pr1
This commit is contained in:
1 parent
1170d4ce15
commit
3749cac06d
1 file changed
+302
+302
@@ -0,0 +1,302 @@
|
|||||||
|
{
|
||||||
|
"cells": [
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 1,
|
||||||
|
"id": "3b2c0591",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 1000x600 with 1 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {},
|
||||||
|
"output_type": "display_data"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"import numpy as np\n",
|
||||||
|
"import matplotlib.pyplot as plt\n",
|
||||||
|
"from sklearn.model_selection import train_test_split\n",
|
||||||
|
"from sklearn.linear_model import LinearRegression\n",
|
||||||
|
"from sklearn.preprocessing import PolynomialFeatures\n",
|
||||||
|
"from sklearn.pipeline import Pipeline\n",
|
||||||
|
"from sklearn.metrics import mean_squared_error, r2_score\n",
|
||||||
|
"\n",
|
||||||
|
"np.random.seed(42)\n",
|
||||||
|
"\n",
|
||||||
|
"m = 100\n",
|
||||||
|
"X = np.sort(np.random.randn(m, 1) * 2 - 1, axis=0)\n",
|
||||||
|
"\n",
|
||||||
|
"y = 3 * X**2 + 2 + X + 1 + np.random.randn(m, 1) * 0.3\n",
|
||||||
|
"\n",
|
||||||
|
"\n",
|
||||||
|
"plt.figure(figsize=(10, 6))\n",
|
||||||
|
"plt.scatter(X, y, color='blue', label=\"Исходные данные (с шумом)\")\n",
|
||||||
|
"plt.xlabel(\"Признак X\")\n",
|
||||||
|
"plt.ylabel(\"Целевая переменная y\")\n",
|
||||||
|
"plt.title(\"Синтетические данные для регрессии\")\n",
|
||||||
|
"plt.legend()\n",
|
||||||
|
"plt.grid(True)\n",
|
||||||
|
"plt.show()"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 5,
|
||||||
|
"id": "0f51a261",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"name": "stdout",
|
||||||
|
"output_type": "stream",
|
||||||
|
"text": [
|
||||||
|
"--- Линейная регрессия (степень 1) ---\n",
|
||||||
|
"MSE на обучении: 147.7139\n",
|
||||||
|
"MSE на тесте: 304.1505\n",
|
||||||
|
"R² на обучении: 0.4254\n",
|
||||||
|
"R² на тесте: 0.5454\n"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 1000x600 with 1 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {},
|
||||||
|
"output_type": "display_data"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
|
||||||
|
"\n",
|
||||||
|
"model = LinearRegression()\n",
|
||||||
|
"model.fit(X_train, y_train)\n",
|
||||||
|
"\n",
|
||||||
|
"y_train_pred = model.predict(X_train)\n",
|
||||||
|
"y_test_pred = model.predict(X_test)\n",
|
||||||
|
"\n",
|
||||||
|
"print(\"--- Линейная регрессия (степень 1) ---\")\n",
|
||||||
|
"print(f\"MSE на обучении: {mean_squared_error(y_train, y_train_pred):.4f}\")\n",
|
||||||
|
"print(f\"MSE на тесте: {mean_squared_error(y_test, y_test_pred):.4f}\")\n",
|
||||||
|
"print(f\"R² на обучении: {r2_score(y_train, y_train_pred):.4f}\")\n",
|
||||||
|
"print(f\"R² на тесте: {r2_score(y_test, y_test_pred):.4f}\")\n",
|
||||||
|
"\n",
|
||||||
|
"# Визуализация\n",
|
||||||
|
"plt.figure(figsize=(10, 6))\n",
|
||||||
|
"plt.scatter(X_train, y_train, color='green', alpha=0.5, label='Обучающая выборка')\n",
|
||||||
|
"plt.scatter(X_test, y_test, color='red', alpha=0.5, label='Тестовая выборка')\n",
|
||||||
|
"X_plot = np.linspace(-1, 1, 100).reshape(-1, 1)\n",
|
||||||
|
"plt.plot(X_plot, model.predict(X_plot), color='black', linewidth=3, label='Модель (прямая)')\n",
|
||||||
|
"plt.legend()\n",
|
||||||
|
"plt.title('Линейная регрессия (недообучение)')\n",
|
||||||
