{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Набор данных \"Ближайшие к Земле объекты\" "
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"# Наборе данных \"Ближайшие к Земле объекты\" \n",
"# \n",
"# 1. Анализ сведений о наборе данных \n",
"# Набор данных о \"ближайших к Земле объектах\" относится к области астрофизики и исследования космических угроз. Проблемная область — мониторинг околоземных объектов (Near-Earth Objects, NEO), которые могут представлять потенциальную опасность для Земли. Цель анализа — оценить характеристики объектов, их орбиты, скорость, и расстояние до Земли, чтобы предсказать возможность столкновения. \n",
"# \n",
"# 2. Анализ содержимого набора данных \n",
"# Объекты наблюдения: \n",
"# Космические тела, которые классифицируются как \"околоземные объекты\". \n",
"# Атрибуты объектов: \n",
"# id — идентификатор объекта. \n",
"# name — название объекта. \n",
"# est_diameter_min и est_diameter_max — оценочные минимальный и максимальный диаметры. \n",
"# relative_velocity — относительная скорость объекта. \n",
"# miss_distance — минимальное расстояние до Земли. \n",
"# orbiting_body — небесное тело, вокруг которого объект совершает движение (обычно Солнце). \n",
"# sentry_object — булевый признак, указывающий, находится ли объект в базе наблюдения системы Sentry. \n",
"# absolute_magnitude — абсолютная звёздная величина объекта. \n",
"# hazardous — булевый признак опасности объекта. \n",
"# Связи между объектами: \n",
"# Объекты могут быть связаны через общее орбитальное тело или группироваться по характеристикам орбит. \n",
"# \n",
"# 3. Примеры бизнес-целей \n",
"# Уведомление о космических угрозах: \n",
"# Построение системы прогнозирования опасных объектов для оперативного предупреждения. \n",
"# Эффект: повышение безопасности через своевременные предупреждения. \n",
"# \n",
"# Оптимизация миссий для изучения астероидов: \n",
"# Идентификация объектов, подходящих для изучения (на основе характеристик орбиты и размера). \n",
"# Эффект: снижение затрат и повышение научной ценности миссий. \n",
"# \n",
"# 4. Примеры целей технического проекта \n",
"# Бизнес-цель: Предсказание опасности объекта. \n",
"# Входные данные: атрибуты объекта (диаметр, скорость, расстояние). \n",
"# Целевой признак: hazardous. \n",
"# \n",
"# Бизнес-цель: Классификация объектов по характеристикам. \n",
"# Входные данные: абсолютная звёздная величина, диаметр, скорость. \n",
"# Целевой признак: принадлежность к определённой группе объектов. "
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import pandas as pd\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"import numpy as np\n",
"\n",
"df = pd.read_csv('../../datasets/neo.csv')\n",
"df = df.drop(columns=['name', 'orbiting_body', 'sentry_object'])\n",
"\n",
"\n",
"#5. Устранение пропущенных данных\n",
" \n",
"#Сведения о пропущенных данных\n",
"print(\"Количество пропущенных значений в каждом столбце:\")\n",
"print(df.isnull().sum())\n",
"\n",
"# Процент пропущенных значений признаков\n",
"for i in df.columns:\n",
" null_rate = df[i].isnull().sum() / len(df) * 100\n",
" if null_rate > 0:\n",
" print(f'{i} Процент пустых значений: %{null_rate:.2f}')\n",
"\n",
"\n",
"#Пропущенных данных в датасете нет\n",
"\n",
"\n",
"\n",
"\n",
"#6. Проблемы набора данных\n",
" #5.1Выбросы: Возможны аномалии в значениях скорости или расстояния.\n",
" #Смещение: Данные могут быть смещены в сторону объектов, которые легче обнаружить (крупные, близкие).\n",
"\n",
"#7. Решения для обнаруженных проблем\n",
" #Выбросы: Идентификация и обработка выбросов через методы (например, IQR или Z-оценка).\n",
" #Смещение: Использование методов балансировки данных, таких как oversampling.\n",
"\n",
"#7.1 Проверка набора данных на выбросы\n",
"# Выбираем столбцы для анализа\n",
"columns_to_check = ['est_diameter_min', 'est_diameter_max', 'relative_velocity', 'miss_distance', 'absolute_magnitude']\n",
"def Emissions(columns_to_check):\n",
"\n",
" # Функция для подсчета выбросов\n",
" def count_outliers(df, columns):\n",
" outliers_count = {}\n",
" for col in columns:\n",
" Q1 = df[col].quantile(0.25)\n",
" Q3 = df[col].quantile(0.75)\n",
" IQR = Q3 - Q1\n",
" lower_bound = Q1 - 1.5 * IQR\n",
" upper_bound = Q3 + 1.5 * IQR\n",
" \n",
" # Считаем количество выбросов\n",
" outliers = df[(df[col] < lower_bound) | (df[col] > upper_bound)]\n",
" outliers_count[col] = len(outliers)\n",
" \n",
" return outliers_count\n",
"\n",
" # Подсчитываем выбросы\n",
