    "    \n",
    "    # 멀티인덱스로 되어 있는 prestige를 컬럼으로 올려 시각화하기 좋은 구조로 만듭니다.\n",
    "    plot_data = grouped['admit_pred'].unstack(level='prestige')\n",
    "    \n",
    "    # 그래프를 그립니다.\n",
    "    plt.figure(figsize=(8, 5))\n",
    "    \n",
    "    # 각 prestige 등급별 색상을 매핑합니다.\n",
    "    color_map = {1: 'r', 2: 'b', 3: 'g', 4: 'y'}\n",
    "    \n",
    "    for col in plot_data.columns:\n",
    "        plt.plot(\n",
    "            plot_data.index, \n",
    "            plot_data[col], \n",
    "            color=color_map.get(col, 'k'), \n",
    "            label=f\"Prestige {col}\"\n",
    "        )\n",
    " \n",
    "    plt.xlabel(variable)\n",
    "    plt.ylabel(\"P(admit=1)\")\n",
    "    plt.legend(loc='upper left')\n",
    "    plt.title(f\"Prob(admit=1) isolating {variable} and prestige\")\n",
    "    plt.grid(True, linestyle='--', alpha=0.5)\n",
    "    plt.show()\n",
    "\n",
    "# 함수 실행\n",
    "isolate_and_plot('gre')\n",
    "isolate_and_plot('gpa')"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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": "3.11.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}
