diff --git a/lab_2/lab2.ipynb b/lab_2/lab2.ipynb index f963d47..ebf885d 100644 --- a/lab_2/lab2.ipynb +++ b/lab_2/lab2.ipynb @@ -13,7 +13,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 1, "metadata": {}, "outputs": [], "source": [ @@ -49,7 +49,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 2, "metadata": {}, "outputs": [ { @@ -97,7 +97,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 3, "metadata": {}, "outputs": [ { @@ -127,7 +127,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 4, "metadata": {}, "outputs": [ { @@ -229,7 +229,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -238,7 +238,7 @@ "" ] }, - "execution_count": 7, + "execution_count": 5, "metadata": {}, "output_type": "execute_result" }, @@ -266,7 +266,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 6, "metadata": {}, "outputs": [ { @@ -275,7 +275,7 @@ "" ] }, - "execution_count": 8, + "execution_count": 6, "metadata": {}, "output_type": "execute_result" }, @@ -303,7 +303,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 7, "metadata": {}, "outputs": [ { @@ -312,7 +312,7 @@ "" ] }, - "execution_count": 9, + "execution_count": 7, "metadata": {}, "output_type": "execute_result" }, @@ -340,7 +340,7 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 8, "metadata": {}, "outputs": [ { @@ -371,7 +371,7 @@ }, { "cell_type": "code", - "execution_count": 52, + "execution_count": 9, "metadata": {}, "outputs": [ { @@ -400,7 +400,7 @@ }, { "cell_type": "code", - "execution_count": 82, + "execution_count": 10, "metadata": {}, "outputs": [ { @@ -409,7 +409,7 @@ "" ] }, - "execution_count": 82, + "execution_count": 10, "metadata": {}, "output_type": "execute_result" }, @@ -430,7 +430,7 @@ }, { "cell_type": "code", - "execution_count": 97, + "execution_count": 11, "metadata": {}, "outputs": [ { @@ -439,7 +439,7 @@ "" ] }, - "execution_count": 97, + "execution_count": 11, "metadata": {}, "output_type": "execute_result" }, @@ -467,7 +467,7 @@ }, { "cell_type": "code", - "execution_count": 146, + "execution_count": 12, "metadata": {}, "outputs": [ { @@ -497,7 +497,7 @@ }, { "cell_type": "code", - "execution_count": 153, + "execution_count": 13, "metadata": {}, "outputs": [ { @@ -506,7 +506,7 @@ "" ] }, - "execution_count": 153, + "execution_count": 13, "metadata": {}, "output_type": "execute_result" }, @@ -553,7 +553,7 @@ }, { "cell_type": "code", - "execution_count": 154, + "execution_count": 14, "metadata": {}, "outputs": [ { @@ -562,7 +562,12 @@ "text": [ "Series([], dtype: int64)\n", "--------------\n", - "Series([], dtype: int64)\n", + "Android_version 404\n", + "Inbuilt_memory 18\n", + "fast_charging 82\n", + "Screen_resolution 2\n", + "Processor 25\n", + "dtype: int64\n", "--------------\n", "Series([], dtype: int64)\n" ] @@ -585,7 +590,7 @@ }, { "cell_type": "code", - "execution_count": 155, + "execution_count": 20, "metadata": {}, "outputs": [ { @@ -594,19 +599,6 @@ "text": [ "Series([], dtype: int64)\n" ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "C:\\Users\\pasha\\AppData\\Local\\Temp\\ipykernel_6832\\3049087464.py:4: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.\n", - "The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.\n", - "\n", - "For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.