" outliers_count = count_outliers(df, columns_to_check)\n",
"\n",
" # Выводим количество выбросов для каждого столбца\n",
" for col, count in outliers_count.items():\n",
" print(f\"Количество выбросов в столбце '{col}': {count}\")\n",
" \n",
" # Создаем гистограммы\n",
" plt.figure(figsize=(15, 10))\n",
" for i, col in enumerate(columns_to_check, 1):\n",
" plt.subplot(2, 3, i)\n",
" sns.histplot(df[col], kde=True)\n",
" plt.title(f'Histogram of {col}')\n",
" plt.tight_layout()\n",
" plt.show()\n",
"Emissions(columns_to_check)\n",
"\n",
"#Признак miss_distance не имеет выбросов, \n",
"#признак absolute_magnitude имеет количество выбросов в приемлемом диапазоне\n",
"#для признаков est_diameter_min, est_diameter_max и relative_velocity необходимо использовать метод решения проблемы выбросов. \n",
"#Воспользуемся методом усреднения значений:\n",
"columns_to_fix = ['est_diameter_min', 'est_diameter_max', 'relative_velocity']\n",
"# Замена нулей и пропущенных значений средним\n",
"for column in columns_to_fix:\n",
" # Расчет среднего, исключая 0 и NaN\n",
" mean_value = df[df[column] > 0][column].mean()\n",
" # Замена NaN и нулей на среднее\n",
" df[column] = df[column].replace(0, np.nan).fillna(mean_value)\n",
"\n",
"#Оценим выбросы в выборке после усреднения:\n",
"Emissions(columns_to_fix)\n",
"\n",
"#Удалось избавиться от выбросов в соответствующих признаках как видно на диаграммах.\n",
"\n",
"\n",
"\n",
"#8. Разбиение данных на выборки\n",
"\n",
"from sklearn.model_selection import train_test_split\n",
"\n",
"train_data, temp_data = train_test_split(df, test_size=0.3, random_state=42)\n",
"val_data, test_data = train_test_split(temp_data, test_size=0.5, random_state=42)\n",
"\n",
"print(\"Обучающая выборка: \", train_data.shape)\n",
"print(train_data.hazardous.value_counts())\n",
"hazardous_counts = train_data['hazardous'].value_counts()\n",
"plt.figure(figsize=(2, 2))\n",
"plt.pie(hazardous_counts, labels=hazardous_counts.index, autopct='%1.1f%%', startangle=90)\n",
"plt.title('Распределение классов hazardous в обучающей выборке')\n",
"plt.show()\n",
"\n",
"print(\"Контрольная выборка: \", val_data.shape)\n",
"print(val_data.hazardous.value_counts())\n",
"hazardous_counts = val_data['hazardous'].value_counts()\n",
"plt.figure(figsize=(2, 2))\n",
"plt.pie(hazardous_counts, labels=hazardous_counts.index, autopct='%1.1f%%', startangle=90)\n",
"plt.title('Распределение классов hazardous в контрольной выборке')\n",
"plt.show()\n",
"\n",
"print(\"Тестовая выборка: \", test_data.shape)\n",
"print(test_data.hazardous.value_counts())\n",
"hazardous_counts = test_data['hazardous'].value_counts()\n",
"plt.figure(figsize=(2, 2))\n",
"plt.pie(hazardous_counts, labels=hazardous_counts.index, autopct='%1.1f%%', startangle=90)\n",
"plt.title('Распределение классов hazardous в тестовой выборке')\n",
"plt.show()\n",
"\n",
"\n",
"#9. Оценить сбалансированность выборок для каждого набора данных. Оценить необходимость использования методов приращения (аугментации) данных. \n",
"#Выводы по сбалансированности\n",
"#Если распределение классов примерно равно (например, 50%/50%), выборка считается сбалансированной, и аугментация данных не требуется.\n",
"#Если один из классов сильно доминирует (например, 90%/10%), выборка несбалансированная, и может потребоваться аугментация данных.\n",
"\n",
"#Данная сборка несбалансированная, и требуется аугментация данных.\n",
"\n",
"\n",
"#10. Выполнить приращение данных методами выборки с избытком (oversampling) и выборки с недостатком (undersampling). Должны быть представлены примеры реализации обоих методов для выборок набора данных. \n",
"\n",
"#10.1\n",
"#Аугментация данных методом оверсемплинга¶\n",
"#Этот метод увеличивает количество примеров меньшинства.\n",
"\n",
"from imblearn.over_sampling import ADASYN\n",
"\n",
"# Создание экземпляра ADASYN\n",
"ada = ADASYN()\n",
"\n",
"# Применение ADASYN\n",
"X_resampled, y_resampled = ada.fit_resample(train_data.drop(columns=['hazardous']), train_data['hazardous'])\n",
"\n",
"# Создание нового DataFrame\n",
"df_train_adasyn = pd.DataFrame(X_resampled)\n",
"df_train_adasyn['hazardous'] = y_resampled # Добавление целевой переменной\n",
"\n",
"# Вывод информации о новой выборке\n",
"print(\"Обучающая выборка после оверсемплинга: \", df_train_adasyn.shape)\n",
"print(df_train_adasyn['hazardous'].value_counts())\n",
"hazardous_counts = df_train_adasyn['hazardous'].value_counts()\n",
"plt.figure(figsize=(2, 2))\n",
