{"id":25860,"date":"2025-04-19T14:01:06","date_gmt":"2025-04-19T06:01:06","guid":{"rendered":"https:\/\/www.wsisp.com\/helps\/25860.html"},"modified":"2025-04-19T14:01:06","modified_gmt":"2025-04-19T06:01:06","slug":"%e6%9c%ba%e5%99%a8%e5%ad%a6%e4%b9%a0%e6%a8%a1%e5%9e%8b%e9%83%a8%e7%bd%b2%ef%bc%9a%e4%bd%bf%e7%94%a8flask-%e5%ba%93%e7%9a%84-python-web-%e6%a1%86%e6%9e%b6%e5%b0%86xgboost%e6%a8%a1%e5%9e%8b%e9%83%a8","status":"publish","type":"post","link":"https:\/\/www.wsisp.com\/helps\/25860.html","title":{"rendered":"\u673a\u5668\u5b66\u4e60\u6a21\u578b\u90e8\u7f72\uff1a\u4f7f\u7528Flask \u5e93\u7684 Python Web \u6846\u67b6\u5c06XGBoost\u6a21\u578b\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\uff08\u7b80\u5355\u5c0f\u6a21\u578b\uff09\u4ece\u6a21\u578b\u8bad\u7ec3\u5230\u90e8\u7f72\u518d\u5230\u672c\u5730\u8c03\u7528"},"content":{"rendered":"

1.XGBoost\u6a21\u578b\u8bad\u7ec3\u9884\u6d4b\u98ce\u901f\u6a21\u578b 2.\u4fdd\u5b58\u8bad\u7ec3\u597d\u7684\u6a21\u578b 3.\u670d\u52a1\u5668\u7aef\u90e8\u7f72\u53ca\u8fd0\u884c 4.\u672c\u5730PyCharm\u8c03\u7528 5.\u4e00\u4e9b\u62a5\u9519\u95ee\u9898\u53ca\u6ce8\u610f<\/p>\n

\u4e00\u3001XGBoost\u6a21\u578b\u8bad\u7ec3\u9884\u6d4b\u98ce\u901f\u6a21\u578b<\/h2>\n

\u8fd9\u91cc\u4e0d\u89e3\u91ca\u4ee3\u7801，\u540e\u9762\u5176\u4ed6\u6587\u7ae0\u8bf4\u660e，\u4f7f\u7528\u4e86\u8d1d\u53f6\u65af\u4f18\u5316\u5bfb\u627e\u6700\u4f73\u53c2\u6570\u7ec4\u5408\u3002<\/p>\n

import<\/span> pandas as<\/span> pd
\nimport<\/span> numpy as<\/span> np
\nfrom<\/span> sklearn.<\/span>model_selection import<\/span> train_test_split
\nfrom<\/span> xgboost import<\/span> XGBRegressor # \u5bfc\u5165 XGBoost \u56de\u5f52\u6a21\u578b<\/span>
\nfrom<\/span> sklearn.<\/span>metrics import<\/span> mean_squared_error,<\/span> mean_absolute_error,<\/span> r2_score
\nimport<\/span> matplotlib.<\/span>pyplot as<\/span> plt
\nfrom<\/span> sklearn.<\/span>preprocessing import<\/span> MinMaxScaler
\nimport<\/span> optuna
\nimport<\/span> joblib
\nimport<\/span> pickle<\/p>\n

# 1. \u52a0\u8f7d\u6570\u636e<\/span>
\ndata1 =<\/span> pd.<\/span>read_csv(<\/span>r'ID01.csv'<\/span>)<\/span><\/p>\n

# 2. \u786e\u4fdd date_time \u5217\u662f datetime \u7c7b\u578b<\/span>
\ndata1[<\/span>'date_time'<\/span>]<\/span> =<\/span> pd.<\/span>to_datetime(<\/span>data1[<\/span>'date_time'<\/span>]<\/span>,<\/span> format<\/span>=<\/span>'%Y\/%m\/%d %H:%M'<\/span>)<\/span><\/p>\n

# 3. \u6dfb\u52a0\u6ede\u540e\u7279\u5f81<\/span>
\nlags =<\/span> 9<\/span> # \u6ede\u540e\u6b65\u957f\u4e3a 12<\/span>
\nfor<\/span> lag in<\/span> range<\/span>(<\/span>1<\/span>,<\/span> lags +<\/span> 1<\/span>)<\/span>:<\/span>
\n data1[<\/span>f'wind_obs_lag_<\/span>{<\/span>lag}<\/span><\/span>'<\/span><\/span>]<\/span> =<\/span> data1[<\/span>'wind_obs'<\/span>]<\/span>.<\/span>shift(<\/span>lag)<\/span><\/p>\n

lags1 =<\/span> 9<\/span> # \u6ede\u540e\u6b65\u957f\u4e3a 12<\/span>
\nfor<\/span> lag in<\/span> range<\/span>(<\/span>1<\/span>,<\/span> lags1 +<\/span> 1<\/span>)<\/span>:<\/span>
\n data1[<\/span>f'ec_lag_<\/span>{<\/span>lag}<\/span><\/span>'<\/span><\/span>]<\/span> =<\/span> data1[<\/span>'ec'<\/span>]<\/span>.<\/span>shift(<\/span>lag)<\/span>
\n# 4. \u6dfb\u52a0\u6ed1\u52a8\u7a97\u53e3\u7279\u5f81<\/span>
\nwindow_size =<\/span> 12<\/span> # \u6ed1\u52a8\u7a97\u53e3\u5927\u5c0f<\/span>
\ndata1[<\/span>'wind_obs_rolling_mean'<\/span>]<\/span> =<\/span> data1[<\/span>'wind_obs'<\/span>]<\/span>.<\/span>shift(<\/span>1<\/span>)<\/span>.<\/span>rolling(<\/span>window=<\/span>window_size,<\/span> min_periods=<\/span>1<\/span>)<\/span>.<\/span>mean(<\/span>)<\/span> # \u6ed1\u52a8\u5e73\u5747\u503c<\/span>
\ndata1[<\/span>'wind_obs_rolling_std'<\/span>]<\/span> =<\/span> data1[<\/span>'wind_obs'<\/span>]<\/span>.<\/span>shift(<\/span>1<\/span>)<\/span>.<\/span>rolling(<\/span>window=<\/span>window_size,<\/span> min_periods=<\/span>1<\/span>)<\/span>.<\/span>std(<\/span>)<\/span> # \u6ed1\u52a8\u6807\u51c6\u5dee<\/span>
\ndata1[<\/span>'wind_obs_rolling_max'<\/span>]<\/span> =<\/span> data1[<\/span>'wind_obs'<\/span>]<\/span>.<\/span>shift(<\/span>1<\/span>)<\/span>.<\/span>rolling(<\/span>window=<\/span>window_size,<\/span> min_periods=<\/span>1<\/span>)<\/span>.<\/span>max<\/span>(<\/span>)<\/span> # \u6700\u5927\u503c<\/span>
\ndata1[<\/span>'wind_obs_rolling_min'<\/span>]<\/span> =<\/span> data1[<\/span>'wind_obs'<\/span>]<\/span>.<\/span>shift(<\/span>1<\/span>)<\/span>.<\/span>rolling(<\/span>window=<\/span>window_size,<\/span> min_periods=<\/span>1<\/span>)<\/span>.<\/span>min<\/span>(<\/span>)<\/span> # \u6700\u5c0f\u503c<\/span>
\nwindow_size1 =<\/span> 12<\/span> # \u6ed1\u52a8\u7a97\u53e3\u5927\u5c0f<\/span>
\ndata1[<\/span>'ec_mean'<\/span>]<\/span> =<\/span> data1[<\/span>'ec'<\/span>]<\/span>.<\/span>shift(<\/span>1<\/span>)<\/span>.<\/span>rolling(<\/span>window=<\/span>window_size1,<\/span> min_periods=<\/span>1<\/span>)<\/span>.<\/span>mean(<\/span>)<\/span> # \u6ed1\u52a8\u5e73\u5747\u503c<\/span>
\ndata1[<\/span>'ec_std'<\/span>]<\/span> =<\/span> data1[<\/span>'ec'<\/span>]<\/span>.<\/span>shift(<\/span>1<\/span>)<\/span>.<\/span>rolling(<\/span>window=<\/span>window_size1,<\/span> min_periods=<\/span>1<\/span>)<\/span>.<\/span>std(<\/span>)<\/span> # \u6ed1\u52a8\u6807\u51c6\u5dee<\/span>
\ndata1[<\/span>'ec_max'<\/span>]<\/span> =<\/span> data1[<\/span>'ec'<\/span>]<\/span>.<\/span>shift(<\/span>1<\/span>)<\/span>.<\/span>rolling(<\/span>window=<\/span>window_size1,<\/span> min_periods=<\/span>1<\/span>)<\/span>.<\/span>max<\/span>(<\/span>)<\/span> # \u6700\u5927\u503c<\/span>
\ndata1[<\/span>'ec_min'<\/span>]<\/span> =<\/span> data1[<\/span>'ec'<\/span>]<\/span>.<\/span>shift(<\/span>1<\/span>)<\/span>.<\/span>rolling(<\/span>window=<\/span>window_size1,<\/span> min_periods=<\/span>1<\/span>)<\/span>.<\/span>min<\/span>(<\/span>)<\/span> # \u6700\u5c0f\u503c<\/span>
\n# 5. \u5b9a\u4e49\u65f6\u95f4\u8303\u56f4\u5e76\u7b5b\u9009\u6570\u636e<\/span>
\nstart_time =<\/span> '2023-06-01 00:00:00'<\/span>
\nend_time =<\/span> '2024-06-30 18:00:00'<\/span>
\ndata1 =<\/span> data1[<\/span>(<\/span>data1[<\/span>'date_time'<\/span>]<\/span> >=<\/span> start_time)<\/span> &<\/span> (<\/span>data1[<\/span>'date_time'<\/span>]<\/span> <=<\/span> end_time)<\/span>]<\/span><\/p>\n

