{"id":1655,"date":"2024-11-23T07:47:00","date_gmt":"2024-11-22T23:47:00","guid":{"rendered":"https:\/\/blog.laoyulaoyu.top\/?p=1655"},"modified":"2024-10-29T14:48:58","modified_gmt":"2024-10-29T06:48:58","slug":"svr-%e8%82%a1%e4%bb%b7%e9%a2%84%e6%b5%8b%e5%ae%9e%e7%94%a8%e6%8c%87%e5%8d%97-%ef%bc%88%e9%99%84%e4%bb%a3%e7%a0%81%ef%bc%89","status":"publish","type":"post","link":"https:\/\/www.laoyulaoyu.com\/index.php\/2024\/11\/23\/svr-%e8%82%a1%e4%bb%b7%e9%a2%84%e6%b5%8b%e5%ae%9e%e7%94%a8%e6%8c%87%e5%8d%97-%ef%bc%88%e9%99%84%e4%bb%a3%e7%a0%81%ef%bc%89\/","title":{"rendered":"SVR \u80a1\u4ef7\u9884\u6d4b\u5b9e\u7528\u6307\u5357 \uff08\u9644\u4ee3\u7801\uff09"},"content":{"rendered":"\n

\u4f5c\u8005\uff1a\u8001\u4f59\u635e\u9c7c<\/a><\/p>\n\n\n\n

\u539f\u521b\u4e0d\u6613\uff0c\u8f6c\u8f7d\u8bf7\u6807\u660e\u51fa\u5904\u53ca\u539f\u4f5c\u8005\u3002<\/mark><\/strong><\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n
\n
\u5199\u5728\u524d\u9762\u7684\u8bdd\uff1a<\/strong>\u672c\u6587\u4ecb\u7ecd\u4e86\u5982\u4f55\u5229\u7528\u652f\u6301\u5411\u91cf\u56de\u5f52\uff08SVR\uff09\u7b97\u6cd5\u6765\u9884\u6d4b\u80a1\u7968\u4ef7\u683c<\/mark>\uff0c\u7279\u522b\u662f\u9488\u5bf9\u65e5\u5185\u4ea4\u6613\u7684\u5e94\u7528\uff0c\u5e76\u901a\u8fc7\u5b9e\u9645\u4ee3\u7801\u793a\u4f8b\u8be6\u7ec6\u8bb2\u89e3\u5982\u4f55\u4f7f\u7528 Python \u4e2d\u7684\u00a0yfinance\u00a0\u548c\u00a0scikit-learn\u00a0\u5e93\u6765\u83b7\u53d6\u6570\u636e\u3001\u51c6\u5907\u6570\u636e\u96c6\u3001\u8bad\u7ec3 SVR \u6a21\u578b\u4ee5\u53ca\u8fdb\u884c\u4ef7\u683c\u9884\u6d4b\u548c\u7ed3\u679c\u53ef\u89c6\u5316\u3002<\/pre>\n<\/blockquote>\n\n\n\n
\u4e00\u3001\u4ec0\u4e48\u662f SVR\uff1f<\/strong><\/h3>\n\n\n\n
\u5728\u6211\u4eec\u5f00\u59cb\u6df1\u5165\u63a2\u8ba8\u4e4b\u524d\uff0c\u8ba9\u6211\u4eec\u5148\u5feb\u901f\u4e86\u89e3\u4e00\u4e0b\u4ec0\u4e48\u662f SVR\u3002<\/p>\n\n\n\n
SVR\uff08\u652f\u6301\u5411\u91cf\u56de\u5f52 Support Vector Regression\uff09<\/strong>\uff0c\u5b83\u662f\u652f\u6301\u5411\u91cf\u673a\uff08SVM\uff09\u5728\u56de\u5f52\u5206\u6790\u4e0a\u7684\u5e94\u7528\u3002\u8fd9\u662f\u4e00\u79cd\u53ef\u4ee5\u9884\u6d4b\u8fde\u7eed\u503c\u7684\u673a\u5668\u76d1\u7763\u5b66\u4e60\u7b97\u6cd5<\/strong>\uff0c\u4e3b\u8981\u7528\u4e8e\u56de\u5f52\u5206\u6790\u548c\u9884\u6d4b\u6570\u503c\u578b\u6570\u636e\u7684\u8f93\u51fa\u3002\u4e0e\u4f20\u7edf\u7684\u7ebf\u6027\u56de\u5f52\u4e0d\u540c\uff0cSVR \u7684\u76ee\u6807\u662f\u5728\u6307\u5b9a\u7684\u5bb9\u5dee\uff08\u03b5\uff09\u8303\u56f4\u5185\u627e\u5230\u6700\u4f73\u62df\u5408\u3002\u8fd9\u4f7f\u5b83\u6210\u4e3a\u9884\u6d4b\u6ce2\u52a8\u80a1\u4ef7\u7684\u5b8c\u7f8e\u5de5\u5177\uff01SVR\u5728\u5904\u7406\u975e\u7ebf\u6027\u95ee\u9898\u548c\u9ad8\u7ef4\u6570\u636e\u65f6\u8868\u73b0\u51fa\u8272\uff0c\u5e7f\u6cdb\u5e94\u7528\u4e8e\u91d1\u878d\u5206\u6790\u3001\u65f6\u95f4\u5e8f\u5217\u9884\u6d4b\u3001\u80a1\u7968\u4ef7\u683c\u8d8b\u52bf\u9884\u6d4b\u4ee5\u53ca\u751f\u7269\u4fe1\u606f\u5b66\u7b49\u9886\u57df\u3002<\/p>\n\n\n\n
\u4e8c\u3001\u6570\u636e\u51c6\u5907\u5de5\u4f5c<\/strong><\/h3>\n\n\n\n
2.1 \u83b7\u53d6\u5f53\u65e5\u6570\u636e<\/strong><\/p>\n\n\n\n
\u5bf9\u4e8e\u65e5\u5185\u4ea4\u6613\uff0c\u6211\u4eec\u611f\u5174\u8da3\u7684\u662f\u76d8\u4e2d\u6570\u636e–\u6bcf\u5206\u949f\u7684\u80a1\u7968\u4ef7\u683c\u3002\u6211\u4eec\u5c06\u4f7f\u7528 yfinance \u5e93\u6765\u83b7\u53d6\u8fd9\u4e9b\u6570\u636e\u3002<\/p>\n\n\n\n
# Install necessary libraries\n!pip install yfinance\n!pip install scikit-learn\n\n# Import libraries\nimport yfinance as yf\nimport numpy as np\nimport pandas as pd\nfrom sklearn.svm import SVR\nfrom sklearn.preprocessing import StandardScaler\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import mean_squared_error\nimport matplotlib.pyplot as plt\nimport matplotlib.dates as mdates\n\n# Fetch intraday data for a stock, e.g., Apple Inc. (AAPL)\nticker = yf.Ticker(\"AAPL\")\nstock_data = ticker.history(period=\"1d\", interval=\"1m\")\nstock_data = stock_data[['Close']]  # Use only the 'Close' price for simplicity\nstock_data.reset_index(inplace=True)<\/code><\/code><\/pre>\n\n\n\n
2.2 \u51c6\u5907\u6570\u636e<\/strong><\/p>\n\n\n\n
\u63a5\u4e0b\u6765\u4e3a SVR \u6a21\u578b\u51c6\u5907\u6570\u636e\uff0c\u8fd9\u5305\u62ec\u521b\u5efa\u4ea4\u6613\u65e5\u7684\u5c0f\u65f6\u548c\u5206\u949f\u7b49\u7279\u5f81\u3002<\/p>\n\n\n\n
# Prepare the data\nstock_data['Hour'] = stock_data['Datetime'].dt.hour\nstock_data['Minute'] = stock_data['Datetime'].dt.minute\n\nX = stock_data[['Hour', 'Minute']]\ny = stock_data['Close']\n\n# Split into training and testing sets\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n\n# Scale the features\nscaler = StandardScaler()\nX_train_scaled = scaler.fit_transform(X_train)\nX_test_scaled = scaler.transform(X_test)<\/code><\/code><\/pre>\n\n\n\n
\u4e09\u3001\u8bad\u7ec3 SVR \u6a21\u578b<\/strong><\/h3>\n\n\n\n
\u6211\u4eec\u5c06\u4f7f\u7528 scikit-learn \u5e93\u8bad\u7ec3 SVR \u6a21\u578b\u3002<\/p>\n\n\n\n
# Initialize and train the SVR model\nsvr_model = SVR(kernel='rbf', C=100, gamma=0.1, epsilon=0.1)\nsvr_model.fit(X_train_scaled, y_train)<\/code><\/code><\/pre>\n\n\n\n
Scikit-learn<\/strong> \u662f\u673a\u5668\u5b66\u4e60\u9886\u57df\u4e2d\u6700\u5b8c\u6574\u3001\u6700\u5177\u5f71\u54cd\u529b\u7684\u5f00\u6e90Python\u673a\u5668\u5b66\u4e60\u5e93\u4e4b\u4e00\uff0c\u662f\u6570\u636e\u5206\u6790\u5e08\u9996\u9009\u7684\u673a\u5668\u5b66\u4e60\u5de5\u5177\u5305\uff0c\u5b83\u63d0\u4f9b\u4e86\u4e30\u5bcc\u800c\u5f3a\u5927\u7684\u5de5\u5177\uff0c\u7528\u4e8e\u6570\u636e\u6316\u6398\u548c\u6570\u636e\u5206\u6790\u3002\u65e0\u8bba\u662f\u521d\u5b66\u8005\u8fd8\u662f\u7ecf\u9a8c\u4e30\u5bcc\u7684\u6570\u636e\u79d1\u5b66\u5bb6\uff0c\u90fd\u53ef\u4ee5\u901a\u8fc7Scikit-learn\u6765\u9ad8\u6548\u5730\u8fdb\u884c\u6570\u636e\u6316\u6398\u548c\u6570\u636e\u5206\u6790\u5de5\u4f5c\u3002<\/p>\n\n\n\n
\u56db\u3001\u8fdb\u884c\u9884\u6d4b\u548c\u7ed8\u5236\u7ed3\u679c\u56fe<\/strong><\/h3>\n\n\n\n
\u6a21\u578b\u8bad\u7ec3\u5b8c\u6210\u540e\uff0c\u6211\u4eec\u5c31\u53ef\u4ee5\u8fdb\u884c\u9884\u6d4b\u5e76\u5c06\u7ed3\u679c\u53ef\u89c6\u5316\u3002\u56fe\u8868\u4e0a\u7684\u7ea2\u70b9\u4ee3\u8868\u6211\u4eec\u6a21\u578b\u7684\u9884\u6d4b\u7ed3\u679c\u3002<\/p>\n\n\n\n
# Predict on the test set\ny_pred = svr_model.predict(X_test_scaled)\n\n# Evaluate the model\nmse = mean_squared_error(y_test, y_pred)\nprint(f'Mean Squared Error: {mse}')\n\n# Plot the results\nplt.figure(figsize=(14, 7))\nplt.plot(stock_data['Datetime'], stock_data['Close'], label='Actual Prices')\nplt.scatter(stock_data['Datetime'].iloc[X_test.index], y_pred, color='red', label='Predicted Prices')\nplt.xlabel('Time')\nplt.ylabel('Stock Price')\nplt.title('SVR Model - Intraday Actual vs Predicted Prices')\nplt.legend()\n\n# Format x-axis to show hours and minutes\nplt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%H:%M'))\nplt.gca().xaxis.set_major_locator(mdates.HourLocator(interval=1))\nplt.gcf().autofmt_xdate()  # Rotation\n\nplt.tight_layout()\nplt.show()<\/code><\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
SVR \u56fe\u8868\u4e2d\u7684\u7ea2\u70b9\u901a\u5e38\u4ee3\u8868 SVR \u6a21\u578b\u751f\u6210\u7684\u9884\u6d4b\u503c\u3002\u5728\u5f53\u65e5\u4ea4\u6613\u548c\u80a1\u7968\u4ef7\u683c\u9884\u6d4b\u4e2d\uff0c\u8fd9\u4e9b\u7ea2\u70b9\u8868\u793a SVR \u6a21\u578b\u5728\u4ea4\u6613\u65e5\u7279\u5b9a\u65f6\u95f4\u70b9\u9884\u6d4b\u7684\u4ef7\u683c\u3002\u5177\u4f53\u6765\u8bf4\uff1a<\/p>\n\n\n\n