"plt.pie(hazardous_counts, labels=hazardous_counts.index, autopct='%1.1f%%', startangle=90)\n",
"plt.title('Распределение классов hazardous в обучающей выборке после оверсемплинга')\n",
"plt.show()\n",
"\n",
"\n",
"#10.2\n",
"#Аугментация данных методом андерсемплинга\n",
"#Этот метод помогает сбалансировать выборку, уменьшая количество экземпляров класса большинства, чтобы привести его в соответствие с классом меньшинства\n",
"\n",
"from imblearn.under_sampling import RandomUnderSampler\n",
"\n",
"rus = RandomUnderSampler()\n",
"\n",
"# Применение RandomUnderSampler\n",
"X_resampled, y_resampled = rus.fit_resample(train_data.drop(columns=['hazardous']), train_data['hazardous'])\n",
"\n",
"# Создание нового DataFrame\n",
"df_train_undersampled = pd.DataFrame(X_resampled)\n",
"df_train_undersampled['hazardous'] = y_resampled # Добавление целевой переменной\n",
"\n",
"# Вывод информации о новой выборке\n",
"print(\"Обучающая выборка после андерсемплинга: \", df_train_undersampled.shape)\n",
"print(df_train_undersampled['hazardous'].value_counts())\n",
"\n",
"# Визуализация распределения классов\n",
"hazardous_counts = df_train_undersampled['hazardous'].value_counts()\n",
"plt.figure(figsize=(2, 2))\n",
"plt.pie(hazardous_counts, labels=hazardous_counts.index, autopct='%1.1f%%', startangle=90)\n",
"plt.title('Распределение классов hazardous в обучающей выборке после андерсемплинга')\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Набор данных \"Продажа домов в округе Кинг, США\""
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"# 1. Анализ сведений о наборе данных \"Продажа домов в округе Кинг, США\"\n",
"# Данный набор данных содержит информацию о работе филиалов супермаркетов. Проблемная область связана с анализом производительности магазинов, оптимизацией их # работы и улучшением ключевых показателей, таких как продажи и привлечение клиентов. Основные вопросы, которые могут возникнуть:\n",
"# \n",
"# 2. Анализ содержимого набора данных\n",
"# Объекты наблюдения\n",
"# Каждая строка представляет собой данные о конкретном магазине.\n",
"# \n",
"# Атрибуты объектов\n",
"# Store ID (идентификатор магазина): уникальный номер, однозначно идентифицирующий магазин.\n",
"# Store_Area (площадь магазина): общая площадь магазина в квадратных метрах.\n",
"# Items_Available (количество доступных товаров): общее количество товаров в магазине.\n",
"# Daily_Customer_Count (ежедневное число покупателей): среднее число покупателей в день.\n",
"# Store_Sales (продажи магазина): общая выручка магазина.\n",
"# \n",
"# Связи между объектами\n",
"# Объекты (магазины) связаны общей бизнес-целью (рост прибыли) и схожими условиями работы (товары, покупатели,площади). Возможные связи:\n",
"# Прямая связь между числом покупателей и выручкой.\n",
"# Влияние площади и ассортимента на посещаемость и продажи.\n",
"# \n",
"# 3. Примеры бизнес-целей и эффекты для бизнеса\n",
"# Бизнес-цели\n",
"# Оптимизация ассортимента: определить, как ассортимент товаров влияет на продажи.\n",
"# Увеличение посещаемости: найти факторы, повышающие интерес покупателей к магазину.\n",
"# Оптимизация площадей: выявить оптимальные площади магазинов для максимальной выручки.\n",
"# Анализ производительности филиалов: сравнить эффективность магазинов и выявить лучшие практики.\n",
"# Эффекты для бизнеса\n",
"# Повышение выручки за счет оптимизации ассортимента и улучшения опыта покупателей.\n",
"# Сокращение затрат на неэффективные площади.\n",
"# Более точное планирование при открытии новых магазинов.\n",
"# \n",
"# 4. Примеры целей технического проекта\n",
"# Цель 1: Оптимизация ассортимента\n",
"# Входные данные: Store_Area, Items_Available, Store_Sales.\n",
"# Целевой признак: Store_Sales.\n",
"# Результат: рекомендации по увеличению или уменьшению ассортимента.\n",
"# Цель 2: Увеличение посещаемости\n",
"# Входные данные: Store_Area, Items_Available, Store_Sales.\n",
"# Целевой признак: Daily_Customer_Count.\n",
"# Результат: прогноз посещаемости на основе характеристик магазина.\n",
"# Цель 3: Оптимизация площадей\n",
"# Входные данные: Store_Area, Store_Sales.\n",
"# Целевой признак: Store_Sales.\n",
"# Результат: определение оптимальной площади для магазина.\n",
"# Цель 4: Анализ производительности филиалов\n",
"# Входные данные: Все атрибуты.\n",
"# Целевой признак: Store_Sales.\n",
"# Результат: классификация магазинов по эффективности и выявление отстающих филиалов."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import pandas as pd\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"import numpy as np\n",