\n", - "\n", - "\n", - " phones_df[column].fillna(mode, inplace=True)\n" - ] } ], "source": [ @@ -630,7 +622,7 @@ }, { "cell_type": "code", - "execution_count": 156, + "execution_count": 21, "metadata": {}, "outputs": [ { @@ -679,7 +671,7 @@ }, { "cell_type": "code", - "execution_count": 157, + "execution_count": 22, "metadata": {}, "outputs": [ { @@ -725,7 +717,7 @@ }, { "cell_type": "code", - "execution_count": 158, + "execution_count": 23, "metadata": {}, "outputs": [ { @@ -769,7 +761,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 24, "metadata": {}, "outputs": [ { @@ -799,6 +791,16 @@ }, "metadata": {}, "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" } ], "source": [ diff --git a/lab_3/lab3.ipynb b/lab_3/lab3.ipynb new file mode 100644 index 0000000..7b71d1c --- /dev/null +++ b/lab_3/lab3.ipynb @@ -0,0 +1,1200 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "1. Определить бизнес-цели\n", + "Бизнес-цели:\n", + " а. Прогнозирование цены страховки\n", + " б. Оценка влияния данных страхователя на цену страховки" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "2. Определить цели технического проета для каждой бизнес-цели\n", + " а. Построить можедь, которая на основе данных страхователя будет предсказывать цену страховки\n", + " б. Провести анализ для выявления факторов, которые наиболее сильно влияют на итоговую цену страховки" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "3. Подготовка данных" + ] + }, + { + "cell_type": "code", + "execution_count": 592, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "2772\n" + ] + } + ], + "source": [ + "import pandas as pd\n", + "df = pd.read_csv(\"../dataset.csv\")\n", + "print(df.shape[0])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "данных достаточно чтобы шумы усреднились" + ] + }, + { + "cell_type": "code", + "execution_count": 593, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "было 2772\n", + "age 39.10966810966811 14.081459420836477\n", + "bmi 30.70134920634921 6.1294486949652205\n", + "children 1.1026753434562546 1.2157555494600176\n", + "charges 13325.498588795157 12200.175109274192\n", + "стало 2710\n" + ] + } + ], + "source": [ + "print(\"было \", df.shape[0])\n", + "for column in df.select_dtypes(include=['int', 'float']).columns:\n", + " mean = df[column].mean()\n", + " std_dev = df[column].std()\n", + " print(column, mean, std_dev)\n", + " \n", + " lower_bound = mean - 3 * std_dev\n", + " upper_bound = mean + 3 * std_dev\n", + " \n", + " df = df[(df[column] <= upper_bound) & (df[column] >= lower_bound)]\n", + " \n", + "print(\"стало \", df.shape[0])\n", + "df = df.reset_index(drop=True)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "были устранены выбросы, отобранные по правилу трех сигм" + ] + }, + { + "cell_type": "code", + "execution_count": 594, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "age 0\n", + "sex 0\n", + "bmi 0\n", + "children 0\n", + "smoker 0\n", + "region 0\n", + "charges 0\n", + "dtype: int64\n" + ] + } + ], + "source": [ + "print(df.isnull().sum())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Пропущенных значений нет" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "4. Разбиение на выборки" + ] + }, + { + "cell_type": "code", + "execution_count": 595, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "1897 406 407\n", + "2710 2710\n" + ] + } + ], + "source": [ + "from sklearn.model_selection import train_test_split\n", + "\n", + "train_df, temp_df = train_test_split(df, test_size=0.3, random_state=52)\n", + "val_df, test_df = train_test_split(temp_df, test_size=0.5, random_state=52)\n", + "\n", + "print(train_df.shape[0], val_df.shape[0], test_df.shape[0])\n", + "print(df.shape[0], train_df.shape[0] + val_df.shape[0] + test_df.shape[0])\n", + "\n", + "test_df = test_df.reset_index(drop=True)\n", + "val_df = val_df.reset_index(drop=True)\n", + "train_df = train_df.reset_index(drop=True)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "данные были разделены на обучающую (70%), контрольную (15%) и тестовую (15%) выборки" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "5. Оценка сбалансированности выборок" + ] + }, + { + "cell_type": "code", + "execution_count": 596, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "\n", + "def draw(data, title):\n", + " sns.histplot(data['charges'], kde=True)\n", + " plt.title(title)\n", + " plt.show()\n", + " \n", + "draw(train_df, 'Распределение цен в обучающей выборке')\n", + "draw(val_df, 'Распределение цен в контрольной выборке')\n", + "draw(test_df, 'Распределение цен в тестовой выборке')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "6,7. Конструирование признаков" + ] + }, + { + "cell_type": "code", + "execution_count": 597, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "['southwest' 'southeast' 'northwest' 'northeast']\n", + "[0 1 3 2 4]\n", + "Index(['age', 'sex', 'bmi', 'children', 'smoker', 'region', 'charges',\n", + " 'smoker_yes', 'sex_male', 'region_northwest', 'region_southeast',\n", + " 'region_southwest', 'children_1', 'children_2', 'children_3',\n", + " 'children_4'],\n", + " dtype='object')\n", + "Index(['age', 'sex', 'bmi', 'children', 'smoker', 'region', 'charges',\n", + " 'smoker_yes', 'sex_male', 'region_northwest', 'region_southeast',\n", + " 'region_southwest', 'children_1', 'children_2', 'children_3',\n", + " 'children_4'],\n", + " dtype='object')\n", + "Index(['age', 'sex', 'bmi', 'children', 'smoker', 'region', 'charges',\n", + " 'smoker_yes', 'sex_male', 'region_northwest', 'region_southeast',\n", + " 'region_southwest', 'children_1', 'children_2', 'children_3',\n", + " 'children_4'],\n", + " dtype='object')\n" + ] + } + ], + "source": [ + "print(df['region'].unique())\n", + "print(df['children'].unique())\n", + "\n", + "from sklearn.preprocessing import OneHotEncoder\n", + "import numpy as np\n", + "\n", + "encoder = OneHotEncoder(sparse_output=False, drop=\"first\")\n", + "\n", + "encoded_values = encoder.fit_transform(train_df[[\"smoker\", \"sex\", \"region\", \"children\"]])\n", + "encoded_columns = encoder.get_feature_names_out([\"smoker\", \"sex\", \"region\", \"children\"])\n", + "encoded_values_df = pd.DataFrame(encoded_values, columns=encoded_columns)\n", + "train_df = pd.concat([train_df, encoded_values_df], axis=1)\n", + "\n", + "encoded_values = encoder.fit_transform(test_df[[\"smoker\", \"sex\", \"region\", \"children\"]])\n", + "encoded_columns = encoder.get_feature_names_out([\"smoker\", \"sex\", \"region\", \"children\"])\n", + "encoded_values_df = pd.DataFrame(encoded_values, columns=encoded_columns)\n", + "test_df = pd.concat([test_df, encoded_values_df], axis=1)\n", + "\n", + "encoded_values = encoder.fit_transform(val_df[[\"smoker\", \"sex\", \"region\", \"children\"]])\n", + "encoded_columns = encoder.get_feature_names_out([\"smoker\", \"sex\", \"region\", \"children\"])\n", + "encoded_values_df = pd.DataFrame(encoded_values, columns=encoded_columns)\n", + "val_df = pd.concat([val_df, encoded_values_df], axis=1)\n", + "\n", + "print(test_df.columns)\n", + "print(val_df.columns)\n", + "print(train_df.columns)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Было совершено унитарное кодирование признаков Пол (sex), Курильщик (smoker) и Регион (region). Полученные признаки были добавлены в исходный сет." + ] + }, + { + "cell_type": "code", + "execution_count": 598, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "age 18 64\n", + "bmi 15.96 49.06\n", + "bmi_category\n", + "overweight 335\n", + "normal weight 66\n", + "underweight 5\n", + "Name: count, dtype: int64\n", + "========================\n", + "bmi_category\n", + "overweight 332\n", + "normal weight 70\n", + "underweight 5\n", + "Name: count, dtype: int64\n", + "========================\n", + "bmi_category\n", + "overweight 1543\n", + "normal weight 324\n", + "underweight 30\n", + "Name: count, dtype: int64\n" + ] + } + ], + "source": [ + "print('age', min(df['age']), max(df['age']))\n", + "#print('charges', min(df['charges']), max(df['charges']))\n", + "print('bmi', min(df['bmi']), max(df['bmi']))\n", + "\n", + "labels_age = ['young', 'middle-aged', 'old']\n", + "labels_bmi = ['underweight', 'normal weight', 'overweight']\n", + "#labels_charges = ['low_charges', 'medium_charges', 'high_charges']\n", + "\n", + "hist_age, bins_age = np.histogram(test_df['age'], bins = [0, 27, 45, 100])\n", + "age_df = pd.concat([test_df['age'], pd.cut(test_df['age'], list(bins_age), labels = labels_age)], axis=1)\n", + "test_df['age_category'] = pd.cut(test_df['age'], bins=bins_age, labels=labels_age)\n", + "\n", + "hist_bmi, bins_bmi = np.histogram(test_df['bmi'], bins = [0, 18.5, 25, 100])\n", + "bmi_df = pd.concat([test_df['bmi'], pd.cut(test_df['bmi'], list(bins_bmi), labels = labels_bmi)], axis=1)\n", + "test_df['bmi_category'] = pd.cut(test_df['bmi'], bins=bins_bmi, labels=labels_bmi)\n", + "\n", + "hist_age, bins_age = np.histogram(train_df['age'], bins = [0, 27, 45, 100])\n", + "age_df = pd.concat([train_df['age'], pd.cut(train_df['age'], list(bins_age), labels = labels_age)], axis=1)\n", + "train_df['age_category'] = pd.cut(train_df['age'], bins=bins_age, labels=labels_age)\n", + "\n", + "hist_bmi, bins_bmi = np.histogram(train_df['bmi'], bins = [0, 18.5, 25, 100])\n", + "bmi_df = pd.concat([train_df['bmi'], pd.cut(train_df['bmi'], list(bins_bmi), labels = labels_bmi)], axis=1)\n", + "train_df['bmi_category'] = pd.cut(train_df['bmi'], bins=bins_bmi, labels=labels_bmi)\n", + "\n", + "hist_age, bins_age = np.histogram(val_df['age'], bins = [0, 27, 45, 100])\n", + "age_df = pd.concat([val_df['age'], pd.cut(val_df['age'], list(bins_age), labels = labels_age)], axis=1)\n", + "val_df['age_category'] = pd.cut(val_df['age'], bins=bins_age, labels=labels_age)\n", + "\n", + "hist_bmi, bins_bmi = np.histogram(val_df['bmi'], bins = [0, 18.5, 25, 100])\n", + "bmi_df = pd.concat([val_df['bmi'], pd.cut(val_df['bmi'], list(bins_bmi), labels = labels_bmi)], axis=1)\n", + "val_df['bmi_category'] = pd.cut(val_df['bmi'], bins=bins_bmi, labels=labels_bmi)\n", + "\n", + "category_counts = val_df['bmi_category'].value_counts()\n", + "print(category_counts)\n", + "print('========================')\n", + "category_counts = test_df['bmi_category'].value_counts()\n", + "print(category_counts)\n", + "print('========================')\n", + "category_counts = train_df['bmi_category'].value_counts()\n", + "print(category_counts)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Была выполнена дискретизация числовых признаков Индекс массы тела (bmi) и Возраст (age)" + ] + }, + { + "cell_type": "code", + "execution_count": 599, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "0 1.0\n", + "1 1.0\n", + "2 0.0\n", + "3 0.0\n", + "4 1.0\n", + "5 0.0\n", + "6 0.0\n", + "7 0.0\n", + "8 1.0\n", + "9 1.0\n", + "10 0.0\n", + "11 0.0\n", + "12 1.0\n", + "13 0.0\n", + "14 0.0\n", + "15 0.0\n", + "16 1.0\n", + "17 1.0\n", + "18 1.0\n", + "19 0.0\n", + "Name: parent_yes, dtype: float64\n", + "========================\n", + "0 1.0\n", + "1 1.0\n", + "2 0.0\n", + "3 1.0\n", + "4 0.0\n", + "5 1.0\n", + "6 0.0\n", + "7 0.0\n", + "8 1.0\n", + "9 0.0\n", + "10 1.0\n", + "11 1.0\n", + "12 0.0\n", + "13 0.0\n", + "14 0.0\n", + "15 1.0\n", + "16 1.0\n", + "17 0.0\n", + "18 1.0\n", + "19 1.0\n", + "Name: parent_yes, dtype: float64\n", + "========================\n", + "0 1.0\n", + "1 0.0\n", + "2 1.0\n", + "3 1.0\n", + "4 1.0\n", + "5 0.0\n", + "6 0.0\n", + "7 1.0\n", + "8 0.0\n", + "9 1.0\n", + "10 0.0\n", + "11 1.0\n", + "12 1.0\n", + "13 0.0\n", + "14 1.0\n", + "15 1.0\n", + "16 1.0\n", + "17 0.0\n", + "18 0.0\n", + "19 