|
"plt.grid(True)\n",
|
||||||
|
"plt.show()"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 6,
|
||||||
|
"id": "cf1d3404",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"name": "stdout",
|
||||||
|
"output_type": "stream",
|
||||||
|
"text": [
|
||||||
|
"\n",
|
||||||
|
"--- Полиномиальная регрессия (степень 2) ---\n",
|
||||||
|
"MSE на обучении: 0.0802\n",
|
||||||
|
"MSE на тесте: 0.0770\n"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 1000x600 with 1 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {},
|
||||||
|
"output_type": "display_data"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"poly_model_2 = Pipeline([\n",
|
||||||
|
" ('poly', PolynomialFeatures(degree=2, include_bias=False)),\n",
|
||||||
|
" ('linear', LinearRegression())\n",
|
||||||
|
"])\n",
|
||||||
|
"poly_model_2.fit(X_train, y_train)\n",
|
||||||
|
"\n",
|
||||||
|
"print(\"\\n--- Полиномиальная регрессия (степень 2) ---\")\n",
|
||||||
|
"print(f\"MSE на обучении: {mean_squared_error(y_train, poly_model_2.predict(X_train)):.4f}\")\n",
|
||||||
|
"print(f\"MSE на тесте: {mean_squared_error(y_test, poly_model_2.predict(X_test)):.4f}\")\n",
|
||||||
|
"\n",
|
||||||
|
"plt.figure(figsize=(10, 6))\n",
|
||||||
|
"plt.scatter(X_train, y_train, color='green', alpha=0.5, label='Обучающая выборка')\n",
|
||||||
|
"plt.scatter(X_test, y_test, color='red', alpha=0.5, label='Тестовая выборка')\n",
|
||||||
|
"plt.plot(X_plot, poly_model_2.predict(X_plot), color='blue', linewidth=3, label='Модель (степень 2)')\n",
|
||||||
|
"plt.legend()\n",
|
||||||
|
"plt.title('Полиномиальная регрессия степени 2 (оптимальная)')\n",
|
||||||
|
"plt.grid(True)\n",
|
||||||
|
"plt.show()"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 7,
|
||||||
|
"id": "961c989a",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"name": "stdout",
|
||||||
|
"output_type": "stream",
|
||||||
|
"text": [
|
||||||
|
"\n",
|
||||||
|
"--- Полиномиальная регрессия (степень 15) ---\n",
|
||||||
|
"MSE на обучении: 22.3938\n",
|
||||||
|
"MSE на тесте: 69942577.6653\n"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 1000x600 with 1 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {},
|
||||||
|
"output_type": "display_data"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"poly_model_15 = Pipeline([\n",
|
||||||
|
" ('poly', PolynomialFeatures(degree=15, include_bias=False)),\n",
|
||||||
|
" ('linear', LinearRegression())\n",
|
||||||
|
"])\n",
|
||||||
|
"poly_model_15.fit(X_train, y_train)\n",
|
||||||
|
"\n",
|
||||||
|
"print(\"\\n--- Полиномиальная регрессия (степень 15) ---\")\n",
|
||||||
|
"print(f\"MSE на обучении: {mean_squared_error(y_train, poly_model_15.predict(X_train)):.4f}\")\n",
|
||||||
|
"print(f\"MSE на тесте: {mean_squared_error(y_test, poly_model_15.predict(X_test)):.4f}\")\n",
|
||||||
|
"\n",
|
||||||
|
"plt.figure(figsize=(10, 6))\n",
|
||||||
|
"plt.scatter(X_train, y_train, color='green', alpha=0.5, label='Обучающая выборка')\n",
|
||||||
|
"plt.scatter(X_test, y_test, color='red', alpha=0.5, label='Тестовая выборка')\n",
|
||||||
|
"plt.plot(X_plot, poly_model_15.predict(X_plot), color='purple', linewidth=3, label='Модель (степень 15)')\n",
|
||||||
|
"plt.legend()\n",
|
||||||
|
"plt.title('Полиномиальная регрессия степени 15 (СИЛЬНОЕ ПЕРЕОБУЧЕНИЕ)')\n",
|
||||||
|
"plt.grid(True)\n",
|