"\n",
"df = pd.read_csv('../../datasets/kc_house_data.csv')\n",
"df = df.drop(columns=['id'])\n",
"\n",
"\n",
"#5. Устранение пропущенных данных\n",
" \n",
"#Сведения о пропущенных данных\n",
"print(\"Количество пропущенных значений в каждом столбце:\")\n",
"print(df.isnull().sum())\n",
"\n",
"# Процент пропущенных значений признаков\n",
"for i in df.columns:\n",
" null_rate = df[i].isnull().sum() / len(df) * 100\n",
" if null_rate > 0:\n",
" print(f'{i} Процент пустых значений: %{null_rate:.2f}')\n",
"\n",
"\n",
"#Пропущенных данных в датасете нет\n",
"\n",
"\n",
"\n",
"\n",
"#6. Проблемы набора данных\n",
" #5.1Выбросы: Возможны аномалии в значениях скорости или расстояния.\n",
" #Смещение: Данные могут быть смещены в сторону объектов, которые легче обнаружить (крупные, близкие).\n",
"\n",
"#7. Решения для обнаруженных проблем\n",
" #Выбросы: Идентификация и обработка выбросов через методы (например, IQR или Z-оценка).\n",
" #Смещение: Использование методов балансировки данных, таких как oversampling.\n",
"\n",
"#7.1 Проверка набора данных на выбросы\n",
"# Выбираем столбцы для анализа\n",
"columns_to_check = ['price', 'sqft_living', 'bathrooms', 'yr_built']\n",
"def Emissions(columns_to_check):\n",
"\n",
" # Функция для подсчета выбросов\n",
" def count_outliers(df, columns):\n",
" outliers_count = {}\n",
" for col in columns:\n",
" Q1 = df[col].quantile(0.25)\n",
" Q3 = df[col].quantile(0.75)\n",
" IQR = Q3 - Q1\n",
" lower_bound = Q1 - 1.5 * IQR\n",
" upper_bound = Q3 + 1.5 * IQR\n",
" \n",
" # Считаем количество выбросов\n",
" outliers = df[(df[col] < lower_bound) | (df[col] > upper_bound)]\n",
" outliers_count[col] = len(outliers)\n",
" \n",
" return outliers_count\n",
"\n",
" # Подсчитываем выбросы\n",
" outliers_count = count_outliers(df, columns_to_check)\n",
"\n",
" # Выводим количество выбросов для каждого столбца\n",
" for col, count in outliers_count.items():\n",
" print(f\"Количество выбросов в столбце '{col}': {count}\")\n",
" \n",
" # Создаем гистограммы\n",
" plt.figure(figsize=(15, 10))\n",
" for i, col in enumerate(columns_to_check, 1):\n",
" plt.subplot(2, 3, i)\n",
" sns.histplot(df[col], kde=True)\n",
" plt.title(f'Histogram of {col}')\n",
" plt.tight_layout()\n",
" plt.show()\n",
"Emissions(columns_to_check)\n",
"\n",
"#Признак yr_built не имеет выбросов, \n",
"#для признаков 'price', 'sqft_living', 'bathrooms' необходимо использовать метод решения проблемы выбросов. \n",
"#Воспользуемся методом удаления наблюдений с такими выбросами:\n",
"# Выбираем столбцы для очистки\n",
"columns_to_fix = ['price', 'sqft_living', 'bathrooms']\n",
"\n",
"# Функция для удаления выбросов\n",
"def remove_outliers(df, columns):\n",
" for col in columns:\n",
" Q1 = df[col].quantile(0.25)\n",
" Q3 = df[col].quantile(0.75)\n",
" IQR = Q3 - Q1\n",
" lower_bound = Q1 - 1.5 * IQR\n",
" upper_bound = Q3 + 1.5 * IQR\n",
" \n",
" # Удаляем строки, содержащие выбросы\n",
" df = df[(df[col] >= lower_bound) & (df[col] <= upper_bound)]\n",
" \n",
" return df\n",
"\n",
"# Удаляем выбросы\n",
"df_cleaned = remove_outliers(df, columns_to_fix)\n",
"\n",
"# Выводим количество удаленных строк\n",
"print(f\"Количество удаленных строк: {len(df) - len(df_cleaned)}\")\n",
"\n",
"df = df_cleaned\n",
"\n",
"#Оценим выбросы в выборке после усреднения:\n",
"Emissions(columns_to_fix)\n",
"\n",
"#Удалось избавиться от выбросов в соответствующих признаках как видно на диаграммах.\n",
"\n",
"\n",
"\n",
"#8. Разбиение данных на выборки\n",
"\n",
"from sklearn.model_selection import train_test_split\n",
"\n",
"train_data, temp_data = train_test_split(df, test_size=0.3, random_state=42)\n",
"val_data, test_data = train_test_split(temp_data, test_size=0.5, random_state=42)\n",
"\n",
"# Средние значения цены\n",
"print(\"Средняя цена в обучающей выборке:\", train_data['price'].mean())\n",
"print(\"Средняя цена в контрольной выборке:\", val_data['price'].mean())\n",
"print(\"Средняя цена в тестовой выборке:\", test_data['price'].mean())\n",
"print()\n",
"\n",
"# Стандартное отклонение цены\n",
"print(\"Стандартное отклонение цены в обучающей выборке:\", train_data['price'].std())\n",
"print(\"Стандартное отклонение цены в контрольной выборке:\", val_data['price'].std())\n",
"print(\"Стандартное отклонение цены в тестовой выборке:\", test_data['price'].std())\n",
"print()\n",
"\n",
"# Проверка распределений по количеству объектов в диапазонах\n",