# 6. \u6309\u65f6\u95f4\u987a\u5e8f\u5212\u5206\u6570\u636e\u96c6<\/span>
\ntrain_start =<\/span> '2023-06-01 00:00:00'<\/span>
\ntrain_end =<\/span> '2024-05-31 18:00:00'<\/span>
\ntest_start =<\/span> '2024-06-01 00:00:00'<\/span>
\ntest_end =<\/span> '2024-06-30 18:00:00'<\/span><\/p>\n

train_data =<\/span> data1[<\/span>(<\/span>data1[<\/span>'date_time'<\/span>]<\/span> >=<\/span> train_start)<\/span> &<\/span> (<\/span>data1[<\/span>'date_time'<\/span>]<\/span> <=<\/span> train_end)<\/span>]<\/span>
\ntest_data =<\/span> data1[<\/span>(<\/span>data1[<\/span>'date_time'<\/span>]<\/span> >=<\/span> test_start)<\/span> &<\/span> (<\/span>data1[<\/span>'date_time'<\/span>]<\/span> <=<\/span> test_end)<\/span>]<\/span><\/p>\n

# 7. \u6570\u636e\u9884\u5904\u7406<\/span>
\n# \u63d0\u53d6\u65f6\u95f4\u7279\u5f81<\/span>
\nfor<\/span> data in<\/span> [<\/span>train_data,<\/span> test_data]<\/span>:<\/span>
\n data.<\/span>loc[<\/span>:<\/span>,<\/span> 'hour'<\/span>]<\/span> =<\/span> data[<\/span>'date_time'<\/span>]<\/span>.<\/span>dt.<\/span>hour
\n data.<\/span>loc[<\/span>:<\/span>,<\/span> 'day'<\/span>]<\/span> =<\/span> data[<\/span>'date_time'<\/span>]<\/span>.<\/span>dt.<\/span>day
\n data.<\/span>loc[<\/span>:<\/span>,<\/span> 'month'<\/span>]<\/span> =<\/span> data[<\/span>'date_time'<\/span>]<\/span>.<\/span>dt.<\/span>month
\n data.<\/span>loc[<\/span>:<\/span>,<\/span> 'year'<\/span>]<\/span> =<\/span> data[<\/span>'date_time'<\/span>]<\/span>.<\/span>dt.<\/span>year<\/p>\n

# \u63d0\u53d6\u5b63\u8282\u7279\u5f81<\/span>
\n def<\/span> get_season<\/span>(<\/span>month)<\/span>:<\/span>
\n if<\/span> month in<\/span> [<\/span>3<\/span>,<\/span> 4<\/span>,<\/span> 5<\/span>]<\/span>:<\/span>
\n return<\/span> 'Spring'<\/span>
\n elif<\/span> month in<\/span> [<\/span>6<\/span>,<\/span> 7<\/span>,<\/span> 8<\/span>]<\/span>:<\/span>
\n return<\/span> 'Summer'<\/span>
\n elif<\/span> month in<\/span> [<\/span>9<\/span>,<\/span> 10<\/span>,<\/span> 11<\/span>]<\/span>:<\/span>
\n return<\/span> 'Autumn'<\/span>
\n else<\/span>:<\/span>
\n return<\/span> 'Winter'<\/span><\/p>\n

data.<\/span>loc[<\/span>:<\/span>,<\/span> 'season'<\/span>]<\/span> =<\/span> data[<\/span>'month'<\/span>]<\/span>.<\/span>apply<\/span>(<\/span>get_season)<\/span>
\n season_mapping =<\/span> {<\/span>'Spring'<\/span>:<\/span> 1<\/span>,<\/span> 'Summer'<\/span>:<\/span> 2<\/span>,<\/span> 'Autumn'<\/span>:<\/span> 3<\/span>,<\/span> 'Winter'<\/span>:<\/span> 4<\/span>}<\/span>
\n data.<\/span>loc[<\/span>:<\/span>,<\/span> 'season'<\/span>]<\/span> =<\/span> data[<\/span>'season'<\/span>]<\/span>.<\/span>map<\/span>(<\/span>season_mapping)<\/span><\/p>\n

# 8. \u63d0\u53d6\u7279\u5f81\u548c\u76ee\u6807\u5217<\/span>
\nX_train =<\/span> train_data[<\/span>[<\/span>'ec'<\/span>,<\/span> 'hour'<\/span>,<\/span> 'day'<\/span>,<\/span> 'month'<\/span>,<\/span> 'year'<\/span>,<\/span> 'season'<\/span>]<\/span> +<\/span> [<\/span>f'wind_obs_lag_<\/span>{<\/span>lag}<\/span><\/span>'<\/span><\/span> for<\/span> lag in<\/span> range<\/span>(<\/span>1<\/span>,<\/span> lags +<\/span> 1<\/span>)<\/span>]<\/span> +<\/span> [<\/span>f'ec_lag_<\/span>{<\/span>lag}<\/span><\/span>'<\/span><\/span> for<\/span> lag in<\/span> range<\/span>(<\/span>1<\/span>,<\/span> lags +<\/span> 1<\/span>)<\/span>]<\/span> +<\/span> [<\/span>'wind_obs_rolling_mean'<\/span>,<\/span> 'wind_obs_rolling_std'<\/span>,<\/span> 'wind_obs_rolling_max'<\/span>,<\/span> 'wind_obs_rolling_min'<\/span>]<\/span> +<\/span>[<\/span>'ec_mean'<\/span>,<\/span> 'ec_std'<\/span>,<\/span> 'ec_max'<\/span>,<\/span> 'ec_min'<\/span>]<\/span>]<\/span>
\ny_train =<\/span> train_data[<\/span>'wind_obs'<\/span>]<\/span><\/p>\n

X_test =<\/span> test_data[<\/span>[<\/span>'ec'<\/span>,<\/span> 'hour'<\/span>,<\/span> 'day'<\/span>,<\/span> 'month'<\/span>,<\/span> 'year'<\/span>,<\/span> 'season'<\/span>]<\/span> +<\/span> [<\/span>f'wind_obs_lag_<\/span>{<\/span>lag}<\/span><\/span>'<\/span><\/span> for<\/span> lag in<\/span> range<\/span>(<\/span>1<\/span>,<\/span> lags +<\/span> 1<\/span>)<\/span>]<\/span> +<\/span> [<\/span>f'ec_lag_<\/span>{<\/span>lag}<\/span><\/span>'<\/span><\/span> for<\/span> lag in<\/span> range<\/span>(<\/span>1<\/span>,<\/span> lags +<\/span> 1<\/span>)<\/span>]<\/span> +<\/span> [<\/span>'wind_obs_rolling_mean'<\/span>,<\/span> 'wind_obs_rolling_std'<\/span>,<\/span> 'wind_obs_rolling_max'<\/span>,<\/span> 'wind_obs_rolling_min'<\/span>]<\/span> +<\/span>[<\/span>'ec_mean'<\/span>,<\/span> 'ec_std'<\/span>,<\/span> 'ec_max'<\/span>,<\/span> 'ec_min'<\/span>]<\/span>]<\/span>
\ny_test =<\/span> test_data[<\/span>'wind_obs'<\/span>]<\/span><\/p>\n

# 9. \u6570\u636e\u5f52\u4e00\u5316（\u6700\u5927\u6700\u5c0f\u5f52\u4e00\u5316）<\/span>
\nscaler =<\/span> MinMaxScaler(<\/span>)<\/span>
\ny_scaler =<\/span> MinMaxScaler(<\/span>)<\/span><\/p>\n

X_train_scaled =<\/span> scaler.<\/span>fit_transform(<\/span>X_train)<\/span>
\nX_test_scaled =<\/span> scaler.<\/span>transform(<\/span>X_test)<\/span><\/p>\n

y_train_scaled =<\/span> y_scaler.<\/span>fit_transform(<\/span>y_train.<\/span>values.<\/span>reshape(<\/span>–<\/span>1<\/span>,<\/span> 1<\/span>)<\/span>)<\/span>.<\/span>flatten(<\/span>)<\/span>
\ny_test_scaled =<\/span> y_scaler.<\/span>transform(<\/span>y_test.<\/span>values.<\/span>reshape(<\/span>–<\/span>1<\/span>,<\/span> 1<\/span>)<\/span>)<\/span>.<\/span>flatten(<\/span>)<\/span><\/p>\n