||||||
|
"plt.show()"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 8,
|
||||||
|
"id": "d028527f",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"name": "stdout",
|
||||||
|
"output_type": "stream",
|
||||||
|
"text": [
|
||||||
|
"Коэффициенты степени 2:\n",
|
||||||
|
"[[0.97310248 2.99355626]]\n",
|
||||||
|
"\n",
|
||||||
|
"Коэффициенты степени 15:\n",
|
||||||
|
"[[-5.08863913e-08 1.06061359e-07 -4.55482834e-07 1.07372888e-06\n",
|
||||||
|
" -3.73711546e-06 8.88098583e-06 -2.70999308e-05 5.83574397e-05\n",
|
||||||
|
" -1.50506332e-04 2.31843565e-04 -4.11713300e-04 -1.16818039e-04\n",
|
||||||
|
" 6.13366087e-05 2.34846121e-05 2.05042712e-06]]\n"
|
||||||
|
]
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"print(\"Коэффициенты степени 2:\")\n",
|
||||||
|
"print(poly_model_2.named_steps['linear'].coef_)\n",
|
||||||
|
"\n",
|
||||||
|
"print(\"\\nКоэффициенты степени 15:\")\n",
|
||||||
|
"print(poly_model_15.named_steps['linear'].coef_)"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 9,
|
||||||
|
"id": "677a2172",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [
|
||||||
|
{
|
||||||
|
"data": {
|
||||||
|
"image/png": "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 truncated
|
||||||
|
"text/plain": [
|
||||||
|
"<Figure size 1000x600 with 1 Axes>"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
"metadata": {},
|
||||||
|
"output_type": "display_data"
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"name": "stdout",
|
||||||
|
"output_type": "stream",
|
||||||
|
"text": [
|
||||||
|
"Оптимальная степень (минимальная тестовая ошибка): 2\n"
|
||||||
|
]
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"source": [
|
||||||
|
"degrees = range(1, 21)\n",
|
||||||
|
"train_errors = []\n",
|
||||||
|
"test_errors = []\n",
|
||||||
|
"\n",
|
||||||
|
"for d in degrees:\n",
|
||||||
|
" model = Pipeline([\n",
|
||||||
|
" ('poly', PolynomialFeatures(degree=d, include_bias=False)),\n",
|
||||||
|
" ('linear', LinearRegression())\n",
|
||||||
|
" ])\n",
|
||||||
|
" model.fit(X_train, y_train)\n",
|
||||||
|
" train_errors.append(mean_squared_error(y_train, model.predict(X_train)))\n",
|
||||||
|
" test_errors.append(mean_squared_error(y_test, model.predict(X_test)))\n",
|
||||||
|
"\n",
|
||||||
|
"plt.figure(figsize=(10, 6))\n",
|
||||||
|
"plt.plot(degrees, train_errors, marker='o', label='Ошибка на обучении (MSE)')\n",
|
||||||
|
"plt.plot(degrees, test_errors, marker='s', label='Ошибка на тесте (MSE)')\n",
|
||||||
|
"plt.xlabel('Степень полинома')\n",
|
||||||
|
"plt.ylabel('Среднеквадратичная ошибка (MSE)')\n",
|
||||||
|
"plt.title('Зависимость ошибки от сложности модели')\n",
|
||||||
|
"plt.legend()\n",
|
||||||
|
"plt.grid(True)\n",
|
||||||
|
"plt.show()\n",
|
||||||
|
"\n",
|
||||||
|
"# Момент начала переобучения\n",
|
||||||
|
"best_degree = degrees[np.argmin(test_errors)]\n",
|
||||||
|
"print(f\"Оптимальная степень (минимальная тестовая ошибка): {best_degree}\")"
|
||||||
|
]
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"metadata": {
|
||||||
|
"kernelspec": {
|
||||||
|
"display_name": "ai-lessons",
|
||||||
|
"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": "3.12.15"
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"nbformat": 4,
|
||||||
|
"nbformat_minor": 5
|
||||||
|
}
|
||||||
Reference in new issue
Block a user