"print(\"Распределение по квартилам (обучающая):\")\n",
"print(train_data['price'].quantile([0.25, 0.5, 0.75]))\n",
"print()\n",
"print(\"Распределение по квартилам (контрольная):\")\n",
"print(val_data['price'].quantile([0.25, 0.5, 0.75]))\n",
"print()\n",
"print(\"Распределение по квартилам (тестовая):\")\n",
"print(test_data['price'].quantile([0.25, 0.5, 0.75]))\n",
"\n",
"# Построение гистограмм для каждой выборки\n",
"plt.figure(figsize=(12, 6))\n",
"\n",
"sns.histplot(train_data['price'], color='blue', label='Train', kde=True)\n",
"sns.histplot(val_data['price'], color='green', label='Validation', kde=True)\n",
"sns.histplot(test_data['price'], color='red', label='Test', kde=True)\n",
"\n",
"plt.legend()\n",
"plt.xlabel('price')\n",
"plt.ylabel('Frequency')\n",
"plt.title('Распределение цены в обучающей, контрольной и тестовой выборках')\n",
"plt.show()\n",
"\n",
"\n",
"#9. Оценить сбалансированность выборок для каждого набора данных. Оценить необходимость использования методов приращения (аугментации) данных. \n",
"#Выводы по сбалансированности\n",
"#Если распределение классов примерно равно (например, 50%/50%), выборка считается сбалансированной, и аугментация данных не требуется.\n",
"#Если один из классов сильно доминирует (например, 90%/10%), выборка несбалансированная, и может потребоваться аугментация данных.\n",
"\n",
"#Выборки оказались недостаточно сбалансированными. Используем методы приращения данных с избытком и с недостатком:\n",
"\n",
"\n",
"#10. Выполнить приращение данных методами выборки с избытком (oversampling) и выборки с недостатком (undersampling). Должны быть представлены примеры реализации обоих методов для выборок набора данных. \n",
"\n",
"#10.1\n",
"#Аугментация данных методом оверсемплинга¶\n",
"#Этот метод увеличивает количество примеров меньшинства.\n",
"\n",
"from imblearn.over_sampling import RandomOverSampler\n",
"from imblearn.under_sampling import RandomUnderSampler\n",
"\n",
"def oversample(df, target_column):\n",
" X = df.drop(target_column, axis=1)\n",
" y = df[target_column]\n",
" \n",
" oversampler = RandomOverSampler(random_state=42)\n",
" x_resampled, y_resampled = oversampler.fit_resample(X, y)\n",
" \n",
" resampled_df = pd.concat([x_resampled, y_resampled], axis=1) \n",
" return resampled_df\n",
"\n",
"def undersample(df, target_column):\n",
" X = df.drop(target_column, axis=1)\n",
" y = df[target_column]\n",
" \n",
" undersampler = RandomUnderSampler(random_state=42)\n",
" x_resampled, y_resampled = undersampler.fit_resample(X, y)\n",
" \n",
" resampled_df = pd.concat([x_resampled, y_resampled], axis=1)\n",
" return resampled_df\n",
"\n",
"train_df_oversampled = oversample(train_data, 'price')\n",
"val_df_oversampled = oversample(val_data, 'price')\n",
"test_df_oversampled = oversample(test_data, 'price')\n",
"\n",
"train_df_undersampled = undersample(train_data, 'price')\n",
"val_df_undersampled = undersample(val_data, 'price')\n",
"test_df_undersampled = undersample(test_data, 'price')\n",
"\n",
"# Построение гистограмм для каждой выборки\n",
"plt.figure(figsize=(12, 6))\n",
"\n",
"sns.histplot(train_df_undersampled['price'], color='blue', label='Train', kde=True)\n",
"sns.histplot(val_df_undersampled['price'], color='green', label='Validation', kde=True)\n",
"sns.histplot(test_df_undersampled['price'], color='red', label='Test', kde=True)\n",
"\n",
"plt.legend()\n",
"plt.xlabel('price')\n",
"plt.ylabel('Frequency')\n",
"plt.title('Распределение цены в обучающей, контрольной и тестовой выборках (андерсемплинг)')\n",
"plt.show()\n",
"\n",
"# Построение гистограмм для каждой выборки\n",
"plt.figure(figsize=(12, 6))\n",
"\n",
"sns.histplot(train_df_oversampled['price'], color='blue', label='Train', kde=True)\n",
"sns.histplot(val_df_oversampled['price'], color='green', label='Validation', kde=True)\n",
"sns.histplot(test_df_oversampled['price'], color='red', label='Test', kde=True)\n",
"\n",
"plt.legend()\n",
"plt.xlabel('price')\n",
"plt.ylabel('Frequency')\n",
"plt.title('Распределение цены в обучающей, контрольной и тестовой выборках (оверсемплинг)')\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Набор данных \"Показатели сердечно-сосудистых заболеваний (за 2022 год)\""
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"# 1. Анализ сведений о наборе данных\n",
"# Этот набор данных из Kaggle, вероятно, относится к исследованию диагностики диабета у женщин индийского происхождения. Набор данных изначально был взят из Национального института диабета и болезней пищеварения и почек (Pima Indians Diabetes Database).\n",
