SooNooClose/AIM-PIbd-31-Kryukov-A-I
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

небольшая правка коментов

Browse Source
This commit is contained in:
Алексей Крюков 2024-11-30 13:59:12 +04:00
parent 39f65612d8
commit 998128cb0f
1 changed files with 9 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
Lab_3/lab3.ipynb
View File

@ -86,7 +86,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 47, "execution_count": null,
"metadata": {}, "metadata": {},
"outputs": [ "outputs": [
{ {
@ -107,8 +107,8 @@
], ],
"source": [ "source": [
"# 3. Проверим датасет на наличие пропусков и удалим строки с недостающими данными\n", "# 3. Проверим датасет на наличие пропусков и удалим строки с недостающими данными\n",
"print(data.isnull().sum()) # Суммируем пропуски по каждому столбцу\n", "print(data.isnull().sum()) \n",
"data.dropna(inplace=True) # Удаляем строки с пропусками\n", "data.dropna(inplace=True) \n",
"print(\"Данные после очистки:\", data.shape)\n" "print(\"Данные после очистки:\", data.shape)\n"
] ]
}, },
@ -124,7 +124,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 48, "execution_count": null,
"metadata": {}, "metadata": {},
"outputs": [ "outputs": [
{ {
@ -144,8 +144,8 @@
} }
], ],
"source": [ "source": [
"print(data.isnull().sum()) # Вывод количества пропусков\n", "print(data.isnull().sum()) \n",
"data.dropna(inplace=True) # Удаление строк с пропущенными значениями\n", "data.dropna(inplace=True) \n",
"print(f\"Количество строк после удаления пропусков: {data.shape[0]}\")\n" "print(f\"Количество строк после удаления пропусков: {data.shape[0]}\")\n"
] ]
}, },
@ -215,7 +215,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 51, "execution_count": null,
"metadata": {}, "metadata": {},
"outputs": [ "outputs": [
{ {
@ -232,7 +232,7 @@
"from sklearn.model_selection import train_test_split\n", "from sklearn.model_selection import train_test_split\n",
"\n", "\n",
"X = data[scaled_columns]\n", "X = data[scaled_columns]\n",
"y = data['Close'] # Заменить на целевую переменную, если другая\n", "y = data['Close']\n",
"X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.4, random_state=42)\n", "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.4, random_state=42)\n",
"X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n", "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",
"\n", "\n",
@ -573,7 +573,6 @@
" dataframe_name=\"stock_data\",\n", " dataframe_name=\"stock_data\",\n",
" dataframe=data,\n", " dataframe=data,\n",
" index=\"datetime\", \n", " index=\"datetime\", \n",
" # Убираем time_index, так как datetime уже используется как индекс\n",
" logical_types={\n", " logical_types={\n",
" \"open\": ww.logical_types.Double, \n", " \"open\": ww.logical_types.Double, \n",
" \"high\": ww.logical_types.Double,\n", " \"high\": ww.logical_types.Double,\n",
@ -594,13 +593,10 @@
"# Выводим имена столбцов в feature_matrix, чтобы убедиться, какие признаки были сгенерированы\n", "# Выводим имена столбцов в feature_matrix, чтобы убедиться, какие признаки были сгенерированы\n",
"print(\"Generated feature columns:\", feature_matrix.columns)\n", "print(\"Generated feature columns:\", feature_matrix.columns)\n",
"\n", "\n",
"# Объединяем новые признаки с исходными данными, добавив суффиксы для дублирующихся столбцов\n",
"data_featuretools = data.join(feature_matrix, lsuffix='_orig', rsuffix='_feature')\n", "data_featuretools = data.join(feature_matrix, lsuffix='_orig', rsuffix='_feature')\n",
"\n", "\n",
"# Масштабирование данных\n",
"scaler = StandardScaler()\n", "scaler = StandardScaler()\n",
"\n", "\n",
"# Используем правильные имена столбцов для масштабирования\n",
"scaled_columns = ['open_orig', 'high_orig', 'low_orig', 'close_orig', 'volume_orig']\n", "scaled_columns = ['open_orig', 'high_orig', 'low_orig', 'close_orig', 'volume_orig']\n",
"\n", "\n",
"data_featuretools[scaled_columns] = scaler.fit_transform(data_featuretools[scaled_columns])\n", "data_featuretools[scaled_columns] = scaler.fit_transform(data_featuretools[scaled_columns])\n",
@ -609,7 +605,7 @@
"\n", "\n",
"feature_columns = [col for col in feature_matrix.columns if 'feature' in col]\n", "feature_columns = [col for col in feature_matrix.columns if 'feature' in col]\n",
"X = data_featuretools[scaled_columns + feature_columns]\n", "X = data_featuretools[scaled_columns + feature_columns]\n",
"y = data_featuretools['close_orig'] # Целевая переменная\n", "y = data_featuretools['close_orig'] \n",
"\n", "\n",
"X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.4, random_state=42)\n", "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.4, random_state=42)\n",
"X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n", "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",