# 8. \u5b9a\u4e49 Optuna \u76ee\u6807\u51fd\u6570<\/span>
\ndef<\/span> objective<\/span>(<\/span>trial)<\/span>:<\/span>
\n # \u5b9a\u4e49\u8d85\u53c2\u6570\u641c\u7d22\u8303\u56f4<\/span>
\n n_estimators =<\/span> trial.<\/span>suggest_int(<\/span>'n_estimators'<\/span>,<\/span> 50<\/span>,<\/span> 700<\/span>)<\/span>
\n learning_rate =<\/span> trial.<\/span>suggest_float(<\/span>'learning_rate'<\/span>,<\/span> 0.01<\/span>,<\/span> 0.2<\/span>)<\/span>
\n max_depth =<\/span> trial.<\/span>suggest_int(<\/span>'max_depth'<\/span>,<\/span> 3<\/span>,<\/span> 30<\/span>)<\/span>
\n min_child_weight =<\/span> trial.<\/span>suggest_int(<\/span>'min_child_weight'<\/span>,<\/span> 1<\/span>,<\/span> 10<\/span>)<\/span>
\n subsample =<\/span> trial.<\/span>suggest_float(<\/span>'subsample'<\/span>,<\/span> 0.5<\/span>,<\/span> 1.0<\/span>)<\/span>
\n colsample_bytree =<\/span> trial.<\/span>suggest_float(<\/span>'colsample_bytree'<\/span>,<\/span> 0.5<\/span>,<\/span> 1.0<\/span>)<\/span><\/p>\n

# \u6784\u5efa XGBoost \u56de\u5f52\u6a21\u578b<\/span>
\n xgb =<\/span> XGBRegressor(<\/span>
\n n_estimators=<\/span>n_estimators,<\/span>
\n learning_rate=<\/span>learning_rate,<\/span>
\n max_depth=<\/span>max_depth,<\/span>
\n min_child_weight=<\/span>min_child_weight,<\/span>
\n subsample=<\/span>subsample,<\/span>
\n colsample_bytree=<\/span>colsample_bytree,<\/span>
\n random_state=<\/span>42<\/span>,<\/span>
\n n_jobs=<\/span>–<\/span>1<\/span>
\n )<\/span><\/p>\n

# \u8bad\u7ec3\u548c\u8bc4\u4f30\u6a21\u578b<\/span>
\n xgb.<\/span>fit(<\/span>X_train_scaled,<\/span> y_train_scaled)<\/span>
\n y_pred =<\/span> xgb.<\/span>predict(<\/span>X_test_scaled)<\/span>
\n rmse =<\/span> np.<\/span>sqrt(<\/span>mean_squared_error(<\/span>y_test_scaled,<\/span> y_pred)<\/span>)<\/span><\/p>\n

return<\/span> rmse<\/p>\n

# 10. \u8fd0\u884c Optuna \u4f18\u5316<\/span>
\nstudy =<\/span> optuna.<\/span>create_study(<\/span>direction=<\/span>'minimize'<\/span>)<\/span>
\nstudy.<\/span>optimize(<\/span>objective,<\/span> n_trials=<\/span>500<\/span>)<\/span><\/p>\n

# \u83b7\u53d6\u6700\u4f73\u8d85\u53c2\u6570<\/span>
\nbest_params =<\/span> study.<\/span>best_params
\nprint<\/span>(<\/span>"\u6700\u4f73\u8d85\u53c2\u6570："<\/span>,<\/span> best_params)<\/span><\/p>\n

# 11. \u4f7f\u7528\u6700\u4f73\u8d85\u53c2\u6570\u8bad\u7ec3\u6a21\u578b<\/span>
\nbest_model =<\/span> XGBRegressor(<\/span>
\n n_estimators=<\/span>best_params[<\/span>'n_estimators'<\/span>]<\/span>,<\/span>
\n learning_rate=<\/span>best_params[<\/span>'learning_rate'<\/span>]<\/span>,<\/span>
\n max_depth=<\/span>best_params[<\/span>'max_depth'<\/span>]<\/span>,<\/span>
\n min_child_weight=<\/span>best_params[<\/span>'min_child_weight'<\/span>]<\/span>,<\/span>
\n subsample=<\/span>best_params[<\/span>'subsample'<\/span>]<\/span>,<\/span>
\n colsample_bytree=<\/span>best_params[<\/span>'colsample_bytree'<\/span>]<\/span>,<\/span>
\n random_state=<\/span>42<\/span>
\n #n_jobs=-1<\/span>
\n)<\/span>
\nbest_model.<\/span>fit(<\/span>X_train_scaled,<\/span> y_train_scaled)<\/span><\/p>\n

# 12. \u9884\u6d4b<\/span>
\ny_train_pred =<\/span> best_model.<\/span>predict(<\/span>X_train_scaled)<\/span>
\ny_test_pred =<\/span> best_model.<\/span>predict(<\/span>X_test_scaled)<\/span><\/p>\n

# \u53cd\u5f52\u4e00\u5316<\/span>
\ny_train_pred =<\/span> y_scaler.<\/span>inverse_transform(<\/span>y_train_pred.<\/span>reshape(<\/span>–<\/span>1<\/span>,<\/span> 1<\/span>)<\/span>)<\/span>.<\/span>flatten(<\/span>)<\/span>
\ny_test_pred =<\/span> y_scaler.<\/span>inverse_transform(<\/span>y_test_pred.<\/span>reshape(<\/span>–<\/span>1<\/span>,<\/span> 1<\/span>)<\/span>)<\/span>.<\/span>flatten(<\/span>)<\/span><\/p>\n

# 13. \u8bc4\u4f30\u6307\u6807<\/span>
\ndef<\/span> evaluate<\/span>(<\/span>y_true,<\/span> y_pred,<\/span> dataset_name)<\/span>:<\/span>
\n rmse =<\/span> np.<\/span>sqrt(<\/span>mean_squared_error(<\/span>y_true,<\/span> y_pred)<\/span>)<\/span>
\n mae =<\/span> mean_absolute_error(<\/span>y_true,<\/span> y_pred)<\/span>
\n mbe =<\/span> np.<\/span>mean(<\/span>y_pred –<\/span> y_true)<\/span> # Mean Bias Error<\/span>
\n r2 =<\/span> r2_score(<\/span>y_true,<\/span> y_pred)<\/span>
\n print<\/span>(<\/span>f"<\/span>{<\/span>dataset_name}<\/span><\/span> RMSE: <\/span>{<\/span>rmse:<\/span>.4f<\/span>}<\/span><\/span>"<\/span><\/span>)<\/span>
\n print<\/span>(<\/span>f"<\/span>{<\/span>dataset_name}<\/span><\/span> MAE: <\/span>{<\/span>mae:<\/span>.4f<\/span>}<\/span><\/span>"<\/span><\/span>)<\/span>
\n print<\/span>(<\/span>f"<\/span>{<\/span>dataset_name}<\/span><\/span> MBE: <\/span>{<\/span>mbe:<\/span>.4f<\/span>}<\/span><\/span>"<\/span><\/span>)<\/span>
\n print<\/span>(<\/span>f"<\/span>{<\/span>dataset_name}<\/span><\/span> R2 Score: <\/span>{<\/span>r2:<\/span>.4f<\/span>}<\/span><\/span>"<\/span><\/span>)<\/span><\/p>\n

# \u8f93\u51fa\u8bad\u7ec3\u96c6\u548c\u6d4b\u8bd5\u96c6\u7684\u8bc4\u4ef7\u6307\u6807<\/span>
\nevaluate(<\/span>y_train,<\/span> y_train_pred,<\/span> "Training Set"<\/span>)<\/span>
\nevaluate(<\/span>y_test,<\/span> y_test_pred,<\/span> "Test Set"<\/span>)<\/span><\/p>\n

# 14. \u53ef\u89c6\u5316<\/span>
\n# \u7ed8\u5236\u8bad\u7ec3\u96c6\u548c\u6d4b\u8bd5\u96c6\u7684\u89c2\u6d4b\u503c\u4e0e\u9884\u6d4b\u503c\u6563\u70b9\u56fe<\/span>
\nplt.<\/span>figure(<\/span>figsize=<\/span>(<\/span>12<\/span>,<\/span> 6<\/span>)<\/span>)<\/span>
\nplt.<\/span>scatter(<\/span>y_test,<\/span> y_test_pred,<\/span> alpha=<\/span>0.5<\/span>,<\/span> label=<\/span>'Test Set'<\/span>)<\/span>
\nplt.<\/span>scatter(<\/span>y_train,<\/span> y_train_pred,<\/span> alpha=<\/span>0.5<\/span>,<\/span> label=<\/span>'Training Set'<\/span>)<\/span>
\nplt.<\/span>plot(<\/span>[<\/span>min<\/span>(<\/span>y_test)<\/span>,<\/span> max<\/span>(<\/span>y_test)<\/span>]<\/span>,<\/span> [<\/span>min<\/span>(<\/span>y_test)<\/span>,<\/span> max<\/span>(<\/span>y_test)<\/span>]<\/span>,<\/span> color=<\/span>'red'<\/span>,<\/span> linestyle=<\/span>'–'<\/span>,<\/span> label=<\/span>'Ideal Line'<\/span>)<\/span>
\nplt.<\/span>xlabel(<\/span>'Observed Wind Speed'<\/span>)<\/span>
\nplt.<\/span>ylabel(<\/span>'Predicted Wind Speed'<\/span>)<\/span>
\nplt.<\/span>title(<\/span>'Observed vs Predicted Wind Speed'<\/span>)<\/span>
\nplt.<\/span>legend(<\/span>)<\/span>
\nplt.<\/span>grid(<\/span>True<\/span>)<\/span>
\nplt.<\/span>show(<\/span>)<\/span><\/p>\n