"# \n",
"# Проблемная область\n",
"# Основная проблемная область — медицинская диагностика. Цель данных — предсказать наличие диабета у пациента на основе медицинских показателей (например, уровня глюкозы, давления, ИМТ) и других факторов.\n",
"# \n",
"# 2. Анализ содержимого набора данных\n",
"# Объекты наблюдения\n",
"# Объектами наблюдения являются медицинские записи пациентов (768 строк). Каждая строка представляет собой данные о состоянии конкретного пациента.\n",
"# \n",
"# Атрибуты объектов\n",
"# Pregnancies (количество беременностей) — числовой признак.\n",
"# Glucose (уровень глюкозы в крови) — числовой признак.\n",
"# BloodPressure (артериальное давление) — числовой признак.\n",
"# SkinThickness (толщина кожной складки) — числовой признак.\n",
"# Insulin (уровень инсулина) — числовой признак.\n",
"# BMI (индекс массы тела) — вещественный признак.\n",
"# DiabetesPedigreeFunction (генетическая предрасположенность к диабету) — вещественный признак.\n",
"# Age (возраст) — числовой признак.\n",
"# Outcome (наличие диабета) — целевой бинарный признак (1 — диабет, 0 — нет).\n",
"# \n",
"# Связи между объектами\n",
"# Объекты наблюдения независимы друг от друга, так как каждый пациент представлен в одном экземпляре.\n",
"# \n",
"# 3. Примеры бизнес-целей\n",
"# Улучшение диагностики диабета\n",
"# Создание инструмента для раннего выявления диабета на основе медицинских данных.\n",
"# Эффект: Снижение затрат на лечение поздних стадий диабета, повышение уровня здоровья пациентов.\n",
"# \n",
"# Оптимизация медицинских услуг\n",
"# Предсказание риска диабета для отдельных групп населения.\n",
"# Эффект: Перераспределение ресурсов на профилактические меры и профилактические программы.\n",
"# \n",
"# Поддержка медицинских решений\n",
"# Автоматизированная рекомендация дополнительных обследований или консультаций для пациентов с высоким риском.\n",
"# Эффект: Ускорение процесса диагностики, уменьшение нагрузки на врачей.\n",
"# \n",
"# 4. Примеры целей технического проекта\n",
"# Цель 1: Улучшение диагностики диабета\n",
"# Вход: Все признаки, кроме Outcome.\n",
"# Целевой признак: Outcome (наличие диабета).\n",
"# Результат: Модель классификации, которая определяет, есть ли у пациента диабет.\n",
"# Цель 2: Оптимизация медицинских услуг\n",
"# Вход: Социально-демографические данные (возраст, количество беременностей) и факторы здоровья (ИМТ, глюкоза).\n",
"# Целевой признак: Вероятность (риск) наличия диабета.\n",
"# Результат: Отчет о вероятности диабета для различных групп пациентов.\n",
"# Цель 3: Поддержка медицинских решений\n",
"# Вход: Все признаки набора данных.\n",
"# Целевой признак: Рекомендации для дальнейших действий (например, \"провести анализ инсулина\", \"назначить консультацию эндокринолога\").\n",
"# Результат: Система поддержки принятия решений для врачей.\n",
"# "
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Pregnancies
\n",
"
Glucose
\n",
"
BloodPressure
\n",
"
SkinThickness
\n",
"
Insulin
\n",
"
BMI
\n",
"
DiabetesPedigreeFunction
\n",
"
Age
\n",
"
Outcome
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
6
\n",
"
148
\n",
"
72
\n",
"
35
\n",
"
0
\n",
"
33.6
\n",
"
0.627
\n",
"
50
\n",
"
1
\n",
"
\n",
"
\n",
"
1
\n",
"
1
\n",
"
85
\n",
"
66
\n",
"
29
\n",
"
0
\n",
"
26.6
\n",
"
0.351
\n",
"
31
\n",
"
0
\n",
"
\n",
"
\n",
"
2
\n",
"
8
\n",
"
183
\n",
"
64
\n",
"
0
\n",
"
0
\n",
"
23.3
\n",
"
0.672
\n",
"
32
\n",
"
1
\n",
"
\n",
"
\n",
"
3
\n",
"
1
\n",
"
89
\n",
"
66
\n",
"
23
\n",
"
94
\n",
"
28.1
\n",
"
0.167
\n",
"
21
\n",
"
0
\n",
"
\n",
"
\n",
"
4
\n",
"
0
\n",
"
137
\n",
"
40
\n",
"
35
\n",
"
168
\n",
"
43.1
\n",
"
2.288
\n",
"
33
\n",
"
1
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Pregnancies Glucose BloodPressure SkinThickness Insulin BMI \\\n",
"0 6 148 72 35 0 33.6 \n",
"1 1 85 66 29 0 26.6 \n",
"2 8 183 64 0 0 23.3 \n",
"3 1 89 66 23 94 28.1 \n",
"4 0 137 40 35 168 43.1 \n",
"\n",
" DiabetesPedigreeFunction Age Outcome \n",
"0 0.627 50 1 \n",
"1 0.351 31 0 \n",
"2 0.672 32 1 \n",
"3 0.167 21 0 \n",
"4 2.288 33 1 "
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df = pd.read_csv('../../datasets/diabetes.csv')\n",
"df.head()\n",
"#df.select_dtypes(include=np.number).columns.tolist()"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Количество пропущенных значений в каждом столбце:\n",
"Pregnancies 0\n",
