\u7ed8\u5236\u9884\u6d4b\u7ed3\u679c<\/strong><\/h2>\nsales_data['Forecast'] = forecast<\/p>\n
sales_data[['Sales', 'Forecast']].plot()<\/p>\n
plt.title('Retail Sales Prediction')<\/p>\n
plt.show()<\/p>\n
<\/code><\/pre>\n<\/p>\n
3. \u5929\u6c14\u9884\u62a5<\/h4>\n<\/p>\n
\u6c14\u8c61\u6570\u636e\u662f\u5178\u578b\u7684\u65f6\u95f4\u5e8f\u5217\u6570\u636e\uff0c\u53ef\u4ee5\u4f7f\u7528Prophet\u6a21\u578b\u8fdb\u884c\u9884\u6d4b\u3002<\/p>\n<\/p>\n
# \u8bfb\u53d6\u6c14\u8c61\u6570\u636e<\/p>\nweather_data = pd.read_csv('weather_data.csv', index_col='Date', parse_dates=True)<\/p>\n
\u521b\u5efaProphet\u6a21\u578b<\/strong><\/h2>\nmodel = Prophet()<\/p>\n
\u51c6\u5907\u6570\u636e<\/strong><\/h2>\nweather_data.reset_index(inplace=True)<\/p>\n
weather_data.rename(columns={'index': 'ds', 'Temperature': 'y'}, inplace=True)<\/p>\n
\u62df\u5408Prophet\u6a21\u578b<\/strong><\/h2>\nmodel.fit(weather_data)<\/p>\n
\u8fdb\u884c\u9884\u6d4b<\/strong><\/h2>\nfuture = model.make_future_dataframe(periods=30, freq='D')<\/p>\n
forecast = model.predict(future)<\/p>\n
\u7ed8\u5236\u9884\u6d4b\u7ed3\u679c<\/strong><\/h2>\nmodel.plot(forecast)<\/p>\n
plt.title('Weather Forecast')<\/p>\n
plt.show()<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u4e5d\u3001\u65f6\u95f4\u5e8f\u5217\u7684\u9ad8\u7ea7\u5904\u7406<\/h3>\n<\/p>\n
\u9664\u4e86\u57fa\u672c\u7684\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u65b9\u6cd5\uff0c\u8fd8\u6709\u4e00\u4e9b\u9ad8\u7ea7\u5904\u7406\u65b9\u6cd5\u53ef\u4ee5\u63d0\u9ad8\u5206\u6790\u548c\u9884\u6d4b\u7684\u51c6\u786e\u6027\u3002<\/p>\n<\/p>\n
1. \u65f6\u95f4\u5e8f\u5217\u7684\u5e73\u7a33\u6027\u68c0\u6d4b<\/h4>\n<\/p>\n
\u5e73\u7a33\u6027\u662f\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u7684\u91cd\u8981\u5047\u8bbe\u3002\u53ef\u4ee5\u4f7f\u7528ADF\uff08Augmented Dickey-Fuller\uff09\u68c0\u9a8c\u6765\u68c0\u6d4b\u65f6\u95f4\u5e8f\u5217\u7684\u5e73\u7a33\u6027\u3002<\/p>\n<\/p>\n
from statsmodels.tsa.stattools import adfuller<\/p>\n\u8fdb\u884cADF\u68c0\u9a8c<\/strong><\/h2>\nresult = adfuller(time_series['Value'])<\/p>\n
print('ADF Statistic:', result[0])<\/p>\n
print('p-value:', result[1])<\/p>\n
<\/code><\/pre>\n<\/p>\n
2. \u65f6\u95f4\u5e8f\u5217\u7684\u5dee\u5206\u5904\u7406<\/h4>\n<\/p>\n
\u5dee\u5206\u5904\u7406\u662f\u4e00\u79cd\u5c06\u975e\u5e73\u7a33\u65f6\u95f4\u5e8f\u5217\u8f6c\u6362\u4e3a\u5e73\u7a33\u65f6\u95f4\u5e8f\u5217\u7684\u65b9\u6cd5\u3002<\/p>\n<\/p>\n
# \u8fdb\u884c\u5dee\u5206\u5904\u7406<\/p>\ntime_series['Differenced'] = time_series['Value'].diff()<\/p>\n
\u7ed8\u5236\u5dee\u5206\u540e\u7684\u65f6\u95f4\u5e8f\u5217<\/strong><\/h2>\ntime_series['Differenced'].plot()<\/p>\n