1.0\n", + "Name: parent_yes, dtype: float64\n" + ] + } + ], + "source": [ + "train_df['parent_yes'] = train_df['children'] > 0\n", + "train_df['parent_yes'] = train_df['parent_yes'].map({True: 1.0, False: 0.0})\n", + "\n", + "test_df['parent_yes'] = test_df['children'] > 0\n", + "test_df['parent_yes'] = test_df['parent_yes'].map({True: 1.0, False: 0.0})\n", + "\n", + "val_df['parent_yes'] = val_df['children'] > 0\n", + "val_df['parent_yes'] = val_df['parent_yes'].map({True: 1.0, False: 0.0})\n", + "\n", + "print(train_df['parent_yes'].head(20))\n", + "print('========================')\n", + "print(test_df['parent_yes'].head(20))\n", + "print('========================')\n", + "print(val_df['parent_yes'].head(20))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Был выполнен ручной синтез признака Родитель, на основе того, есть ли дети у страхователя или нет" + ] + }, + { + "cell_type": "code", + "execution_count": 600, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "0 0.695652\n", + "1 0.500000\n", + "2 0.239130\n", + "3 0.565217\n", + "4 0.978261\n", + "5 0.304348\n", + "6 0.847826\n", + "7 0.630435\n", + "8 0.804348\n", + "9 0.173913\n", + "Name: age_norm, dtype: float64\n", + "========================\n", + "0 0.586957\n", + "1 0.413043\n", + "2 0.847826\n", + "3 0.956522\n", + "4 0.521739\n", + "5 0.869565\n", + "6 0.021739\n", + "7 0.130435\n", + "8 0.565217\n", + "9 0.500000\n", + "Name: age_norm, dtype: float64\n", + "========================\n", + "0 0.217391\n", + "1 0.173913\n", + "2 0.760870\n", + "3 0.108696\n", + "4 0.326087\n", + "5 0.065217\n", + "6 0.282609\n", + "7 0.260870\n", + "8 0.152174\n", + "9 0.565217\n", + "Name: age_norm, dtype: float64\n" + ] + } + ], + "source": [ + "from sklearn import preprocessing\n", + "\n", + "scaler = preprocessing.MinMaxScaler()\n", + "test_df['age_norm'] = scaler.fit_transform(test_df[['age']])\n", + "print(test_df['age_norm'].head(10))\n", + "print('========================')\n", + "train_df['age_norm'] = scaler.fit_transform(train_df[['age']])\n", + "print(train_df['age_norm'].head(10))\n", + "print('========================')\n", + "val_df['age_norm'] = scaler.fit_transform(val_df[['age']])\n", + "print(val_df['age_norm'].head(10))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Было выполнено маштабирование признака Возраст (age) на основе нормировки." + ] + }, + { + "cell_type": "code", + "execution_count": 601, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "0 0.414950\n", + "1 -0.153961\n", + "2 1.268317\n", + "3 1.623887\n", + "4 0.201608\n", + "5 1.339431\n", + "6 -1.434012\n", + "7 -1.078443\n", + "8 0.343836\n", + "9 0.130494\n", + "Name: age_stand, dtype: float64\n", + "========================\n", + "0 0.727622\n", + "1 0.101529\n", + "2 -0.733262\n", + "3 0.310226\n", + "4 1.631978\n", + "5 -0.524564\n", + "6 1.214583\n", + "7 0.518924\n", + "8 1.075451\n", + "9 -0.941960\n", + "Name: age_stand, dtype: float64\n", + "========================\n", + "0 -0.766548\n", + "1 -0.907530\n", + "2 0.995731\n", + "3 -1.119003\n", + "4 -0.414092\n", + "5 -1.259986\n", + "6 -0.555074\n", + "7 -0.625565\n", + "8 -0.978021\n", + "9 0.361310\n", + "Name: age_stand, dtype: float64\n" + ] + } + ], + "source": [ + "scaler = preprocessing.StandardScaler()\n", + "train_df['age_stand'] = scaler.fit_transform(train_df[['age']])\n", + "print(train_df['age_stand'].head(10))\n", + "print('========================')\n", + "test_df['age_stand'] = scaler.fit_transform(test_df[['age']])\n", + "print(test_df['age_stand'].head(10))\n", + "print('========================')\n", + "val_df['age_stand'] = scaler.fit_transform(val_df[['age']])\n", + "print(val_df['age_stand'].head(10))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Было выполнено маштабирование признака Возраст (age) на основе стандартизации." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "8. Использование Featuretools" + ] + }, + { + "cell_type": "code", + "execution_count": 602, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "c:\\ulstu\\mii\\AIM-PIbd-31-Barsukov-P-O\\aimenv\\Lib\\site-packages\\featuretools\\entityset\\entityset.py:1733: UserWarning: index index not found in dataframe, creating new integer column\n", + " warnings.warn(\n", + "c:\\ulstu\\mii\\AIM-PIbd-31-Barsukov-P-O\\aimenv\\Lib\\site-packages\\woodwork\\type_sys\\utils.py:33: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n", + " pd.to_datetime(\n", + "c:\\ulstu\\mii\\AIM-PIbd-31-Barsukov-P-O\\aimenv\\Lib\\site-packages\\woodwork\\type_sys\\utils.py:33: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n", + " pd.to_datetime(\n", + "c:\\ulstu\\mii\\AIM-PIbd-31-Barsukov-P-O\\aimenv\\Lib\\site-packages\\woodwork\\type_sys\\utils.py:33: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n", + " pd.to_datetime(\n", + "c:\\ulstu\\mii\\AIM-PIbd-31-Barsukov-P-O\\aimenv\\Lib\\site-packages\\woodwork\\type_sys\\utils.py:33: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n", + " pd.to_datetime(\n", + "c:\\ulstu\\mii\\AIM-PIbd-31-Barsukov-P-O\\aimenv\\Lib\\site-packages\\woodwork\\type_sys\\utils.py:33: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n", + " pd.to_datetime(\n", + "c:\\ulstu\\mii\\AIM-PIbd-31-Barsukov-P-O\\aimenv\\Lib\\site-packages\\woodwork\\type_sys\\utils.py:33: UserWarning: Could not infer format, so each element will be parsed individually, falling back to `dateutil`. To ensure parsing is consistent and as-expected, please specify a format.\n", + " pd.to_datetime(\n", + "c:\\ulstu\\mii\\AIM-PIbd-31-Barsukov-P-O\\aimenv\\Lib\\site-packages\\featuretools\\synthesis\\deep_feature_synthesis.py:169: UserWarning: Only one dataframe in entityset, changing max_depth to 1 since deeper features cannot be created\n", + " warnings.warn(\n", + "c:\\ulstu\\mii\\AIM-PIbd-31-Barsukov-P-O\\aimenv\\Lib\\site-packages\\featuretools\\synthesis\\dfs.py:321: UnusedPrimitiveWarning: Some specified primitives were not used during DFS:\n", + " agg_primitives: ['count', 'max', 'mean', 'median', 'min', 'std', 'sum']\n", + "This may be caused by a using a value of max_depth that is too small, not setting interesting values, or it may indicate no compatible columns for the primitive were found in the data. If the DFS call contained multiple instances of a primitive in the list above, none of them were used.\n", + " warnings.warn(warning_msg, UnusedPrimitiveWarning)\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
agesexbmichildrensmokerregionchargesage + bmiage + chargesage + children...age * childrenbmi * chargesbmi * childrencharges * childrenage - bmiage - chargesage - childrenbmi - chargesbmi - childrencharges - children
index
019female27.9000Truesouthwest16884.9240046.90016903.9240019.0...0.0471089.3796000.000.0000-8.900-16865.9240019.0-16857.0240027.90016884.92400
118male33.7701Falsesoutheast1725.5523051.7701743.5523019.0...18.058271.90117133.771725.5523-15.770-1707.5523017.0-1691.7823032.7701724.55230
228male33.0003Falsesoutheast4449.4620061.0004477.4620031.0...84.0146832.24600099.0013348.3860-5.000-4421.4620025.0-4416.4620030.0004446.46200
333male22.7050Falsenorthwest21984.4706155.70522017.4706133.0...0.0499157.4052000.000.000010.295-21951.4706133.0-21961.7656122.70521984.47061
432male28.8800Falsenorthwest3866.8552060.8803898.8552032.0...0.0111674.7781760.000.00003.120-3834.8552032.0-3837.9752028.8803866.85520
..................................................................