# 15. \u53ef\u89c6\u5316\u9884\u6d4b\u7ed3\u679c<\/span>
\n# \u7ed8\u5236\u6d4b\u8bd5\u96c6\u7684\u5b9e\u9645\u503c\u548c\u9884\u6d4b\u503c\u66f2\u7ebf<\/span>
\nplt.<\/span>figure(<\/span>figsize=<\/span>(<\/span>12<\/span>,<\/span> 6<\/span>)<\/span>)<\/span>
\nplt.<\/span>plot(<\/span>y_test,<\/span> label=<\/span>'Observed'<\/span>,<\/span> color=<\/span>'blue'<\/span>)<\/span>
\nplt.<\/span>plot(<\/span>y_test_pred,<\/span> label=<\/span>'Predicted'<\/span>,<\/span> color=<\/span>'red'<\/span>)<\/span>
\nplt.<\/span>legend(<\/span>)<\/span>
\nplt.<\/span>title(<\/span>'Observed vs Predicted Wind Speed (Test Set)'<\/span>)<\/span>
\nplt.<\/span>xlabel(<\/span>'Time'<\/span>)<\/span>
\nplt.<\/span>ylabel(<\/span>'Wind Speed'<\/span>)<\/span>
\nplt.<\/span>grid(<\/span>True<\/span>)<\/span>
\nplt.<\/span>show(<\/span>)<\/span><\/p>\n

# \u7ed8\u5236\u8bad\u7ec3\u96c6\u7684\u5b9e\u9645\u503c\u548c\u9884\u6d4b\u503c\u66f2\u7ebf<\/span>
\nplt.<\/span>figure(<\/span>figsize=<\/span>(<\/span>12<\/span>,<\/span> 6<\/span>)<\/span>)<\/span>
\nplt.<\/span>plot(<\/span>y_train,<\/span> label=<\/span>'Observed'<\/span>,<\/span> color=<\/span>'blue'<\/span>)<\/span>
\nplt.<\/span>plot(<\/span>y_train_pred,<\/span> label=<\/span>'Predicted'<\/span>,<\/span> color=<\/span>'red'<\/span>)<\/span>
\nplt.<\/span>legend(<\/span>)<\/span>
\nplt.<\/span>title(<\/span>'Observed vs Predicted Wind Speed (Training Set)'<\/span>)<\/span>
\nplt.<\/span>xlabel(<\/span>'Time'<\/span>)<\/span>
\nplt.<\/span>ylabel(<\/span>'Wind Speed'<\/span>)<\/span>
\nplt.<\/span>grid(<\/span>True<\/span>)<\/span>
\nplt.<\/span>show(<\/span>)<\/span><\/p>\n

# 16. \u8f93\u51fa\u7279\u5f81\u91cd\u8981\u6027<\/span>
\n# \u83b7\u53d6\u7279\u5f81\u540d\u79f0<\/span>
\nfeature_names =<\/span> X_train.<\/span>columns<\/p>\n

# \u83b7\u53d6\u7279\u5f81\u91cd\u8981\u6027<\/span>
\nfeature_importances =<\/span> best_model.<\/span>feature_importances_<\/p>\n

# \u5c06\u7279\u5f81\u540d\u79f0\u4e0e\u91cd\u8981\u6027\u5bf9\u5e94\u8d77\u6765<\/span>
\nfeature_importance =<\/span> pd.<\/span>DataFrame(<\/span>{<\/span>'Feature'<\/span>:<\/span> feature_names,<\/span> 'Importance'<\/span>:<\/span> feature_importances}<\/span>)<\/span><\/p>\n

# \u6309\u91cd\u8981\u6027\u6392\u5e8f，\u4ee5\u67e5\u770b\u6700\u91cd\u8981\u7684\u7279\u5f81<\/span>
\nfeature_importance =<\/span> feature_importance.<\/span>sort_values(<\/span>by=<\/span>'Importance'<\/span>,<\/span> ascending=<\/span>False<\/span>)<\/span><\/p>\n

# \u8f93\u51fa\u7279\u5f81\u91cd\u8981\u6027<\/span>
\nprint<\/span>(<\/span>"\u7279\u5f81\u91cd\u8981\u6027："<\/span>)<\/span>
\nprint<\/span>(<\/span>feature_importance)<\/span><\/p>\n

# \u53ef\u89c6\u5316\u7279\u5f81\u91cd\u8981\u6027<\/span>
\nplt.<\/span>figure(<\/span>figsize=<\/span>(<\/span>10<\/span>,<\/span> 6<\/span>)<\/span>)<\/span>
\nplt.<\/span>barh(<\/span>feature_importance[<\/span>'Feature'<\/span>]<\/span>,<\/span> feature_importance[<\/span>'Importance'<\/span>]<\/span>,<\/span> color=<\/span>'skyblue'<\/span>)<\/span>
\nplt.<\/span>xlabel(<\/span>'Importance'<\/span>)<\/span>
\nplt.<\/span>ylabel(<\/span>'Feature'<\/span>)<\/span>
\nplt.<\/span>title(<\/span>'Feature Importance'<\/span>)<\/span>
\nplt.<\/span>show(<\/span>)<\/span><\/p>\n

# 17. \u4fdd\u5b58\u6a21\u578b\u548c\u9884\u5904\u7406\u5bf9\u8c61<\/span>
\n# \u4f7f\u7528 joblib \u4fdd\u5b58\u6a21\u578b<\/span>
\njoblib.<\/span>dump(<\/span>best_model,<\/span> 'xgb_wind_speed_model.joblib'<\/span>)<\/span><\/p>\n

# \u4f7f\u7528 pickle \u4fdd\u5b58\u6a21\u578b（\u53e6\u4e00\u79cd\u65b9\u5f0f）<\/span>
\nwith<\/span> open<\/span>(<\/span>'xgb_wind_speed_model.pkl'<\/span>,<\/span> 'wb'<\/span>)<\/span> as<\/span> f:<\/span>
\n pickle.<\/span>dump(<\/span>best_model,<\/span> f)<\/span><\/p>\n

# \u4fdd\u5b58 scaler \u5bf9\u8c61（\u4ee5\u4fbf\u540e\u7eed\u5bf9\u65b0\u6570\u636e\u8fdb\u884c\u76f8\u540c\u7684\u9884\u5904\u7406）<\/span>
\njoblib.<\/span>dump(<\/span>scaler,<\/span> 'scaler.joblib'<\/span>)<\/span>
\njoblib.<\/span>dump(<\/span>y_scaler,<\/span> 'y_scaler.joblib'<\/span>)<\/span><\/p>\n

print<\/span>(<\/span>"\u6a21\u578b\u548c\u9884\u5904\u7406\u5bf9\u8c61\u5df2\u4fdd\u5b58"<\/span>)<\/span><\/p>\n

\u4e8c\u3001\u4fdd\u5b58\u8bad\u7ec3\u597d\u7684\u6a21\u578b<\/h2>\n

\u8fd9\u91cc\u6211\u76f4\u63a5\u628a\u4ee3\u7801\u5408\u4e00\u8d77，\u6240\u6709\u7ed3\u675f\u5728\u6587\u4ef6\u4e0b\u751f\u6210\u56db\u4e2a\u6587\u4ef6\u3002\u8fd9\u91cc\u540d\u79f0\u6211\u540e\u9762\u4fee\u6539\u4e86 \"\u5728\u8fd9\u91cc\u63d2\u5165\u56fe\u7247\u63cf\u8ff0\"<\/p>\n

\u4e09\u3001\u670d\u52a1\u5668\u7aef\u90e8\u7f72<\/h2>\n

\u8981\u5c06\u8bad\u7ec3\u597d\u7684XGBoost\u6a21\u578b\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\u5e76\u5728\u672c\u5730PyCharm\u8c03\u7528，\u4f60\u53ef\u4ee5\u4f7f\u7528Flask\u6216FastAPI\u521b\u5efa\u4e00\u4e2a\u7b80\u5355\u7684Web\u670d\u52a1\u3002<\/p>\n

1. \u521b\u5efa\u670d\u52a1\u5668\u7aef\u9879\u76ee\u7ed3\u6784（PyCharm\u4e0b\u65b0\u5efa\u6587\u4ef6\u5939model_api）<\/h3>\n

\u6309\u4e4b\u524d\u6587\u4ef6\u4fee\u6539\u540d\u5b57\u548c\u521b\u5efa\u65b0\u6587\u4ef6\u3002<\/p>\n

model_api\/
\n\u2502
\n\u251c\u2500\u2500 app.py # \u4e3b\u5e94\u7528\u6587\u4ef6<\/span>
\n\u251c\u2500\u2500 requirements.txt # \u4f9d\u8d56\u6587\u4ef6<\/span>
\n\u251c\u2500\u2500 models\/
\n\u2502 \u251c\u2500\u2500 xgb_model.joblib # \u4fdd\u5b58\u7684\u6a21\u578b\u6587\u4ef6<\/span>
\n\u2502 \u251c\u2500\u2500 scaler.joblib # \u7279\u5f81\u7f29\u653e\u5668<\/span>
\n\u2502 \u2514\u2500\u2500 y_scaler.joblib # \u76ee\u6807\u53d8\u91cf\u7f29\u653e\u5668<\/span><\/p>\n

2.\u521b\u5efaFlask API (app.py)<\/h3>\n

\u8fd9\u4e2a\u540e\u9762\u662f\u5728\u670d\u52a1\u5668\u4e2d\u8fd0\u884c，\u9047\u5230\u597d\u51e0\u6b21\u62a5\u9519，\u6700\u540e\u7edf\u4e00\u8bf4\u660e\u3002<\/p>\n

from<\/span> flask import<\/span> Flask,<\/span> request,<\/span> jsonify
\nimport<\/span> joblib
\nimport<\/span> numpy as<\/span> np
\nimport<\/span> pandas as<\/span> pd
\nfrom<\/span> datetime import<\/span> datetime
\nfrom<\/span> sklearn.<\/span>preprocessing import<\/span> MinMaxScaler<\/p>\n

app =<\/span> Flask(<\/span>__name__)<\/span><\/p>\n

# \u52a0\u8f7d\u6a21\u578b\u548c\u9884\u5904\u7406\u5bf9\u8c61<\/span>
\nmodel =<\/span> joblib.<\/span>load(<\/span>'models\/xgb_model.joblib'<\/span>)<\/span>
\nscaler =<\/span> joblib.<\/span>load(<\/span>'models\/scaler.joblib'<\/span>)<\/span>
\ny_scaler =<\/span> joblib.<\/span>load(<\/span>'models\/y_scaler.joblib'<\/span>)<\/span><\/p>\n

# \u5b9a\u4e49\u4e0e\u8bad\u7ec3\u65f6\u76f8\u540c\u7684\u53c2\u6570<\/span>
\nlags =<\/span> 9<\/span> # \u6ede\u540e\u6b65\u957f<\/span>
\nwindow_size =<\/span> 12<\/span> # \u6ed1\u52a8\u7a97\u53e3\u5927\u5c0f<\/span><\/p>\n

def<\/span> prepare_features<\/span>(<\/span>input_data)<\/span>:<\/span>
\n """\u51c6\u5907\u7279\u5f81，\u4e0e\u8bad\u7ec3\u65f6\u5b8c\u5168\u76f8\u540c\u7684\u9884\u5904\u7406\u6b65\u9aa4"""<\/span>
\n # \u5c06\u8f93\u5165\u6570\u636e\u8f6c\u6362\u4e3aDataFrame<\/span>
\n data =<\/span> pd.<\/span>DataFrame(<\/span>[<\/span>input_data]<\/span>)<\/span><\/p>\n

# \u89e3\u6790\u65e5\u671f\u65f6\u95f4<\/span>
\n data[<\/span>'date_time'<\/span>]<\/span> =<\/span> pd.<\/span>to_datetime(<\/span>data[<\/span>'date_time'<\/span>]<\/span>)<\/span>
\n data[<\/span>'hour'<\/span>]<\/span> =<\/span> data[<\/span>'date_time'<\/span>]<\/span>.<\/span>dt.<\/span>hour
\n data[<\/span>'day'<\/span>]<\/span> =<\/span> data[<\/span>'date_time'<\/span>]<\/span>.<\/span>dt.<\/span>day
\n data[<\/span>'month'<\/span>]<\/span> =<\/span> data[<\/span>'date_time'<\/span>]<\/span>.<\/span>dt.<\/span>month
\n data[<\/span>'year'<\/span>]<\/span> =<\/span> data[<\/span>'date_time'<\/span>]<\/span>.<\/span>dt.<\/span>year<\/p>\n

# \u5b63\u8282\u7279\u5f81<\/span>
\n def<\/span> get_season<\/span>(<\/span>month)<\/span>:<\/span>
\n if<\/span> month in<\/span> [<\/span>3<\/span>,<\/span> 4<\/span>,<\/span> 5<\/span>]<\/span>:<\/span>
\n return<\/span> 'Spring'<\/span>
\n elif<\/span> month in<\/span> [<\/span>6<\/span>,<\/span> 7<\/span>,<\/span> 8<\/span>]<\/span>:<\/span>
\n return<\/span> 'Summer'<\/span>
\n elif<\/span> month in<\/span> [<\/span>9<\/span>,<\/span> 10<\/span>,<\/span> 11<\/span>]<\/span>:<\/span>
\n return<\/span> 'Autumn'<\/span>
\n else<\/span>:<\/span>
\n return<\/span> 'Winter'<\/span><\/p>\n

data[<\/span>'season'<\/span>]<\/span> =<\/span> data[<\/span>'month'<\/span>]<\/span>.<\/span>apply<\/span>(<\/span>get_season)<\/span>
\n season_mapping =<\/span> {<\/span>'Spring'<\/span>:<\/span> 1<\/span>,<\/span> 'Summer'<\/span>:<\/span> 2<\/span>,<\/span> 'Autumn'<\/span>:<\/span> 3<\/span>,<\/span> 'Winter'<\/span>:<\/span> 4<\/span>}<\/span>
\n data[<\/span>'season'<\/span>]<\/span> =<\/span> data[<\/span>'season'<\/span>]<\/span>.<\/span>map<\/span>(<\/span>season_mapping)<\/span><\/p>\n

# \u6dfb\u52a0\u6ede\u540e\u7279\u5f81 – \u8fd9\u4e9b\u503c\u5e94\u8be5\u7531\u5ba2\u6237\u7aef\u63d0\u4f9b<\/span>
\n for<\/span> lag in<\/span> range<\/span>(<\/span>1<\/span>,<\/span> lags +<\/span> 1<\/span>)<\/span>:<\/span>
\n if<\/span> f'wind_obs_lag_<\/span>{<\/span>lag}<\/span><\/span>'<\/span><\/span> not<\/span> in<\/span> data.<\/span>columns:<\/span>
\n raise<\/span> ValueError(<\/span>f"Missing required feature: wind_obs_lag_<\/span>{<\/span>lag}<\/span><\/span>"<\/span><\/span>)<\/span>
\n if<\/span> f'ec_lag_<\/span>{<\/span>lag}<\/span><\/span>'<\/span><\/span> not<\/span> in<\/span> data.<\/span>columns:<\/span>
\n raise<\/span> ValueError(<\/span>f"Missing required feature: ec_lag_<\/span>{<\/span>lag}<\/span><\/span>"<\/span><\/span>)<\/span><\/p>\n

# \u6dfb\u52a0\u6ed1\u52a8\u7a97\u53e3\u7279\u5f81 – \u8fd9\u4e9b\u503c\u5e94\u8be5\u7531\u5ba2\u6237\u7aef\u8ba1\u7b97\u540e\u63d0\u4f9b<\/span>
\n window_features =<\/span> [<\/span>
\n 'wind_obs_rolling_mean'<\/span>,<\/span> 'wind_obs_rolling_std'<\/span>,<\/span>
\n 'wind_obs_rolling_max'<\/span>,<\/span> 'wind_obs_rolling_min'<\/span>,<\/span>
\n 'ec_mean'<\/span>,<\/span> 'ec_std'<\/span>,<\/span> 'ec_max'<\/span>,<\/span> 'ec_min'<\/span>
\n ]<\/span><\/p>\n

for<\/span> feature in<\/span> window_features:<\/span>
\n if<\/span> feature not<\/span> in<\/span> data.<\/span>columns:<\/span>
\n raise<\/span> ValueError(<\/span>f"Missing required window feature: <\/span>{<\/span>feature}<\/span><\/span>"<\/span><\/span>)<\/span><\/p>\n

# \u9009\u62e9\u6a21\u578b\u9700\u8981\u7684\u6240\u6709\u7279\u5f81\u5217<\/span>
\n features =<\/span> [<\/span>
\n 'ec'<\/span>,<\/span> 'hour'<\/span>,<\/span> 'day'<\/span>,<\/span> 'month'<\/span>,<\/span> 'year'<\/span>,<\/span> 'season'<\/span>
\n ]<\/span> +<\/span> [<\/span>f'wind_obs_lag_<\/span>{<\/span>lag}<\/span><\/span>'<\/span><\/span> for<\/span> lag in<\/span> range<\/span>(<\/span>1<\/span>,<\/span> lags +<\/span> 1<\/span>)<\/span>]<\/span> \\\\
\n +<\/span> [<\/span>f'ec_lag_<\/span>{<\/span>lag}<\/span><\/span>'<\/span><\/span> for<\/span> lag in<\/span> range<\/span>(<\/span>1<\/span>,<\/span> lags +<\/span> 1<\/span>)<\/span>]<\/span> \\\\
\n +<\/span> [<\/span>
\n 'wind_obs_rolling_mean'<\/span>,<\/span> 'wind_obs_rolling_std'<\/span>,<\/span>
\n 'wind_obs_rolling_max'<\/span>,<\/span> 'wind_obs_rolling_min'<\/span>,<\/span>
\n 'ec_mean'<\/span>,<\/span> 'ec_std'<\/span>,<\/span> 'ec_max'<\/span>,<\/span> 'ec_min'<\/span>
\n ]<\/span><\/p>\n

# \u786e\u4fdd\u6240\u6709\u7279\u5f81\u90fd\u5b58\u5728<\/span>
\n missing_features =<\/span> [<\/span>f for<\/span> f in<\/span> features if<\/span> f not<\/span> in<\/span> data.<\/span>columns]<\/span>
\n if<\/span> missing_features:<\/span>
\n raise<\/span> ValueError(<\/span>f"Missing features: <\/span>{<\/span>missing_features}<\/span><\/span>"<\/span><\/span>)<\/span><\/p>\n

# \u7f29\u653e\u7279\u5f81<\/span>
\n features_scaled =<\/span> scaler.<\/span>transform(<\/span>data[<\/span>features]<\/span>)<\/span><\/p>\n

return<\/span> features_scaled<\/p>\n

@app.<\/span>route<\/span>(<\/span>'\/predict'<\/span>,<\/span> methods=<\/span>[<\/span>'POST'<\/span>]<\/span>)<\/span>
\ndef<\/span> predict<\/span>(<\/span>)<\/span>:<\/span>
\n """\u9884\u6d4b\u7aef\u70b9"""<\/span>
\n try<\/span>:<\/span>
\n # \u83b7\u53d6JSON\u6570\u636e<\/span>
\n data =<\/span> request.<\/span>get_json(<\/span>)<\/span><\/p>\n

# \u51c6\u5907\u7279\u5f81<\/span>
\n features =<\/span> prepare_features(<\/span>data)<\/span><\/p>\n

# \u9884\u6d4b<\/span>
\n prediction_scaled =<\/span> model.<\/span>predict(<\/span>features)<\/span><\/p>\n

# \u53cd\u5f52\u4e00\u5316<\/span>
\n prediction =<\/span> y_scaler.<\/span>inverse_transform(<\/span>prediction_scaled.<\/span>reshape(<\/span>–<\/span>1<\/span>,<\/span> 1<\/span>)<\/span>)<\/span>[<\/span>0<\/span>]<\/span>[<\/span>0<\/span>]<\/span><\/p>\n

# \u8fd4\u56de\u7ed3\u679c<\/span>
\n return<\/span> jsonify(<\/span>{<\/span>
\n 'prediction'<\/span>:<\/span> float<\/span>(<\/span>prediction)<\/span>,<\/span> # \u8f6c\u6362\u4e3aPython\u539f\u751ffloat\u7c7b\u578b<\/span>
\n 'status'<\/span>:<\/span> 'success'<\/span>
\n }<\/span>)<\/span><\/p>\n

except<\/span> ValueError as<\/span> e:<\/span>
\n return<\/span> jsonify(<\/span>{<\/span>
\n 'error'<\/span>:<\/span> str<\/span>(<\/span>e)<\/span>,<\/span>
\n 'status'<\/span>:<\/span> 'error'<\/span>,<\/span>
\n 'message'<\/span>:<\/span> '\u7f3a\u5c11\u5fc5\u8981\u7684\u7279\u5f81\u5b57\u6bb5，\u8bf7\u786e\u4fdd\u63d0\u4f9b\u6240\u6709\u6ede\u540e\u7279\u5f81\u548c\u6ed1\u52a8\u7a97\u53e3\u7279\u5f81'<\/span>
\n }<\/span>)<\/span>,<\/span> 400<\/span><\/p>\n

except<\/span> Exception as<\/span> e:<\/span>
\n return<\/span> jsonify(<\/span>{<\/span>
\n 'error'<\/span>:<\/span> str<\/span>(<\/span>e)<\/span>,<\/span>
\n 'status'<\/span>:<\/span> 'error'<\/span>
\n }<\/span>)<\/span>,<\/span> 500<\/span><\/p>\n

if<\/span> __name__ ==<\/span> '__main__'<\/span>:<\/span>
\n app.<\/span>run(<\/span>host=<\/span>'0.0.0.0'<\/span>,<\/span> port=<\/span>5000<\/span>,<\/span> debug=<\/span>True<\/span>)<\/span><\/p>\n

3.\u521b\u5efa requirements.txt<\/h3>\n

\u8fd9\u4e2a\u6587\u672c\u7528\u9014\u662f\u5b89\u88c5\u9879\u76ee\u6240\u9700\u7684\u6240\u6709\u4f9d\u8d56\u5305，\u7f57\u5217\u4e86\u9879\u76ee\u6240\u9700\u7684\u6240\u6709 Python \u5305\u53ca\u5176\u7248\u672c\u53f7，\u8fd9\u4e2a\u62a5\u9519\u597d\u51e0\u6b21，\u4e3b\u8981\u539f\u56e0\u662f\u670d\u52a1\u5668\u7aef\u548c\u672c\u5730\u7aef\u5b89\u88c5\u5e93\u7248\u672c\u4e0d\u4e00\u81f4\u3002\u3002<\/p>\n

flask<\/span>==<\/span>3.1<\/span>.0
\njoblib<\/span>==<\/span>1.4<\/span>.2
\nnumpy<\/span>==<\/span>1.24<\/span>.3
\npandas<\/span>==<\/span>2.0<\/span>.3
\nscikit-learn==<\/span>1.3<\/span>.2
\nxgboost<\/span>==<\/span>2.1<\/span>.1<\/p>\n

4. \u90e8\u7f72\u5230\u670d\u52a1\u5668\u53ca\u8fd0\u884c<\/h3>\n

1.\u5c06\u6574\u4e2a\u9879\u76ee\u6587\u4ef6\u5939\u4e0a\u4f20\u5230\u670d\u52a1\u5668 \u628a\u4e4b\u524dmodel_api\u6587\u4ef6\u5939\u4e0a\u4f20\u5230\u670d\u52a1\u5668\u4e2d\u3002<\/p>\n

2.\u5728\u670d\u52a1\u5668\u4e0a\u521b\u5efa\u865a\u62df\u73af\u5883\u5e76\u5b89\u88c5\u4f9d\u8d56 \u8fd9\u5c31\u4f7f\u7528\u5230\u4e86requirements.txt\u8fd9\u4e2a\u6587\u4ef6\u4f5c\u7528，\u88c5\u8d77\u6765\u5f88\u5feb\u3002<\/p>\n

#1. conda\u521b\u5efa\u865a\u62df\u73af\u5883\/\u4e5f\u53ef\u4ee5python\u521b\u5efa\u73af\u5883，\u73af\u5883\u540dWY<\/span>
\n (<\/span>base)<\/span>$ conda create –name WY python<\/span>=<\/span>3.9<\/span><\/p>\n

# \u73af\u5883\u521b\u5efa\u4e2d\u8f93\u5165y\u5b89\u88c5\u51e0\u4e2a\u5305<\/span><\/p>\n

#2.\u6fc0\u6d3b\u73af\u5883<\/span>
\n(<\/span>base)<\/span>$ conda activate WY<\/p>\n

#3.\u5b89\u88c5\u4f9d\u8d56\u5305<\/span>
\npip install<\/span> -r requirements.txt<\/p>\n

5.\u8fd0\u884c\u670d\u52a1<\/h3>\n

python app.py<\/p>\n

\u5f53\u7136\u8fd9\u91cc\u62a5\u4e86\u51e0\u6b21\u9519\u8bef，\u540e\u9762\u8bf4\u3002\u8fd9\u91cc\u7ed9\u51fa\u8fd0\u884c\u6210\u529f\u622a\u56fe\u3002\u8fd8\u6709\u8b66\u544a \"\u5728\u8fd9\u91cc\u63d2\u5165\u56fe\u7247\u63cf\u8ff0\"<\/p>\n

\u56db\u3001\u672c\u5730PyCharm\u8c03\u7528<\/h2>\n

1. \u521b\u5efa\u5ba2\u6237\u7aef\u4ee3\u7801 (client.py)<\/h3>\n

\u8fd9\u91cc\u90fd\u662f\u6211\u7ed9\u7684\u503c，\u771f\u7684\u9884\u6d4b\u901a\u8fc7CSV\u8f93\u5165\u6570\u636e\u3002 \u4ee3\u7801\u6700\u540e\u6709\u4e2a\u670d\u52a1\u5668\u7ed9\u7684URL\u662f\u591a\u5c11<\/p>\n

import<\/span> requests
\nimport<\/span> json
\nfrom<\/span> datetime import<\/span> datetime<\/p>\n

# \u793a\u4f8b\u6570\u636e – \u5ba2\u6237\u7aef\u9700\u8981\u8ba1\u7b97\u6240\u6709\u6ede\u540e\u548c\u6ed1\u52a8\u7a97\u53e3\u7279\u5f81<\/span>
\nsample_data =<\/span> {<\/span>
\n "date_time"<\/span>:<\/span> "2024-06-01 12:00:00"<\/span>,<\/span>
\n "ec"<\/span>:<\/span> 5.3<\/span>,<\/span>
\n "hour"<\/span>:<\/span> 12<\/span>,<\/span>
\n "day"<\/span>:<\/span> 1<\/span>,<\/span>
\n "month"<\/span>:<\/span> 6<\/span>,<\/span>
\n "year"<\/span>:<\/span> 2024<\/span>,<\/span>
\n "season"<\/span>:<\/span> 2<\/span>,<\/span>
\n # \u6ede\u540e\u7279\u5f81<\/span>
\n "wind_obs_lag_1"<\/span>:<\/span> 4.8<\/span>,<\/span>
\n "wind_obs_lag_2"<\/span>:<\/span> 4.5<\/span>,<\/span>
\n "wind_obs_lag_3"<\/span>:<\/span> 4.2<\/span>,<\/span>
\n "wind_obs_lag_4"<\/span>:<\/span> 4.0<\/span>,<\/span>
\n "wind_obs_lag_5"<\/span>:<\/span> 3.9<\/span>,<\/span>
\n "wind_obs_lag_6"<\/span>:<\/span> 3.8<\/span>,<\/span>
\n "wind_obs_lag_7"<\/span>:<\/span> 3.7<\/span>,<\/span>
\n "wind_obs_lag_8"<\/span>:<\/span> 3.6<\/span>,<\/span>
\n "wind_obs_lag_9"<\/span>:<\/span> 3.5<\/span>,<\/span>
\n "ec_lag_1"<\/span>:<\/span> 5.1<\/span>,<\/span>
\n "ec_lag_2"<\/span>:<\/span> 5.0<\/span>,<\/span>
\n "ec_lag_3"<\/span>:<\/span> 4.9<\/span>,<\/span>
\n "ec_lag_4"<\/span>:<\/span> 4.8<\/span>,<\/span>
\n "ec_lag_5"<\/span>:<\/span> 4.7<\/span>,<\/span>
\n "ec_lag_6"<\/span>:<\/span> 4.6<\/span>,<\/span>
\n "ec_lag_7"<\/span>:<\/span> 4.5<\/span>,<\/span>
\n "ec_lag_8"<\/span>:<\/span> 4.4<\/span>,<\/span>
\n "ec_lag_9"<\/span>:<\/span> 4.3<\/span>,<\/span>
\n # \u6ed1\u52a8\u7a97\u53e3\u7279\u5f81<\/span>
\n "wind_obs_rolling_mean"<\/span>:<\/span> 4.2<\/span>,<\/span>
\n "wind_obs_rolling_std"<\/span>:<\/span> 0.5<\/span>,<\/span>
\n "wind_obs_rolling_max"<\/span>:<\/span> 5.0<\/span>,<\/span>
\n "wind_obs_rolling_min"<\/span>:<\/span> 3.5<\/span>,<\/span>
\n "ec_mean"<\/span>:<\/span> 4.8<\/span>,<\/span>
\n "ec_std"<\/span>:<\/span> 0.3<\/span>,<\/span>
\n "ec_max"<\/span>:<\/span> 5.2<\/span>,<\/span>
\n "ec_min"<\/span>:<\/span> 4.5<\/span>
\n}<\/span><\/p>\n

response =<\/span> requests.<\/span>post(<\/span>
\n "http:\/\/123.45.67.8:5000\/predict"<\/span>,<\/span>%<\/span>\u8fd9\u91cc\u770b\u4f60\u670d\u52a1\u5668\u7ed9\u7684URL\u662f\u591a\u5c11
\n headers=<\/span>{<\/span>"Content-Type"<\/span>:<\/span> "application\/json"<\/span>}<\/span>,<\/span>
\n data=<\/span>json.<\/span>dumps(<\/span>sample_data)<\/span>
\n)<\/span><\/p>\n

print<\/span>(<\/span>response.<\/span>json(<\/span>)<\/span>)<\/span><\/p>\n

2. \u5728PyCharm\u4e2d\u8fd0\u884c\u5ba2\u6237\u7aef<\/h3>\n

\u5728PyCharm\u4e2d\u521b\u5efa\u65b0\u9879\u76ee \u6dfb\u52a0client.py\u6587\u4ef6 \u786e\u4fdd\u5b89\u88c5\u4e86requests\u5e93 (PyCharm\u901a\u5e38\u4f1a\u81ea\u52a8\u63d0\u793a\u5b89\u88c5) \u8fd0\u884c\u5ba2\u6237\u7aef\u4ee3\u7801<\/p>\n

\u6210\u529f\u7684\u94fe\u63a5\u7684\u622a\u56fe\u6211\u8fd9\u8fb9\u5f00\u4e86VPN\u9700\u4fdd\u8bc1\u5728\u540c\u4e00\u7f51\u7edc\u3002 \u670d\u52a1\u5668\u7aef \"\u5728\u8fd9\u91cc\u63d2\u5165\u56fe\u7247\u63cf\u8ff0\" \u672c\u5730\u7aef \"\u5728\u8fd9\u91cc\u63d2\u5165\u56fe\u7247\u63cf\u8ff0\"<\/p>\n

\u4e94\u3001\u4e00\u4e9b\u62a5\u9519\u95ee\u9898\u548c\u6ce8\u610f<\/h2>\n

1.\u670d\u52a1\u5668\u7aef\u8fd0\u884capp.py\u62a5\u9519，\u5f04\u6e05\u695a\u7248\u672c\u53f7\u3002<\/h3>\n

\u8fd9\u90e8\u5206\u62a5\u9519\u95ee\u9898 （1）\u4e8e Flask \u548c Werkzeug \u7684\u7248\u672c\u4e0d\u517c\u5bb9\u5bfc\u81f4\u7684<\/p>\n

pip install<\/span> –upgrade Flask Werkzeug<\/p>\n

（2）\u6a21\u578b\u6587\u4ef6\u7684\u683c\u5f0f\u4e0e\u5f53\u524d XGBoost \u7248\u672c\u4e0d\u517c\u5bb9 \u6838\u5bf9\u4e24\u4e2a\u7248\u672c，\u91cd\u65b0\u5b89\u88c5\u5e93<\/p>\n

pip install<\/span> xgboost<\/span>==<\/span>2.1<\/span>.1<\/p>\n

（3）\u7ee7\u7eed\u62a5\u9519\u66f4\u65b0\u5176\u4ed6\u5e93<\/p>\n

pip install<\/span> –upgrade joblib numpy scikit-learn<\/p>\n

2.\u670d\u52a1\u5668\u7aef\u9884\u5904\u7406\u4e2d\u4ee3\u7801app.py\u548c\u672c\u5730\u7aefclient.py\u4e2d\u9884\u5904\u7406\u4ee3\u7801\u4e0e\u8bad\u7ec3XGB\u6a21\u578b\u9884\u5904\u7406\u4fdd\u6301\u4e00\u81f4\u3002<\/h3>\n

3.\u9632\u706b\u5899，URL\u4e0d\u901a，\u5c1d\u8bd5ping 123.45.67.8<\/h3>\n

4.\u7b80\u5355\u7684\u90e8\u7f72，\u540e\u9762\u53ef\u4ee5\u52a0\u5165API\u8ba4\u8bc1\u548c\u5176\u4ed6\u65b9\u6cd5\u90e8\u7f72<\/h3>\n","protected":false},"excerpt":{"rendered":"

\u6587\u7ae0\u6d4f\u89c8\u9605\u8bfb798\u6b21\uff0c\u70b9\u8d5e15\u6b21\uff0c\u6536\u85cf8\u6b21\u3002\u786e\u4fdd\u670d\u52a1\u5668\u9632\u706b\u5899\u5f00\u653e\u4e865000\u7aef\u53e3\uff08\u6216\u4f60\u4f7f\u7528\u7684\u5176\u4ed6\u7aef\u53e3\uff09\uff0c\u5bf9\u4e8e\u751f\u4ea7\u73af\u5883\uff0c\u8003\u8651\u4f7f\u7528Nginx\u4f5c\u4e3a\u53cd\u5411\u4ee3\u7406\uff0c\u6dfb\u52a0\u9002\u5f53\u7684\u9519\u8bef\u5904\u7406\u548c\u65e5\u5fd7\u8bb0\u5f55\uff0c\u6839\u636e\u4f60\u7684\u5b9e\u9645\u6a21\u578b\u8c03\u6574\u7279\u5f81\u9884\u5904\u7406\u6b65\u9aa4\uff0c\u8003\u8651\u6dfb\u52a0\u8f93\u5165\u6570\u636e\u9a8c\u8bc1\uff0c\u8fd9\u6837\u4f60\u5c31\u53ef\u4ee5\u5728\u672c\u5730PyCharm\u4e2d\u8f7b\u677e\u8c03\u7528\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\u7684\u6a21\u578b\u4e86\u3002_1\u67087\u65e5-\u673a\u5668\u5b66\u4e60\u7ebf\u4e0a\u5f00\u53d1\u4e0e\u6a21\u578b\u90e8\u7f72<\/p>\n","protected":false},"author":2,"featured_media":25856,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[153,2058,81,207],"topic":[],"class_list":["post-25860","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-server","tag-flask","tag-linuxubuntu","tag-python","tag-207"],"yoast_head":"\n\u673a\u5668\u5b66\u4e60\u6a21\u578b\u90e8\u7f72\uff1a\u4f7f\u7528Flask \u5e93\u7684 Python Web \u6846\u67b6\u5c06XGBoost\u6a21\u578b\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\uff08\u7b80\u5355\u5c0f\u6a21\u578b\uff09\u4ece\u6a21\u578b\u8bad\u7ec3\u5230\u90e8\u7f72\u518d\u5230\u672c\u5730\u8c03\u7528 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.wsisp.com\/helps\/25860.html\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"\u673a\u5668\u5b66\u4e60\u6a21\u578b\u90e8\u7f72\uff1a\u4f7f\u7528Flask \u5e93\u7684 Python Web \u6846\u67b6\u5c06XGBoost\u6a21\u578b\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\uff08\u7b80\u5355\u5c0f\u6a21\u578b\uff09\u4ece\u6a21\u578b\u8bad\u7ec3\u5230\u90e8\u7f72\u518d\u5230\u672c\u5730\u8c03\u7528 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"og:description\" content=\"\u6587\u7ae0\u6d4f\u89c8\u9605\u8bfb798\u6b21\uff0c\u70b9\u8d5e15\u6b21\uff0c\u6536\u85cf8\u6b21\u3002\u786e\u4fdd\u670d\u52a1\u5668\u9632\u706b\u5899\u5f00\u653e\u4e865000\u7aef\u53e3\uff08\u6216\u4f60\u4f7f\u7528\u7684\u5176\u4ed6\u7aef\u53e3\uff09\uff0c\u5bf9\u4e8e\u751f\u4ea7\u73af\u5883\uff0c\u8003\u8651\u4f7f\u7528Nginx\u4f5c\u4e3a\u53cd\u5411\u4ee3\u7406\uff0c\u6dfb\u52a0\u9002\u5f53\u7684\u9519\u8bef\u5904\u7406\u548c\u65e5\u5fd7\u8bb0\u5f55\uff0c\u6839\u636e\u4f60\u7684\u5b9e\u9645\u6a21\u578b\u8c03\u6574\u7279\u5f81\u9884\u5904\u7406\u6b65\u9aa4\uff0c\u8003\u8651\u6dfb\u52a0\u8f93\u5165\u6570\u636e\u9a8c\u8bc1\uff0c\u8fd9\u6837\u4f60\u5c31\u53ef\u4ee5\u5728\u672c\u5730PyCharm\u4e2d\u8f7b\u677e\u8c03\u7528\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\u7684\u6a21\u578b\u4e86\u3002_1\u67087\u65e5-\u673a\u5668\u5b66\u4e60\u7ebf\u4e0a\u5f00\u53d1\u4e0e\u6a21\u578b\u90e8\u7f72\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.wsisp.com\/helps\/25860.html\" \/>\n<meta property=\"og:site_name\" content=\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\" \/>\n<meta property=\"article:published_time\" content=\"2025-04-19T06:01:06+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2025\/04\/20250419060104-68033c20c8981.png\" \/>\n<meta name=\"author\" content=\"admin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"9 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/25860.html\",\"url\":\"https:\/\/www.wsisp.com\/helps\/25860.html\",\"name\":\"\u673a\u5668\u5b66\u4e60\u6a21\u578b\u90e8\u7f72\uff1a\u4f7f\u7528Flask \u5e93\u7684 Python Web \u6846\u67b6\u5c06XGBoost\u6a21\u578b\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\uff08\u7b80\u5355\u5c0f\u6a21\u578b\uff09\u4ece\u6a21\u578b\u8bad\u7ec3\u5230\u90e8\u7f72\u518d\u5230\u672c\u5730\u8c03\u7528 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"isPartOf\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\"},\"datePublished\":\"2025-04-19T06:01:06+00:00\",\"dateModified\":\"2025-04-19T06:01:06+00:00\",\"author\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.wsisp.com\/helps\/25860.html#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.wsisp.com\/helps\/25860.html\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/25860.html#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/www.wsisp.com\/helps\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"\u673a\u5668\u5b66\u4e60\u6a21\u578b\u90e8\u7f72\uff1a\u4f7f\u7528Flask \u5e93\u7684 Python Web \u6846\u67b6\u5c06XGBoost\u6a21\u578b\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\uff08\u7b80\u5355\u5c0f\u6a21\u578b\uff09\u4ece\u6a21\u578b\u8bad\u7ec3\u5230\u90e8\u7f72\u518d\u5230\u672c\u5730\u8c03\u7528\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#website\",\"url\":\"https:\/\/www.wsisp.com\/helps\/\",\"name\":\"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3\",\"description\":\"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41\",\"name\":\"admin\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"contentUrl\":\"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery\",\"caption\":\"admin\"},\"sameAs\":[\"http:\/\/wp.wsisp.com\"],\"url\":\"https:\/\/www.wsisp.com\/helps\/author\/admin\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"\u673a\u5668\u5b66\u4e60\u6a21\u578b\u90e8\u7f72\uff1a\u4f7f\u7528Flask \u5e93\u7684 Python Web \u6846\u67b6\u5c06XGBoost\u6a21\u578b\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\uff08\u7b80\u5355\u5c0f\u6a21\u578b\uff09\u4ece\u6a21\u578b\u8bad\u7ec3\u5230\u90e8\u7f72\u518d\u5230\u672c\u5730\u8c03\u7528 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.wsisp.com\/helps\/25860.html","og_locale":"zh_CN","og_type":"article","og_title":"\u673a\u5668\u5b66\u4e60\u6a21\u578b\u90e8\u7f72\uff1a\u4f7f\u7528Flask \u5e93\u7684 Python Web \u6846\u67b6\u5c06XGBoost\u6a21\u578b\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\uff08\u7b80\u5355\u5c0f\u6a21\u578b\uff09\u4ece\u6a21\u578b\u8bad\u7ec3\u5230\u90e8\u7f72\u518d\u5230\u672c\u5730\u8c03\u7528 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","og_description":"\u6587\u7ae0\u6d4f\u89c8\u9605\u8bfb798\u6b21\uff0c\u70b9\u8d5e15\u6b21\uff0c\u6536\u85cf8\u6b21\u3002\u786e\u4fdd\u670d\u52a1\u5668\u9632\u706b\u5899\u5f00\u653e\u4e865000\u7aef\u53e3\uff08\u6216\u4f60\u4f7f\u7528\u7684\u5176\u4ed6\u7aef\u53e3\uff09\uff0c\u5bf9\u4e8e\u751f\u4ea7\u73af\u5883\uff0c\u8003\u8651\u4f7f\u7528Nginx\u4f5c\u4e3a\u53cd\u5411\u4ee3\u7406\uff0c\u6dfb\u52a0\u9002\u5f53\u7684\u9519\u8bef\u5904\u7406\u548c\u65e5\u5fd7\u8bb0\u5f55\uff0c\u6839\u636e\u4f60\u7684\u5b9e\u9645\u6a21\u578b\u8c03\u6574\u7279\u5f81\u9884\u5904\u7406\u6b65\u9aa4\uff0c\u8003\u8651\u6dfb\u52a0\u8f93\u5165\u6570\u636e\u9a8c\u8bc1\uff0c\u8fd9\u6837\u4f60\u5c31\u53ef\u4ee5\u5728\u672c\u5730PyCharm\u4e2d\u8f7b\u677e\u8c03\u7528\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\u7684\u6a21\u578b\u4e86\u3002_1\u67087\u65e5-\u673a\u5668\u5b66\u4e60\u7ebf\u4e0a\u5f00\u53d1\u4e0e\u6a21\u578b\u90e8\u7f72","og_url":"https:\/\/www.wsisp.com\/helps\/25860.html","og_site_name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","article_published_time":"2025-04-19T06:01:06+00:00","og_image":[{"url":"https:\/\/www.wsisp.com\/helps\/wp-content\/uploads\/2025\/04\/20250419060104-68033c20c8981.png"}],"author":"admin","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"admin","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"9 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.wsisp.com\/helps\/25860.html","url":"https:\/\/www.wsisp.com\/helps\/25860.html","name":"\u673a\u5668\u5b66\u4e60\u6a21\u578b\u90e8\u7f72\uff1a\u4f7f\u7528Flask \u5e93\u7684 Python Web \u6846\u67b6\u5c06XGBoost\u6a21\u578b\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\uff08\u7b80\u5355\u5c0f\u6a21\u578b\uff09\u4ece\u6a21\u578b\u8bad\u7ec3\u5230\u90e8\u7f72\u518d\u5230\u672c\u5730\u8c03\u7528 - \u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","isPartOf":{"@id":"https:\/\/www.wsisp.com\/helps\/#website"},"datePublished":"2025-04-19T06:01:06+00:00","dateModified":"2025-04-19T06:01:06+00:00","author":{"@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41"},"breadcrumb":{"@id":"https:\/\/www.wsisp.com\/helps\/25860.html#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.wsisp.com\/helps\/25860.html"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.wsisp.com\/helps\/25860.html#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.wsisp.com\/helps"},{"@type":"ListItem","position":2,"name":"\u673a\u5668\u5b66\u4e60\u6a21\u578b\u90e8\u7f72\uff1a\u4f7f\u7528Flask \u5e93\u7684 Python Web \u6846\u67b6\u5c06XGBoost\u6a21\u578b\u90e8\u7f72\u5728\u670d\u52a1\u5668\u4e0a\uff08\u7b80\u5355\u5c0f\u6a21\u578b\uff09\u4ece\u6a21\u578b\u8bad\u7ec3\u5230\u90e8\u7f72\u518d\u5230\u672c\u5730\u8c03\u7528"}]},{"@type":"WebSite","@id":"https:\/\/www.wsisp.com\/helps\/#website","url":"https:\/\/www.wsisp.com\/helps\/","name":"\u7f51\u7855\u4e92\u8054\u5e2e\u52a9\u4e2d\u5fc3","description":"\u9999\u6e2f\u670d\u52a1\u5668_\u9999\u6e2f\u4e91\u670d\u52a1\u5668\u8d44\u8baf_\u670d\u52a1\u5668\u5e2e\u52a9\u6587\u6863_\u670d\u52a1\u5668\u6559\u7a0b","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.wsisp.com\/helps\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/358e386c577a3ab51c4493330a20ad41","name":"admin","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.wsisp.com\/helps\/#\/schema\/person\/image\/","url":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","contentUrl":"https:\/\/gravatar.wp-china-yes.net\/avatar\/?s=96&d=mystery","caption":"admin"},"sameAs":["http:\/\/wp.wsisp.com"],"url":"https:\/\/www.wsisp.com\/helps\/author\/admin"}]}},"_links":{"self":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/25860","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/comments?post=25860"}],"version-history":[{"count":0,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/posts\/25860\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/media\/25856"}],"wp:attachment":[{"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/media?parent=25860"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/categories?post=25860"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/tags?post=25860"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/www.wsisp.com\/helps\/wp-json\/wp\/v2\/topic?post=25860"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}