plt.title('Differenced Time Series')<\/p>\n
plt.show()<\/p>\n
<\/code><\/pre>\n<\/p>\n
3. \u65f6\u95f4\u5e8f\u5217\u7684\u5b63\u8282\u6027\u8c03\u6574<\/h4>\n<\/p>\n
\u5b63\u8282\u6027\u8c03\u6574\u662f\u53bb\u9664\u65f6\u95f4\u5e8f\u5217\u4e2d\u5b63\u8282\u6027\u6210\u5206\u7684\u65b9\u6cd5\uff0c\u53ef\u4ee5\u4f7f\u7528seasonal_decompose()<\/code>\u51fd\u6570\u7684\u5b63\u8282\u6027\u6210\u5206\u8fdb\u884c\u8c03\u6574\u3002<\/p>\n<\/p>\n# \u8fdb\u884c\u5b63\u8282\u6027\u8c03\u6574<\/p>\ntime_series['Seasonally_Adjusted'] = time_series['Value'] - result.seasonal<\/p>\n
\u7ed8\u5236\u5b63\u8282\u6027\u8c03\u6574\u540e\u7684\u65f6\u95f4\u5e8f\u5217<\/strong><\/h2>\ntime_series['Seasonally_Adjusted'].plot()<\/p>\n
plt.title('Seasonally Adjusted Time Series')<\/p>\n
plt.show()<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u5341\u3001\u65f6\u95f4\u5e8f\u5217\u7684\u672a\u6765\u53d1\u5c55\u65b9\u5411<\/h3>\n<\/p>\n
\u968f\u7740\u6570\u636e\u79d1\u5b66\u548c\u673a\u5668\u5b66\u4e60\u7684\u53d1\u5c55\uff0c\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u4e5f\u5728\u4e0d\u65ad\u8fdb\u6b65\u3002\u4ee5\u4e0b\u662f\u51e0\u4e2a\u672a\u6765\u7684\u53d1\u5c55\u65b9\u5411\u3002<\/p>\n<\/p>\n
1. \u6df1\u5ea6\u5b66\u4e60\u5728\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u4e2d\u7684\u5e94\u7528<\/h4>\n<\/p>\n
\u6df1\u5ea6\u5b66\u4e60\u5728\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u4e2d\u7684\u5e94\u7528\u8d8a\u6765\u8d8a\u5e7f\u6cdb\uff0c\u7279\u522b\u662fRNN\uff08\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\uff09\u548cLSTM\uff08\u957f\u77ed\u671f\u8bb0\u5fc6\u7f51\u7edc\uff09\u5728\u5904\u7406\u65f6\u95f4\u5e8f\u5217\u6570\u636e\u65b9\u9762\u8868\u73b0\u51fa\u8272\u3002<\/p>\n<\/p>\n
import tensorflow as tf<\/p>\nfrom tensorflow.keras.models import Sequential<\/p>\n
from tensorflow.keras.layers import LSTM, Dense<\/p>\n
\u51c6\u5907\u6570\u636e<\/strong><\/h2>\nX = time_series[['Lag_1', 'Lag_2']].values[2:]<\/p>\n
y = time_series['Value'].values[2:]<\/p>\n
\u6784\u5efaLSTM\u6a21\u578b<\/strong><\/h2>\nmodel = Sequential()<\/p>\n
model.add(LSTM(50, input_shape=(X.shape[1], 1)))<\/p>\n
model.add(Dense(1))<\/p>\n
model.compile(loss='mean_squared_error', optimizer='adam')<\/p>\n
\u8bad\u7ec3\u6a21\u578b<\/strong><\/h2>\nmodel.fit(X, y, epochs=100, batch_size=1, verbose=2)<\/p>\n
\u8fdb\u884c\u9884\u6d4b<\/strong><\/h2>\nforecast = model.predict(X[-12:])<\/p>\n
<\/code><\/pre>\n<\/p>\n
2. \u65f6\u95f4\u5e8f\u5217\u5206\u6790\u5de5\u5177\u548c\u5e73\u53f0\u7684\u53d1\u5c55<\/h4>\n<\/p>\n
\u968f\u7740\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u7684\u9700\u6c42\u589e\u52a0\uff0c\u8d8a\u6765\u8d8a\u591a\u7684\u5de5\u5177\u548c\u5e73\u53f0\u6d8c\u73b0\u51fa\u6765\uff0c\u5982Facebook\u7684Prophet\u3001Google\u7684TensorFlow\u7b49\u3002\u8fd9\u4e9b\u5de5\u5177\u548c\u5e73\u53f0\u63d0\u4f9b\u4e86\u5f3a\u5927\u7684\u529f\u80fd\u548c\u6613\u7528\u7684\u754c\u9762\uff0c\u5e2e\u52a9\u6570\u636e\u79d1\u5b66\u5bb6\u548c\u5206\u6790\u5e08\u66f4\u9ad8\u6548\u5730\u8fdb\u884c\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u3002<\/p>\n<\/p>\n
3. \u65f6\u95f4\u5e8f\u5217\u5206\u6790\u4e0e\u5927\u6570\u636e\u6280\u672f\u7684\u7ed3\u5408<\/h4>\n<\/p>\n
\u968f\u7740\u5927\u6570\u636e\u6280\u672f\u7684\u53d1\u5c55\uff0c\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u4e5f\u9010\u6e10\u4e0e\u5927\u6570\u636e\u6280\u672f\u7ed3\u5408\uff0c\u901a\u8fc7\u5206\u5e03\u5f0f\u8ba1\u7b97\u548c\u5b58\u50a8\u6280\u672f\uff0c\u53ef\u4ee5\u5904\u7406\u66f4\u5927\u89c4\u6a21\u7684\u65f6\u95f4\u5e8f\u5217\u6570\u636e\uff0c\u63d0\u9ad8\u5206\u6790\u548c\u9884\u6d4b\u7684\u51c6\u786e\u6027\u548c\u6548\u7387\u3002<\/p>\n<\/p>\n
\u7efc\u4e0a\u6240\u8ff0\uff0cPython\u65f6\u95f4\u5e8f\u5217\u7684\u5904\u7406\u65b9\u6cd5\u4e30\u5bcc\u591a\u6837\uff0c\u5305\u62ec\u6570\u636e\u6e05\u6d17\u3001\u65f6\u95f4\u5e8f\u5217\u5206\u89e3\u3001\u65f6\u95f4\u5e8f\u5217\u5efa\u6a21\u3001\u6a21\u578b\u8bc4\u4ef7\u3001\u7279\u5f81\u5de5\u7a0b\u7b49\u591a\u4e2a\u6b65\u9aa4\u3002\u901a\u8fc7\u4f7f\u7528pandas\u3001statsmodels\u3001fbprophet\u7b49\u5e93\uff0c\u53ef\u4ee5\u9ad8\u6548\u5730\u8fdb\u884c\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u548c\u9884\u6d4b\u3002\u540c\u65f6\uff0c\u968f\u7740\u6df1\u5ea6\u5b66\u4e60\u548c\u5927\u6570\u636e\u6280\u672f\u7684\u53d1\u5c55\uff0c\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u4e5f\u5728\u4e0d\u65ad\u8fdb\u6b65\uff0c\u672a\u6765\u5c06\u6709\u66f4\u591a\u7684\u5e94\u7528\u548c\u53d1\u5c55\u65b9\u5411\u3002<\/p>\n<\/p>\n
\u76f8\u5173\u95ee\u7b54FAQs\uff1a<\/strong><\/h2>\n \u5982\u4f55\u5728Python\u4e2d\u8bfb\u53d6\u65f6\u95f4\u5e8f\u5217\u6570\u636e\uff1f<\/strong>
\u8981\u8bfb\u53d6\u65f6\u95f4\u5e8f\u5217\u6570\u636e\uff0c\u53ef\u4ee5\u4f7f\u7528Pandas\u5e93\u4e2d\u7684read_csv<\/code>\u51fd\u6570\uff0c\u7ed3\u5408parse_dates<\/code>\u53c2\u6570\u5c06\u65e5\u671f\u5217\u89e3\u6790\u4e3a\u65e5\u671f\u65f6\u95f4\u683c\u5f0f\u3002\u786e\u4fdd\u5728\u8bfb\u53d6\u6570\u636e\u65f6\u6307\u5b9a\u9002\u5f53\u7684\u65e5\u671f\u89e3\u6790\u683c\u5f0f\uff0c\u4ee5\u4fbfPandas\u80fd\u591f\u6b63\u786e\u8bc6\u522b\u65e5\u671f\u3002<\/p>\nPython\u4e2d\u6709\u54ea\u4e9b\u5e93\u53ef\u4ee5\u5904\u7406\u65f6\u95f4\u5e8f\u5217\u6570\u636e\uff1f<\/strong>
\u5904\u7406\u65f6\u95f4\u5e8f\u5217\u6570\u636e\u7684\u5e38\u7528\u5e93\u5305\u62ecPandas\u3001NumPy\u548cStatsmodels\u3002Pandas\u63d0\u4f9b\u4e86\u4e30\u5bcc\u7684\u6570\u636e\u7ed3\u6784\u548c\u51fd\u6570\u6765\u5904\u7406\u65f6\u95f4\u5e8f\u5217\uff0cNumPy\u5219\u53ef\u4ee5\u7528\u4e8e\u6570\u503c\u8ba1\u7b97\uff0c\u800cStatsmodels\u5219\u9002\u5408\u8fdb\u884c\u65f6\u95f4\u5e8f\u5217\u5206\u6790\u548c\u5efa\u6a21\u3002<\/p>\n\u5982\u4f55\u5728Python\u4e2d\u8fdb\u884c\u65f6\u95f4\u5e8f\u5217\u7684\u91cd\u91c7\u6837\uff1f<\/strong>
\u91cd\u91c7\u6837\u53ef\u4ee5\u901a\u8fc7Pandas\u7684resample<\/code>\u65b9\u6cd5\u5b9e\u73b0\u3002\u8be5\u65b9\u6cd5\u5141\u8bb8\u5c06\u65f6\u95f4\u5e8f\u5217\u6570\u636e\u8f6c\u6362\u4e3a\u4e0d\u540c\u7684\u9891\u7387\uff0c\u4f8b\u5982\u4ece\u65e5\u9891\u7387\u8f6c\u6362\u4e3a\u6708\u9891\u7387\u3002\u5728\u91cd\u91c7\u6837\u65f6\uff0c\u53ef\u4ee5\u6307\u5b9a\u805a\u5408\u51fd\u6570\uff0c\u5982mean<\/code>\u3001sum<\/code>\u7b49\uff0c\u4ee5\u8ba1\u7b97\u65b0\u7684\u65f6\u95f4\u6bb5\u5185\u7684\u503c\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"\u5904\u7406Python\u65f6\u95f4\u5e8f\u5217\u7684\u65b9\u6cd5\u6709\uff1a\u4f7f\u7528pandas\u5e93\u3001\u8fdb\u884c\u6570\u636e\u6e05\u6d17\u3001\u8fdb\u884c\u65f6\u95f4\u5e8f\u5217\u5206\u89e3\u3001\u8fdb\u884c\u65f6\u95f4\u5e8f\u5217\u5efa\u6a21\u3002\u5176\u4e2d\uff0c […]","protected":false},"author":3,"featured_media":1172656,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[37],"tags":[],"acf":[],"_links":{"self":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1172647"}],"collection":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/comments?post=1172647"}],"version-history":[{"count":"1","href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1172647\/revisions"}],"predecessor-version":[{"id":1172659,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1172647\/revisions\/1172659"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media\/1172656"}],"wp:attachment":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media?parent=1172647"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/categories?post=1172647"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/tags?post=1172647"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
![\"\u5982\u4f55\u5904\u7406python\u65f6\u95f4\u5e8f\u5217\"](\"https:\/\/cdn-kb.worktile.com\/kb\/wp-content\/uploads\/2024\/04\/26074636\/8337c299-55cf-4228-abb1-b25a2d904200.webp\")
<\/p>\n