270547female45.3201Falsesoutheast8569.8618092.3208616.8618048.0...47.0388386.13677645.328569.86181.680-8522.8618046.0-8524.5418044.3208568.86180
270621female34.6000Falsesouthwest2020.1770055.6002041.1770021.0...0.069898.1242000.000.0000-13.600-1999.1770021.0-1985.5770034.6002020.17700
270719male26.0301Truenorthwest16450.8947045.03016469.8947020.0...19.0428216.78904126.0316450.8947-7.030-16431.8947018.0-16424.8647025.03016449.89470
270823male18.7150Falsenorthwest21595.3822941.71521618.3822923.0...0.0404157.5795570.000.00004.285-21572.3822923.0-21576.6672918.71521595.38229
270954male31.6000Falsesouthwest9850.4320085.6009904.4320054.0...0.0311273.6512000.000.000022.400-9796.4320054.0-9818.8320031.6009850.43200
\n", + "

2710 rows × 37 columns

\n", + "
" + ], + "text/plain": [ + " age sex bmi children smoker region charges \\\n", + "index \n", + "0 19 female 27.900 0 True southwest 16884.92400 \n", + "1 18 male 33.770 1 False southeast 1725.55230 \n", + "2 28 male 33.000 3 False southeast 4449.46200 \n", + "3 33 male 22.705 0 False northwest 21984.47061 \n", + "4 32 male 28.880 0 False northwest 3866.85520 \n", + "... ... ... ... ... ... ... ... \n", + "2705 47 female 45.320 1 False southeast 8569.86180 \n", + "2706 21 female 34.600 0 False southwest 2020.17700 \n", + "2707 19 male 26.030 1 True northwest 16450.89470 \n", + "2708 23 male 18.715 0 False northwest 21595.38229 \n", + "2709 54 male 31.600 0 False southwest 9850.43200 \n", + "\n", + " age + bmi age + charges age + children ... age * children \\\n", + "index ... \n", + "0 46.900 16903.92400 19.0 ... 0.0 \n", + "1 51.770 1743.55230 19.0 ... 18.0 \n", + "2 61.000 4477.46200 31.0 ... 84.0 \n", + "3 55.705 22017.47061 33.0 ... 0.0 \n", + "4 60.880 3898.85520 32.0 ... 0.0 \n", + "... ... ... ... ... ... \n", + "2705 92.320 8616.86180 48.0 ... 47.0 \n", + "2706 55.600 2041.17700 21.0 ... 0.0 \n", + "2707 45.030 16469.89470 20.0 ... 19.0 \n", + "2708 41.715 21618.38229 23.0 ... 0.0 \n", + "2709 85.600 9904.43200 54.0 ... 0.0 \n", + "\n", + " bmi * charges bmi * children charges * children age - bmi \\\n", + "index \n", + "0 471089.379600 0.00 0.0000 -8.900 \n", + "1 58271.901171 33.77 1725.5523 -15.770 \n", + "2 146832.246000 99.00 13348.3860 -5.000 \n", + "3 499157.405200 0.00 0.0000 10.295 \n", + "4 111674.778176 0.00 0.0000 3.120 \n", + "... ... ... ... ... \n", + "2705 388386.136776 45.32 8569.8618 1.680 \n", + "2706 69898.124200 0.00 0.0000 -13.600 \n", + "2707 428216.789041 26.03 16450.8947 -7.030 \n", + "2708 404157.579557 0.00 0.0000 4.285 \n", + "2709 311273.651200 0.00 0.0000 22.400 \n", + "\n", + " age - charges age - children bmi - charges bmi - children \\\n", + "index \n", + "0 -16865.92400 19.0 -16857.02400 27.900 \n", + "1 -1707.55230 17.0 -1691.78230 32.770 \n", + "2 -4421.46200 25.0 -4416.46200 30.000 \n", + "3 -21951.47061 33.0 -21961.76561 22.705 \n", + "4 -3834.85520 32.0 -3837.97520 28.880 \n", + "... ... ... ... ... \n", + "2705 -8522.86180 46.0 -8524.54180 44.320 \n", + "2706 -1999.17700 21.0 -1985.57700 34.600 \n", + "2707 -16431.89470 18.0 -16424.86470 25.030 \n", + "2708 -21572.38229 23.0 -21576.66729 18.715 \n", + "2709 -9796.43200 54.0 -9818.83200 31.600 \n", + "\n", + " charges - children \n", + "index \n", + "0 16884.92400 \n", + "1 1724.55230 \n", + "2 4446.46200 \n", + "3 21984.47061 \n", + "4 