"Glucose 0\n",
"BloodPressure 0\n",
"SkinThickness 0\n",
"Insulin 0\n",
"BMI 0\n",
"DiabetesPedigreeFunction 0\n",
"Age 0\n",
"Outcome 0\n",
"dtype: int64\n",
"Количество выбросов в столбце 'Pregnancies': 4\n",
"Количество выбросов в столбце 'Glucose': 5\n",
"Количество выбросов в столбце 'BloodPressure': 45\n",
"Количество выбросов в столбце 'SkinThickness': 1\n",
"Количество выбросов в столбце 'Insulin': 34\n",
"Количество выбросов в столбце 'BMI': 19\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"date 0\n",
"price 0\n",
"bedrooms 0\n",
"bathrooms 0\n",
"sqft_living 0\n",
"sqft_lot 0\n",
"floors 0\n",
"waterfront 0\n",
"view 0\n",
"condition 0\n",
"grade 0\n",
"sqft_above 0\n",
"sqft_basement 0\n",
"yr_built 0\n",
"yr_renovated 0\n",
"zipcode 0\n",
"lat 0\n",
"long 0\n",
"sqft_living15 0\n",
"sqft_lot15 0\n",
"dtype: int64\n",
"Обучающая выборка: (537, 9)\n",
"Outcome\n",
"0 349\n",
"1 188\n",
"Name: count, dtype: int64\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Контрольная выборка: (115, 9)\n",
"Outcome\n",
"0 78\n",
"1 37\n",
"Name: count, dtype: int64\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Тестовая выборка: (116, 9)\n",
"Outcome\n",
"0 73\n",
"1 43\n",
"Name: count, dtype: int64\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Обучающая выборка после оверсемплинга: (677, 9)\n",
"Outcome\n",
"0 349\n",
"1 328\n",
"Name: count, dtype: int64\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Обучающая выборка после андерсемплинга: (376, 9)\n",
"Outcome\n",
"0 188\n",
"1 188\n",
"Name: count, dtype: int64\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import pandas as pd\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"import numpy as np\n",
"\n",
"df = pd.read_csv('../../datasets/diabetes.csv')\n",
"#df = df.drop(columns=['name', 'orbiting_body', 'sentry_object'])\n",
"\n",
"\n",
"#5. Устранение пропущенных данных\n",
" \n",
"#Сведения о пропущенных данных\n",
"print(\"Количество пропущенных значений в каждом столбце:\")\n",
"print(df.isnull().sum())\n",
"\n",
"# Процент пропущенных значений признаков\n",
"for i in df.columns:\n",
" null_rate = df[i].isnull().sum() / len(df) * 100\n",
" if null_rate > 0:\n",
" print(f'{i} Процент пустых значений: %{null_rate:.2f}')\n",
"\n",
"#Пропущенных данных в датасете нет\n",
"\n",
"\n",
"\n",
"#6. Проблемы набора данных\n",
" #5.1Выбросы: Возможны аномалии в значениях скорости или расстояния.\n",
" #Смещение: Данные могут быть смещены в сторону объектов, которые легче обнаружить (крупные, близкие).\n",
"\n",
"#7. Решения для обнаруженных проблем\n",
" #Выбросы: Идентификация и обработка выбросов через методы (например, IQR или Z-оценка).\n",
" #Смещение: Использование методов балансировки данных, таких как oversampling.\n",
"\n",
"#7.1 Проверка набора данных на выбросы\n",
"# Выбираем столбцы для анализа\n",
"columns_to_check = ['Pregnancies','Glucose','BloodPressure','SkinThickness','Insulin','BMI']\n",
"def Emissions(columns_to_check):\n",
"\n",
" # Функция для подсчета выбросов\n",
" def count_outliers(df, columns):\n",
" outliers_count = {}\n",
" for col in columns:\n",
" Q1 = df[col].quantile(0.25)\n",
" Q3 = df[col].quantile(0.75)\n",
" IQR = Q3 - Q1\n",
" lower_bound = Q1 - 1.5 * IQR\n",
" upper_bound = Q3 + 1.5 * IQR\n",
" \n",
" # Считаем количество выбросов\n",
" outliers = df[(df[col] < lower_bound) | (df[col] > upper_bound)]\n",
" outliers_count[col] = len(outliers)\n",
" \n",
" return outliers_count\n",
"\n",
" # Подсчитываем выбросы\n",
" outliers_count = count_outliers(df, columns_to_check)\n",
"\n",
" # Выводим количество выбросов для каждого столбца\n",
" for col, count in outliers_count.items():\n",
" print(f\"Количество выбросов в столбце '{col}': {count}\")\n",
" \n",
" # Создаем гистограммы\n",
" plt.figure(figsize=(15, 10))\n",
" for i, col in enumerate(columns_to_check, 1):\n",
" plt.subplot(2, 3, i)\n",
" sns.histplot(df[col], kde=True)\n",
" plt.title(f'Histogram of {col}')\n",
" plt.tight_layout()\n",
" plt.show()\n",
"Emissions(columns_to_check)\n",
"\n",
"#Признакb 'Pregnancies','Glucose','BloodPressure','SkinThickness','Insulin','BMI' имеют количество выбросов в приемлемом диапазоне\n",
"#Не нужно проводить устранять проблему пропущенных данных.\n",
"\n",
"#Подстановка константного значения для пропущенных данных. (пример так как кол-во выбросов в пределах нормы)\n",
"constant_value = 0 # Например, подставим 0\n",
"df.fillna(constant_value, inplace=True)\n",
"print(df_cleaned.isna().sum())\n",
"\n",
"\n",
"#8. Разбиение данных на выборки\n",
"\n",
"train_data, temp_data = train_test_split(df, test_size=0.3, random_state=42)\n",
"val_data, test_data = train_test_split(temp_data, test_size=0.5, random_state=42)\n",
"\n",
"print(\"Обучающая выборка: \", train_data.shape)\n",
"print(train_data.Outcome.value_counts())\n",
"Outcome_counts = train_data['Outcome'].value_counts()\n",
"plt.figure(figsize=(2, 2))\n",
"plt.pie(Outcome_counts, labels=Outcome_counts.index, autopct='%1.1f%%', startangle=90)\n",
"plt.title('Распределение классов Outcome в обучающей выборке')\n",
"plt.show()\n",
"\n",
"print(\"Контрольная выборка: \", val_data.shape)\n",
"print(val_data.Outcome.value_counts())\n",
"Outcome_counts = val_data['Outcome'].value_counts()\n",
"plt.figure(figsize=(2, 2))\n",
"plt.pie(Outcome_counts, labels=Outcome_counts.index, autopct='%1.1f%%', startangle=90)\n",
"plt.title('Распределение классов Outcome в контрольной выборке')\n",
"plt.show()\n",
"\n",
"print(\"Тестовая выборка: \", test_data.shape)\n",
"print(test_data.Outcome.value_counts())\n",
"Outcome_counts = test_data['Outcome'].value_counts()\n",
"plt.figure(figsize=(2, 2))\n",
"plt.pie(Outcome_counts, labels=Outcome_counts.index, autopct='%1.1f%%', startangle=90)\n",
"plt.title('Распределение классов Outcome в тестовой выборке')\n",
"plt.show()\n",
"\n",
"\n",
"#9. Оценить сбалансированность выборок для каждого набора данных. Оценить необходимость использования методов приращения (аугментации) данных. \n",
"#Выводы по сбалансированности\n",
"#Если распределение классов примерно равно (например, 50%/50%), выборка считается сбалансированной, и аугментация данных не требуется.\n",
"#Если один из классов сильно доминирует (например, 90%/10%), выборка несбалансированная, и может потребоваться аугментация данных.\n",
"\n",
"#Данная сборка несбалансированная, и требуется аугментация данных.\n",
"\n",
"\n",
"#10. Выполнить приращение данных методами выборки с избытком (oversampling) и выборки с недостатком (undersampling). Должны быть представлены примеры реализации обоих методов для выборок набора данных. \n",
"\n",
"#10.1\n",
"#Аугментация данных методом оверсемплинга¶\n",
"#Этот метод увеличивает количество примеров меньшинства.\n",
"\n",
"from imblearn.over_sampling import ADASYN\n",
"\n",
"# Создание экземпляра ADASYN\n",
"ada = ADASYN()\n",
"\n",
"# Применение ADASYN\n",
"X_resampled, y_resampled = ada.fit_resample(train_data.drop(columns=['Outcome']), train_data['Outcome'])\n",
"\n",
"# Создание нового DataFrame\n",
"df_train_adasyn = pd.DataFrame(X_resampled)\n",
"df_train_adasyn['Outcome'] = y_resampled # Добавление целевой переменной\n",
"\n",
"# Вывод информации о новой выборке\n",
"print(\"Обучающая выборка после оверсемплинга: \", df_train_adasyn.shape)\n",
"print(df_train_adasyn['Outcome'].value_counts())\n",
"Outcome_counts = df_train_adasyn['Outcome'].value_counts()\n",
"plt.figure(figsize=(2, 2))\n",
"plt.pie(Outcome_counts, labels=Outcome_counts.index, autopct='%1.1f%%', startangle=90)\n",
"plt.title('Распределение классов Outcome в обучающей выборке после оверсемплинга')\n",
"plt.show()\n",
"\n",
"\n",
"#10.2\n",
"#Аугментация данных методом андерсемплинга\n",
"#Этот метод помогает сбалансировать выборку, уменьшая количество экземпляров класса большинства, чтобы привести его в соответствие с классом меньшинства\n",
"\n",
"from imblearn.under_sampling import RandomUnderSampler\n",
"\n",
"rus = RandomUnderSampler()\n",
"\n",
"# Применение RandomUnderSampler\n",
"X_resampled, y_resampled = rus.fit_resample(train_data.drop(columns=['Outcome']), train_data['Outcome'])\n",
"\n",
"# Создание нового DataFrame\n",
"df_train_undersampled = pd.DataFrame(X_resampled)\n",
"df_train_undersampled['Outcome'] = y_resampled # Добавление целевой переменной\n",
"\n",
"# Вывод информации о новой выборке\n",
"print(\"Обучающая выборка после андерсемплинга: \", df_train_undersampled.shape)\n",
"print(df_train_undersampled['Outcome'].value_counts())\n",
"\n",
"# Визуализация распределения классов\n",
"Outcome_counts = df_train_undersampled['Outcome'].value_counts()\n",
"plt.figure(figsize=(2, 2))\n",
"plt.pie(Outcome_counts, labels=Outcome_counts.index, autopct='%1.1f%%', startangle=90)\n",
"plt.title('Распределение классов Outcome в обучающей выборке после андерсемплинга')\n",
"plt.show()"
]
}
],
"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": "3.12.8"
}
},
"nbformat": 4,
"nbformat_minor": 2
}