3866.85520 \n", + "... ... \n", + "2705 8568.86180 \n", + "2706 2020.17700 \n", + "2707 16449.89470 \n", + "2708 21595.38229 \n", + "2709 9850.43200 \n", + "\n", + "[2710 rows x 37 columns]" + ] + }, + "execution_count": 602, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "import featuretools as ft\n", + "\n", + "es = ft.EntitySet(id='insurance')\n", + "\n", + "es = es.add_dataframe(dataframe_name=\"insurance_data\", dataframe=df, index='index')\n", + "\n", + "agg_primitives = [\"sum\", \"mean\", \"median\", \"std\", \"max\", \"min\", \"count\"]\n", + "trans_primitives = [\"add_numeric\", \"multiply_numeric\", \"divide_numeric\", \"subtract_numeric\"]\n", + "\n", + "feature_matrix, feature_defs = ft.dfs(\n", + " entityset=es,\n", + " target_dataframe_name='insurance_data',\n", + " agg_primitives=agg_primitives,\n", + " trans_primitives=trans_primitives,\n", + " max_depth=2\n", + ")\n", + "feature_matrix" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Были сконструированы признаки с помощью Featuretools" + ] + }, + { + "cell_type": "code", + "execution_count": 603, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Index(['age', 'bmi', 'children', 'smoker_yes', 'sex_male', 'region_northwest',\n", + " 'region_southeast', 'region_southwest', 'children_1', 'children_2',\n", + " 'children_3', 'children_4', 'parent_yes', 'age_norm', 'age_stand'],\n", + " dtype='object')\n", + "0.02249455451965332 33039788.648656577 0.7496335938888106\n" + ] + }, + { + "data": { + "text/plain": [ + "Text(0, 0.5, 'Прогнозируемая цена')" + ] + }, + "execution_count": 603, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "train_X = train_df.drop(\"charges\", axis=1)\n", + "train_Y = train_df['charges']\n", + "\n", + "test_X = test_df.drop(\"charges\", axis=1)\n", + "test_Y = test_df['charges']\n", + "\n", + "val_X = val_df.drop(\"charges\", axis=1)\n", + "val_Y = val_df['charges']\n", + "\n", + "train_X = train_X.drop(['smoker', 'sex', 'region', 'age_category', 'bmi_category'], axis=1)\n", + "test_X = test_X.drop(['smoker', 'sex', 'region', 'age_category', 'bmi_category'], axis=1)\n", + "val_X = val_X.drop(['smoker', 'sex', 'region', 'age_category', 'bmi_category'], axis=1)\n", + "\n", + "print(train_X.columns)\n", + "\n", + "from sklearn.linear_model import LinearRegression\n", + "import time \n", + "from sklearn.metrics import mean_squared_error, r2_score\n", + "\n", + "model = LinearRegression()\n", + "\n", + "start_time = time.time()\n", + "model.fit(train_X, train_Y)\n", + "train_time = time.time() - start_time\n", + "\n", + "val_predictions = model.predict(val_X)\n", + "mse = mean_squared_error(val_Y, val_predictions)\n", + "r2 = r2_score(val_Y, val_predictions)\n", + "\n", + "print(train_time, mse, r2)\n", + "\n", + "plt.scatter(val_Y, val_predictions, alpha=0.5)\n", + "plt.plot([val_Y.min(), val_Y.max()], [val_Y.min(), val_Y.max()], 'k--', lw=1)\n", + "plt.xlabel('Фактическая цена')\n", + "plt.ylabel('Прогнозируемая цена')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Обученная модель довольно точно предсказывает цены ниже двух тысяч, для цен более двух тысяч модель занижает или завышает цены" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "aimenv", + "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": "3.12.5" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +}