{"id":4225,"date":"2026-03-01T10:15:15","date_gmt":"2026-03-01T02:15:15","guid":{"rendered":"http:\/\/viplao.com\/?p=4225"},"modified":"2026-03-01T10:15:17","modified_gmt":"2026-03-01T02:15:17","slug":"%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96","status":"publish","type":"post","link":"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/","title":{"rendered":"\u6570\u636e\u6280\u80fd\uff08SKILLS\uff09 \uff1aDay 6\uff1a\u9500\u552e\u9884\u6d4b\u4e0e\u5e93\u5b58\u4f18\u5316"},"content":{"rendered":"\n<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_71 counter-hierarchy ez-toc-counter ez-toc-grey ez-toc-container-direction\">\n<div class=\"ez-toc-title-container\">\n<p class=\"ez-toc-title\" style=\"cursor:inherit\">\u6587\u7ae0\u76ee\u5f55<\/p>\n<span class=\"ez-toc-title-toggle\"><a href=\"#\" class=\"ez-toc-pull-right ez-toc-btn ez-toc-btn-xs ez-toc-btn-default ez-toc-toggle\" aria-label=\"Toggle Table of Content\"><span class=\"ez-toc-js-icon-con\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #999;color:#999\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #999;color:#999\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/span><\/a><\/span><\/div>\n<nav><ul class='ez-toc-list ez-toc-list-level-1 eztoc-toggle-hide-by-default' ><ul class='ez-toc-list-level-3' ><li class='ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/#%E8%AF%BE%E7%A8%8B%E7%9B%AE%E6%A0%87\" title=\"\u8bfe\u7a0b\u76ee\u6807\">\u8bfe\u7a0b\u76ee\u6807<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/#%E6%A0%B8%E5%BF%83%E6%8A%80%E8%83%BD\" title=\"\u6838\u5fc3\u6280\u80fd\">\u6838\u5fc3\u6280\u80fd<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/#%E8%AF%BE%E4%BB%B6%E5%86%85%E5%AE%B9\" title=\"\u8bfe\u4ef6\u5185\u5bb9\">\u8bfe\u4ef6\u5185\u5bb9<\/a><ul class='ez-toc-list-level-4' ><li class='ez-toc-heading-level-4'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/#1_%E9%94%80%E5%94%AE%E9%A2%84%E6%B5%8B%E6%A8%A1%E5%9E%8B\" title=\"1. \u9500\u552e\u9884\u6d4b\u6a21\u578b\">1. \u9500\u552e\u9884\u6d4b\u6a21\u578b<\/a><\/li><\/ul><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/#%E5%AE%89%E8%A3%85%E8%A6%81%E6%B1%82\" title=\"\u5b89\u88c5\u8981\u6c42\">\u5b89\u88c5\u8981\u6c42<\/a><\/li><\/ul><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/#%E8%AF%BE%E7%A8%8B%E6%80%BB%E7%BB%93\" title=\"\u8bfe\u7a0b\u603b\u7ed3\">\u8bfe\u7a0b\u603b\u7ed3<\/a><ul class='ez-toc-list-level-3' ><li class='ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-7\" href=\"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/#4-6%E5%A4%A9%E6%8A%80%E8%83%BD%E6%8E%8C%E6%8F%A1%E6%B8%85%E5%8D%95\" title=\"4-6\u5929\u6280\u80fd\u638c\u63e1\u6e05\u5355\">4-6\u5929\u6280\u80fd\u638c\u63e1\u6e05\u5355<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-8\" href=\"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/#%E8%BF%90%E8%A1%8C%E8%AF%B4%E6%98%8E\" title=\"\u8fd0\u884c\u8bf4\u660e\">\u8fd0\u884c\u8bf4\u660e<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-9\" href=\"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/#%E4%B8%9A%E5%8A%A1%E5%BA%94%E7%94%A8%E4%BB%B7%E5%80%BC\" title=\"\u4e1a\u52a1\u5e94\u7528\u4ef7\u503c\">\u4e1a\u52a1\u5e94\u7528\u4ef7\u503c<\/a><\/li><\/ul><\/li><\/ul><\/nav><\/div>\n<h3 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"%E8%AF%BE%E7%A8%8B%E7%9B%AE%E6%A0%87\"><\/span><strong>\u8bfe\u7a0b\u76ee\u6807<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h3>\n\n\n\n<p>\u638c\u63e1\u65f6\u95f4\u5e8f\u5217\u9884\u6d4b\u3001\u5e93\u5b58\u4f18\u5316\u6a21\u578b\u3001\u81ea\u52a8\u62a5\u8868\u751f\u6210\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"%E6%A0%B8%E5%BF%83%E6%8A%80%E8%83%BD\"><\/span><strong>\u6838\u5fc3\u6280\u80fd<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h3>\n\n\n\n<ul>\n<li>\u65f6\u95f4\u5e8f\u5217\u9884\u6d4b\uff08ARIMA\u3001Prophet\uff09<\/li>\n\n\n\n<li>\u5e93\u5b58\u4f18\u5316\u4e0e\u5b89\u5168\u5e93\u5b58\u8ba1\u7b97<\/li>\n\n\n\n<li>\u81ea\u52a8\u5316\u62a5\u8868\u751f\u6210<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"%E8%AF%BE%E4%BB%B6%E5%86%85%E5%AE%B9\"><\/span><strong>\u8bfe\u4ef6\u5185\u5bb9<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h3>\n\n\n\n<h4 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"1_%E9%94%80%E5%94%AE%E9%A2%84%E6%B5%8B%E6%A8%A1%E5%9E%8B\"><\/span><strong>1. \u9500\u552e\u9884\u6d4b\u6a21\u578b<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h4>\n\n\n\n<pre class=\"wp-block-preformatted\">python<code># \u5b89\u88c5\u4f9d\u8d56\uff1apip install pandas numpy matplotlib statsmodels scikit-learn prophet\nimport pandas as pd\nimport numpy as np\nimport matplotlib.pyplot as plt\nfrom datetime import datetime, timedelta\nimport warnings\nwarnings.filterwarnings('ignore')\n\nclass SalesForecaster:\n    def __init__(self, sales_data):\n        \"\"\"\n        \u521d\u59cb\u5316\u9500\u552e\u9884\u6d4b\u5668\n        \n        \u53c2\u6570:\n        sales_data: \u9500\u552e\u6570\u636e\uff0c\u9700\u5305\u542b\u65e5\u671f\u548c\u9500\u552e\u989d\n        \"\"\"\n        self.sales_data = sales_data.copy()\n        self.forecast_results = {}\n        \n        # \u786e\u4fdd\u65e5\u671f\u683c\u5f0f\n        if 'date' in self.sales_data.columns:\n            self.sales_data['date'] = pd.to_datetime(self.sales_data['date'])\n        elif 'order_date' in self.sales_data.columns:\n            self.sales_data['date'] = pd.to_datetime(self.sales_data['order_date'])\n            self.sales_data = self.sales_data.rename(columns={'order_date': 'date'})\n    \n    def prepare_time_series(self, freq='D'):\n        \"\"\"\u51c6\u5907\u65f6\u95f4\u5e8f\u5217\u6570\u636e\"\"\"\n        # \u6309\u65e5\u671f\u805a\u5408\u9500\u552e\u989d\n        if 'order_amount' in self.sales_data.columns:\n            ts_data = self.sales_data.groupby('date')['order_amount'].sum().reset_index()\n        elif 'sales' in self.sales_data.columns:\n            ts_data = self.sales_data.groupby('date')['sales'].sum().reset_index()\n        else:\n            # \u5047\u8bbe\u6709quantity\u548cunit_price\n            self.sales_data['sales'] = self.sales_data['quantity'] * self.sales_data['unit_price']\n            ts_data = self.sales_data.groupby('date')['sales'].sum().reset_index()\n        \n        # \u8bbe\u7f6e\u65e5\u671f\u7d22\u5f15\n        ts_data = ts_data.set_index('date')\n        \n        # \u91cd\u91c7\u6837\u5230\u6307\u5b9a\u9891\u7387\n        ts_data = ts_data.resample(freq).sum()\n        \n        # \u586b\u5145\u7f3a\u5931\u503c\n        ts_data = ts_data.fillna(method='ffill').fillna(0)\n        \n        return ts_data\n    \n    def moving_average_forecast(self, ts_data, window=7, forecast_days=30):\n        \"\"\"\u79fb\u52a8\u5e73\u5747\u9884\u6d4b\"\"\"\n        # \u8ba1\u7b97\u79fb\u52a8\u5e73\u5747\n        ma = ts_data.rolling(window=window).mean()\n        \n        # \u9884\u6d4b\uff1a\u4f7f\u7528\u6700\u540ewindow\u5929\u7684\u5e73\u5747\u503c\n        last_values = ts_data.iloc[-window:].values.flatten()\n        forecast = np.full(forecast_days, np.mean(last_values))\n        \n        # \u521b\u5efa\u9884\u6d4b\u65e5\u671f\n        last_date = ts_data.index[-1]\n        forecast_dates = pd.date_range(\n            start=last_date + pd.Timedelta(days=1),\n            periods=forecast_days,\n            freq=ts_data.index.freq\n        )\n        \n        forecast_df = pd.DataFrame({\n            'date': forecast_dates,\n            'forecast': forecast,\n            'model': 'Moving Average'\n        })\n        \n        self.forecast_results['moving_average'] = {\n            'forecast': forecast_df,\n            'ma_values': ma,\n            'window': window\n        }\n        \n        return forecast_df\n    \n    def exponential_smoothing_forecast(self, ts_data, alpha=0.3, forecast_days=30):\n        \"\"\"\u6307\u6570\u5e73\u6ed1\u9884\u6d4b\"\"\"\n        from statsmodels.tsa.holtwinters import SimpleExpSmoothing\n        \n        # \u62df\u5408\u6a21\u578b\n        model = SimpleExpSmoothing(ts_data)\n        fitted_model = model.fit(smoothing_level=alpha, optimized=False)\n        \n        # \u9884\u6d4b\n        forecast = fitted_model.forecast(forecast_days)\n        \n        # \u521b\u5efa\u9884\u6d4b\u6570\u636e\u6846\n        last_date = ts_data.index[-1]\n        forecast_dates = pd.date_range(\n            start=last_date + pd.Timedelta(days=1),\n            periods=forecast_days,\n            freq=ts_data.index.freq\n        )\n        \n        forecast_df = pd.DataFrame({\n            'date': forecast_dates,\n            'forecast': forecast.values,\n            'model': 'Exponential Smoothing'\n        })\n        \n        self.forecast_results['exponential_smoothing'] = {\n            'forecast': forecast_df,\n            'fitted_values': fitted_model.fittedvalues,\n            'alpha': alpha\n        }\n        \n        return forecast_df\n    \n    def arima_forecast(self, ts_data, order=(1,1,1), forecast_days=30):\n        \"\"\"ARIMA\u9884\u6d4b\"\"\"\n        from statsmodels.tsa.arima.model import ARIMA\n        \n        try:\n            # \u62df\u5408ARIMA\u6a21\u578b\n            model = ARIMA(ts_data, order=order)\n            fitted_model = model.fit()\n            \n            # \u9884\u6d4b\n            forecast = fitted_model.forecast(steps=forecast_days)\n            \n            # \u521b\u5efa\u9884\u6d4b\u6570\u636e\u6846\n            last_date = ts_data.index[-1]\n            forecast_dates = pd.date_range(\n                start=last_date + pd.Timedelta(days=1),\n                periods=forecast_days,\n                freq=ts_data.index.freq\n            )\n            \n            forecast_df = pd.DataFrame({\n                'date': forecast_dates,\n                'forecast': forecast.values,\n                'model': f'ARIMA{order}'\n            })\n            \n            self.forecast_results['arima'] = {\n                'forecast': forecast_df,\n                'fitted_values': fitted_model.fittedvalues,\n                'order': order,\n                'aic': fitted_model.aic,\n                'bic': fitted_model.bic\n            }\n            \n            return forecast_df\n            \n        except Exception as e:\n            print(f\"ARIMA\u6a21\u578b\u62df\u5408\u5931\u8d25: {e}\")\n            return None\n    \n    def prophet_forecast(self, ts_data, forecast_days=30):\n        \"\"\"Prophet\u9884\u6d4b\uff08\u9700\u8981\u5b89\u88c5fbprophet\uff09\"\"\"\n        try:\n            from prophet import Prophet\n            \n            # \u51c6\u5907Prophet\u683c\u5f0f\u6570\u636e\n            prophet_data = ts_data.reset_index()\n            prophet_data.columns = ['ds', 'y']\n            \n            # \u521b\u5efa\u5e76\u62df\u5408\u6a21\u578b\n            model = Prophet(\n                yearly_seasonality=True,\n                weekly_seasonality=True,\n                daily_seasonality=False\n            )\n            model.fit(prophet_data)\n            \n            # \u521b\u5efa\u672a\u6765\u65e5\u671f\n            future = model.make_future_dataframe(periods=forecast_days, freq='D')\n            \n            # \u9884\u6d4b\n            forecast = model.predict(future)\n            \n            # \u63d0\u53d6\u9884\u6d4b\u7ed3\u679c\n            forecast_df = forecast[['ds', 'yhat', 'yhat_lower', 'yhat_upper']].tail(forecast_days)\n            forecast_df.columns = ['date', 'forecast', 'lower_bound', 'upper_bound']\n            forecast_df['model'] = 'Prophet'\n            \n            self.forecast_results['prophet'] = {\n                'forecast': forecast_df,\n                'model_object': model,\n                'components': forecast[['ds', 'trend', 'yearly', 'weekly']]\n            }\n            \n            return forecast_df\n            \n        except ImportError:\n            print(\"Prophet\u672a\u5b89\u88c5\uff0c\u8df3\u8fc7\u6b64\u65b9\u6cd5\")\n            print(\"\u5b89\u88c5\u547d\u4ee4: pip install prophet\")\n            return None\n        except Exception as e:\n            print(f\"Prophet\u6a21\u578b\u62df\u5408\u5931\u8d25: {e}\")\n            return None\n    \n    def evaluate_forecast(self, actual, forecast, model_name):\n        \"\"\"\u8bc4\u4f30\u9884\u6d4b\u6548\u679c\"\"\"\n        from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n        \n        # \u786e\u4fdd\u957f\u5ea6\u4e00\u81f4\n        min_len = min(len(actual), len(forecast))\n        actual = actual[:min_len]\n        forecast = forecast[:min_len]\n        \n        metrics = {\n            'MAE': mean_absolute_error(actual, forecast),\n            'RMSE': np.sqrt(mean_squared_error(actual, forecast)),\n            'MAPE': np.mean(np.abs((actual - forecast) \/ actual)) * 100,\n            'R2': r2_score(actual, forecast)\n        }\n        \n        print(f\"\\n{model_name} \u9884\u6d4b\u6548\u679c\u8bc4\u4f30:\")\n        print(f\"  MAE\uff08\u5e73\u5747\u7edd\u5bf9\u8bef\u5dee\uff09: {metrics['MAE']:.2f}\")\n        print(f\"  RMSE\uff08\u5747\u65b9\u6839\u8bef\u5dee\uff09: {metrics['RMSE']:.2f}\")\n        print(f\"  MAPE\uff08\u5e73\u5747\u7edd\u5bf9\u767e\u5206\u6bd4\u8bef\u5dee\uff09: {metrics['MAPE']:.2f}%\")\n        print(f\"  R\u00b2\uff08\u51b3\u5b9a\u7cfb\u6570\uff09: {metrics['R2']:.4f}\")\n        \n        return metrics\n    \n    def visualize_forecasts(self, ts_data, forecast_days=30):\n        \"\"\"\u53ef\u89c6\u5316\u9884\u6d4b\u7ed3\u679c\"\"\"\n        fig, axes = plt.subplots(2, 2, figsize=(15, 10))\n        \n        # \u539f\u59cb\u65f6\u95f4\u5e8f\u5217\n        ax1 = axes[0, 0]\n        ts_data.plot(ax=ax1, color='blue', linewidth=2, label='\u5b9e\u9645\u9500\u552e\u989d')\n        ax1.set_title('\u539f\u59cb\u9500\u552e\u65f6\u95f4\u5e8f\u5217', fontsize=12, fontweight='bold')\n        ax1.set_xlabel('\u65e5\u671f')\n        ax1.set_ylabel('\u9500\u552e\u989d')\n        ax1.legend()\n        ax1.grid(True, alpha=0.3)\n        \n        # \u79fb\u52a8\u5e73\u5747\u9884\u6d4b\n        if 'moving_average' in self.forecast_results:\n            ax2 = axes[0, 1]\n            ts_data.plot(ax=ax2, color='blue', alpha=0.5, label='\u5b9e\u9645\u503c')\n            self.forecast_results['moving_average']['ma_values'].plot(\n                ax=ax2, color='red', linewidth=2, label='\u79fb\u52a8\u5e73\u5747'\n            )\n            forecast_df = self.forecast_results['moving_average']['forecast']\n            ax2.plot(forecast_df['date'], forecast_df['forecast'], \n                    color='green', linewidth=2, linestyle='--', label='\u9884\u6d4b')\n            ax2.set_title('\u79fb\u52a8\u5e73\u5747\u9884\u6d4b', fontsize=12, fontweight='bold')\n            ax2.legend()\n            ax2.grid(True, alpha=0.3)\n        \n        # \u6307\u6570\u5e73\u6ed1\u9884\u6d4b\n        if 'exponential_smoothing' in self.forecast_results:\n            ax3 = axes[1, 0]\n            ts_data.plot(ax=ax3, color='blue', alpha=0.5, label='\u5b9e\u9645\u503c')\n            self.forecast_results['exponential_smoothing']['fitted_values'].plot(\n                ax=ax3, color='orange', linewidth=2, label='\u62df\u5408\u503c'\n            )\n            forecast_df = self.forecast_results['exponential_smoothing']['forecast']\n            ax3.plot(forecast_df['date'], forecast_df['forecast'], \n                    color='green', linewidth=2, linestyle='--', label='\u9884\u6d4b')\n            ax3.set_title('\u6307\u6570\u5e73\u6ed1\u9884\u6d4b', fontsize=12, fontweight='bold')\n            ax3.legend()\n            ax3.grid(True, alpha=0.3)\n        \n        # \u591a\u6a21\u578b\u5bf9\u6bd4\n        ax4 = axes[1, 1]\n        ts_data.tail(60).plot(ax=ax4, color='black', linewidth=2, label='\u5b9e\u9645\u503c')\n        \n        colors = ['red', 'green', 'blue', 'purple']\n        for i, (model_name, result) in enumerate(self.forecast_results.items()):\n            if 'forecast' in result:\n                forecast_df = result['forecast']\n                ax4.plot(forecast_df['date'], forecast_df['forecast'], \n                        color=colors[i % len(colors)], linewidth=2, \n                        linestyle='--', label=f'{model_name}\u9884\u6d4b')\n        \n        ax4.set_title('\u591a\u6a21\u578b\u9884\u6d4b\u5bf9\u6bd4', fontsize=12, fontweight='bold')\n        ax4.legend()\n        ax4.grid(True, alpha=0.3)\n        \n        plt.tight_layout()\n        plt.show()\n\nclass InventoryOptimizer:\n    def __init__(self, sales_data, lead_time=7, service_level=0.95):\n        \"\"\"\n        \u5e93\u5b58\u4f18\u5316\u5668\n        \n        \u53c2\u6570:\n        sales_data: \u9500\u552e\u6570\u636e\n        lead_time: \u8865\u8d27\u63d0\u524d\u671f\uff08\u5929\uff09\n        service_level: \u670d\u52a1\u6c34\u5e73\uff08\u5e93\u5b58\u6ee1\u8db3\u7387\uff09\n        \"\"\"\n        self.sales_data = sales_data.copy()\n        self.lead_time = lead_time\n        self.service_level = service_level\n        self.z_score = self._calculate_z_score(service_level)\n    \n    def _calculate_z_score(self, service_level):\n        \"\"\"\u8ba1\u7b97\u670d\u52a1\u6c34\u5e73\u5bf9\u5e94\u7684Z\u503c\"\"\"\n        from scipy import stats\n        return stats.norm.ppf(service_level)\n    \n    def calculate_demand_stats(self, product_id=None, freq='D'):\n        \"\"\"\u8ba1\u7b97\u9700\u6c42\u7edf\u8ba1\"\"\"\n        if product_id:\n            product_data = self.sales_data[self.sales_data['product_id'] == product_id]\n        else:\n            product_data = self.sales_data\n        \n        # \u6309\u65e5\u671f\u805a\u5408\n        if 'date' in product_data.columns:\n            date_col = 'date'\n        elif 'order_date' in product_data.columns:\n            date_col = 'order_date'\n            product_data = product_data.rename(columns={'order_date': 'date'})\n        \n        # \u8ba1\u7b97\u6bcf\u65e5\u9700\u6c42\n        daily_demand = product_data.groupby('date')['quantity'].sum().resample(freq).sum()\n        \n        stats = {\n            'mean_demand': daily_demand.mean(),\n            'std_demand': daily_demand.std(),\n            'cv_demand': daily_demand.std() \/ daily_demand.mean() if daily_demand.mean() &gt; 0 else 0,\n            'min_demand': daily_demand.min(),\n            'max_demand': daily_demand.max(),\n            'median_demand': daily_demand.median()\n        }\n        \n        return stats, daily_demand\n    \n    def calculate_eoq(self, annual_demand, ordering_cost, holding_cost_per_unit):\n        \"\"\"\u8ba1\u7b97\u7ecf\u6d4e\u8ba2\u8d27\u6279\u91cf\uff08EOQ\uff09\"\"\"\n        eoq = np.sqrt((2 * annual_demand * ordering_cost) \/ holding_cost_per_unit)\n        return eoq\n    \n    def calculate_safety_stock(self, demand_mean, demand_std, lead_time):\n        \"\"\"\u8ba1\u7b97\u5b89\u5168\u5e93\u5b58\"\"\"\n        safety_stock = self.z_score * demand_std * np.sqrt(lead_time)\n        return safety_stock\n    \n    def calculate_reorder_point(self, demand_mean, demand_std, lead_time):\n        \"\"\"\u8ba1\u7b97\u518d\u8ba2\u8d27\u70b9\"\"\"\n        safety_stock = self.calculate_safety_stock(demand_mean, demand_std, lead_time)\n        reorder_point = (demand_mean * lead_time) + safety_stock\n        return reorder_point\n    \n    def optimize_inventory(self, product_id, ordering_cost=50, holding_cost_rate=0.2, unit_cost=100):\n        \"\"\"\u4f18\u5316\u5355\u4e2a\u4ea7\u54c1\u7684\u5e93\u5b58\"\"\"\n        # \u8ba1\u7b97\u9700\u6c42\u7edf\u8ba1\n        stats, daily_demand = self.calculate_demand_stats(product_id)\n        \n        # \u5e74\u5316\u9700\u6c42\n        annual_demand = stats['mean_demand'] * 365\n        \n        # \u8ba1\u7b97EOQ\n        holding_cost_per_unit = unit_cost * holding_cost_rate\n        eoq = self.calculate_eoq(annual_demand, ordering_cost, holding_cost_per_unit)\n        \n        # \u8ba1\u7b97\u5b89\u5168\u5e93\u5b58\u548c\u518d\u8ba2\u8d27\u70b9\n        safety_stock = self.calculate_safety_stock(\n            stats['mean_demand'], \n            stats['std_demand'], \n            self.lead_time\n        )\n        \n        reorder_point = self.calculate_reorder_point(\n            stats['mean_demand'], \n            stats['std_demand'], \n            self.lead_time\n        )\n        \n        # \u5e93\u5b58\u7b56\u7565\n        strategy = {\n            'product_id': product_id,\n            'annual_demand': annual_demand,\n            'mean_daily_demand': stats['mean_demand'],\n            'std_daily_demand': stats['std_demand'],\n            'demand_cv': stats['cv_demand'],\n            'eoq': eoq,\n            'safety_stock': safety_stock,\n            'reorder_point': reorder_point,\n            'max_inventory': eoq + safety_stock,\n            'ordering_cost': ordering_cost,\n            'holding_cost_per_unit': holding_cost_per_unit,\n            'lead_time': self.lead_time,\n            'service_level': self.service_level\n        }\n        \n        return strategy\n    \n    def visualize_inventory_analysis(self, product_id):\n        \"\"\"\u53ef\u89c6\u5316\u5e93\u5b58\u5206\u6790\"\"\"\n        stats, daily_demand = self.calculate_demand_stats(product_id)\n        strategy = self.optimize_inventory(product_id)\n        \n        fig, axes = plt.subplots(2, 2, figsize=(15, 10))\n        \n        # 1. \u9700\u6c42\u5206\u5e03\n        ax1 = axes[0, 0]\n        daily_demand.plot(kind='hist', bins=20, ax=ax1, alpha=0.7, color='blue')\n        ax1.axvline(stats['mean_demand'], color='red', linestyle='--', linewidth=2, label='\u5e73\u5747\u9700\u6c42')\n        ax1.axvline(stats['mean_demand'] + stats['std_demand'], color='orange', linestyle=':', linewidth=2, label='\u00b11\u6807\u51c6\u5dee')\n        ax1.axvline(stats['mean_demand'] - stats['std_demand'], color='orange', linestyle=':', linewidth=2)\n        ax1.set_title(f'{product_id} \u9700\u6c42\u5206\u5e03', fontsize=12, fontweight='bold')\n        ax1.set_xlabel('\u65e5\u9700\u6c42\u91cf')\n        ax1.set_ylabel('\u9891\u6b21')\n        ax1.legend()\n        ax1.grid(True, alpha=0.3)\n        \n        # 2. \u65f6\u95f4\u5e8f\u5217\u9700\u6c42\n        ax2 = axes[0, 1]\n        daily_demand.plot(ax=ax2, color='green', linewidth=1)\n        ax2.axhline(y=stats['mean_demand'], color='red', linestyle='--', linewidth=2, label='\u5e73\u5747\u9700\u6c42')\n        ax2.fill_between(daily_demand.index, \n                        stats['mean_demand'] - stats['std_demand'],\n                        stats['mean_demand'] + stats['std_demand'],\n                        alpha=0.2, color='orange', label='\u6807\u51c6\u5dee\u8303\u56f4')\n        ax2.set_title(f'{product_id} \u9700\u6c42\u65f6\u95f4\u5e8f\u5217', fontsize=12, fontweight='bold')\n        ax2.set_xlabel('\u65e5\u671f')\n        ax2.set_ylabel('\u9700\u6c42\u91cf')\n        ax2.legend()\n        ax2.grid(True, alpha=0.3)\n        \n        # 3. \u5e93\u5b58\u7b56\u7565\u793a\u610f\u56fe\n        ax3 = axes[1, 0]\n        # \u6a21\u62df\u5e93\u5b58\u53d8\u5316\n        days = 100\n        inventory_level = []\n        current_inventory = strategy['max_inventory']\n        demand_series = np.random.normal(stats['mean_demand'], stats['std_demand'], days)\n        \n        for demand in demand_series:\n            current_inventory = max(0, current_inventory - demand)\n            if current_inventory &lt;= strategy['reorder_point']:\n                current_inventory += strategy['eoq']\n            inventory_level.append(current_inventory)\n        \n        ax3.plot(range(days), inventory_level, color='blue', linewidth=2, label='\u5e93\u5b58\u6c34\u5e73')\n        ax3.axhline(y=strategy['reorder_point'], color='red', linestyle='--', \n                   linewidth=2, label='\u518d\u8ba2\u8d27\u70b9')\n        ax3.axhline(y=strategy['safety_stock'], color='orange', linestyle=':', \n                   linewidth=2, label='\u5b89\u5168\u5e93\u5b58')\n        ax3.set_title('\u5e93\u5b58\u7b56\u7565\u6a21\u62df', fontsize=12, fontweight='bold')\n        ax3.set_xlabel('\u5929\u6570')\n        ax3.set_ylabel('\u5e93\u5b58\u91cf')\n        ax3.legend()\n        ax3.grid(True, alpha=0.3)\n        \n        # 4. \u6210\u672c\u5206\u6790\n        ax4 = axes[1, 1]\n        order_quantities = np.linspace(strategy['eoq'] * 0.5, strategy['eoq'] * 2, 50)\n        ordering_costs = (annual_demand \/ order_quantities) * strategy['ordering_cost']\n        holding_costs = (order_quantities \/ 2) * strategy['holding_cost_per_unit']\n        total_costs = ordering_costs + holding_costs\n        \n        ax4.plot(order_quantities, ordering_costs, label='\u8ba2\u8d27\u6210\u672c', color='blue')\n        ax4.plot(order_quantities, holding_costs, label='\u6301\u6709\u6210\u672c', color='green')\n        ax4.plot(order_quantities, total_costs, label='\u603b\u6210\u672c', color='red', linewidth=2)\n        ax4.axvline(x=strategy['eoq'], color='black', linestyle='--', \n                   linewidth=2, label='EOQ')\n        ax4.set_title('\u6210\u672c\u5206\u6790', fontsize=12, fontweight='bold')\n        ax4.set_xlabel('\u8ba2\u8d27\u6279\u91cf')\n        ax4.set_ylabel('\u6210\u672c')\n        ax4.legend()\n        ax4.grid(True, alpha=0.3)\n        \n        plt.tight_layout()\n        plt.show()\n        \n        return strategy\n\n# \u6848\u4f8b\u6570\u636e\u811a\u672c\ndef create_forecast_sample_data():\n    \"\"\"\u751f\u6210\u9884\u6d4b\u5206\u6790\u793a\u4f8b\u6570\u636e\"\"\"\n    np.random.seed(42)\n    \n    # \u751f\u62102\u5e74\u7684\u65e5\u9500\u552e\u6570\u636e\n    dates = pd.date_range('2022-01-01', '2023-12-31', freq='D')\n    n_days = len(dates)\n    \n    # \u57fa\u7840\u8d8b\u52bf + \u5b63\u8282\u6027 + \u968f\u673a\u566a\u58f0\n    trend = np.linspace(1000, 5000, n_days)\n    yearly_seasonality = 1000 * np.sin(2 * np.pi * np.arange(n_days) \/ 365)\n    weekly_seasonality = 500 * np.sin(2 * np.pi * np.arange(n_days) \/ 7)\n    noise = np.random.normal(0, 200, n_days)\n    \n    sales = trend + yearly_seasonality + weekly_seasonality + noise\n    sales = np.maximum(sales, 0)  # \u786e\u4fdd\u975e\u8d1f\n    \n    # \u521b\u5efaDataFrame\n    sales_data = pd.DataFrame({\n        'date': dates,\n        'sales': sales.round(2),\n        'quantity': (sales \/ 100).astype(int) + np.random.randint(1, 10, n_days)\n    })\n    \n    # \u6dfb\u52a0\u4ea7\u54c1\u4fe1\u606f\n    products = ['P001', 'P002', 'P003', 'P004', 'P005']\n    product_sales = []\n    \n    for date, total_sales, total_qty in zip(dates, sales, sales_data['quantity']):\n        for product in products:\n            # \u5206\u914d\u9500\u552e\u5230\u4e0d\u540c\u4ea7\u54c1\n            product_share = np.random.dirichlet(np.ones(len(products)))\n            product_sales.append({\n                'date': date,\n                'product_id': product,\n                'sales': total_sales * product_share[products.index(product)],\n                'quantity': max(1, int(total_qty * product_share[products.index(product)]))\n            })\n    \n    product_sales_df = pd.DataFrame(product_sales)\n    \n    # \u4fdd\u5b58\u6570\u636e\n    sales_data.to_csv('daily_sales_forecast.csv', index=False)\n    product_sales_df.to_csv('product_sales_detail.csv', index=False)\n    \n    print(\"\u2705 \u9884\u6d4b\u793a\u4f8b\u6570\u636e\u5df2\u751f\u6210:\")\n    print(f\"  \u65e5\u9500\u552e\u6570\u636e: {sales_data.shape}\")\n    print(f\"  \u4ea7\u54c1\u660e\u7ec6\u6570\u636e: {product_sales_df.shape}\")\n    \n    return sales_data, product_sales_df\n\n# \u81ea\u52a8\u62a5\u8868\u751f\u6210\nclass AutoReportGenerator:\n    def __init__(self, forecast_results, inventory_strategies):\n        self.forecast_results = forecast_results\n        self.inventory_strategies = inventory_strategies\n    \n    def generate_html_report(self, filename='sales_forecast_report.html'):\n        \"\"\"\u751f\u6210HTML\u683c\u5f0f\u62a5\u544a\"\"\"\n        html_content = \"\"\"\n        &lt;!DOCTYPE html&gt;\n        &lt;html&gt;\n        &lt;head&gt;\n            &lt;meta charset=\"UTF-8\"&gt;\n            &lt;title&gt;\u7535\u5546\u9500\u552e\u9884\u6d4b\u4e0e\u5e93\u5b58\u4f18\u5316\u62a5\u544a&lt;\/title&gt;\n            &lt;style&gt;\n                body { font-family: Arial, sans-serif; margin: 40px; }\n                .header { text-align: center; padding: 20px; background: #f0f0f0; }\n                .section { margin: 30px 0; padding: 20px; border: 1px solid #ddd; }\n                .metric { display: inline-block; margin: 10px 20px; padding: 10px; background: #e8f4f8; }\n                table { width: 100%; border-collapse: collapse; margin: 20px 0; }\n                th, td { border: 1px solid #ddd; padding: 8px; text-align: left; }\n                th { background-color: #f2f2f2; }\n                .forecast { background: #f9f9f9; padding: 15px; }\n                .inventory { background: #f0f8ff; padding: 15px; }\n            &lt;\/style&gt;\n        &lt;\/head&gt;\n        &lt;body&gt;\n            &lt;div class=\"header\"&gt;\n                &lt;h1&gt;\ud83d\udcca \u7535\u5546\u9500\u552e\u9884\u6d4b\u4e0e\u5e93\u5b58\u4f18\u5316\u62a5\u544a&lt;\/h1&gt;\n                &lt;p&gt;\u751f\u6210\u65f6\u95f4: \"\"\" + datetime.now().strftime('%Y-%m-%d %H:%M:%S') + \"\"\"&lt;\/p&gt;\n            &lt;\/div&gt;\n        \"\"\"\n        \n        # \u6dfb\u52a0\u9884\u6d4b\u7ed3\u679c\n        html_content += \"\"\"\n            &lt;div class=\"section\"&gt;\n                &lt;h2&gt;\ud83d\udcc8 \u9500\u552e\u9884\u6d4b\u7ed3\u679c&lt;\/h2&gt;\n                &lt;div class=\"forecast\"&gt;\n        \"\"\"\n        \n        for model_name, result in self.forecast_results.items():\n            if 'forecast' in result:\n                forecast_df = result['forecast']\n                html_content += f\"\"\"\n                    &lt;h3&gt;{model_name} \u9884\u6d4b&lt;\/h3&gt;\n                    &lt;p&gt;\u672a\u676530\u5929\u9884\u6d4b\u9500\u552e\u989d: \u00a5{forecast_df['forecast'].sum():,.2f}&lt;\/p&gt;\n                    &lt;p&gt;\u65e5\u5747\u9884\u6d4b: \u00a5{forecast_df['forecast'].mean():,.2f}&lt;\/p&gt;\n                \"\"\"\n        \n        html_content += \"\"\"\n                &lt;\/div&gt;\n            &lt;\/div&gt;\n        \"\"\"\n        \n        # \u6dfb\u52a0\u5e93\u5b58\u7b56\u7565\n        html_content += \"\"\"\n            &lt;div class=\"section\"&gt;\n                &lt;h2&gt;\ud83d\udce6 \u5e93\u5b58\u4f18\u5316\u7b56\u7565&lt;\/h2&gt;\n                &lt;div class=\"inventory\"&gt;\n                    &lt;table&gt;\n                        &lt;tr&gt;\n                            &lt;th&gt;\u4ea7\u54c1ID&lt;\/th&gt;\n                            &lt;th&gt;\u5e74\u9700\u6c42\u91cf&lt;\/th&gt;\n                            &lt;th&gt;EOQ&lt;\/th&gt;\n                            &lt;th&gt;\u5b89\u5168\u5e93\u5b58&lt;\/th&gt;\n                            &lt;th&gt;\u518d\u8ba2\u8d27\u70b9&lt;\/th&gt;\n                            &lt;th&gt;\u6700\u5927\u5e93\u5b58&lt;\/th&gt;\n                        &lt;\/tr&gt;\n        \"\"\"\n        \n        for strategy in self.inventory_strategies:\n            html_content += f\"\"\"\n                        &lt;tr&gt;\n                            &lt;td&gt;{strategy['product_id']}&lt;\/td&gt;\n                            &lt;td&gt;{strategy['annual_demand']:,.0f}&lt;\/td&gt;\n                            &lt;td&gt;{strategy['eoq']:,.1f}&lt;\/td&gt;\n                            &lt;td&gt;{strategy['safety_stock']:,.1f}&lt;\/td&gt;\n                            &lt;td&gt;{strategy['reorder_point']:,.1f}&lt;\/td&gt;\n                            &lt;td&gt;{strategy['max_inventory']:,.1f}&lt;\/td&gt;\n                        &lt;\/tr&gt;\n            \"\"\"\n        \n        html_content += \"\"\"\n                    &lt;\/table&gt;\n                &lt;\/div&gt;\n            &lt;\/div&gt;\n            \n            &lt;div class=\"section\"&gt;\n                &lt;h2&gt;\ud83d\udca1 \u5efa\u8bae\u4e0e\u884c\u52a8\u8ba1\u5212&lt;\/h2&gt;\n                &lt;ol&gt;\n                    &lt;li&gt;&lt;strong&gt;\u91c7\u8d2d\u5efa\u8bae\uff1a&lt;\/strong&gt;\u6839\u636eEOQ\u548c\u5b89\u5168\u5e93\u5b58\u5236\u5b9a\u91c7\u8d2d\u8ba1\u5212&lt;\/li&gt;\n                    &lt;li&gt;&lt;strong&gt;\u5e93\u5b58\u76d1\u63a7\uff1a&lt;\/strong&gt;\u8bbe\u7f6e\u518d\u8ba2\u8d27\u70b9\u9884\u8b66\u673a\u5236&lt;\/li&gt;\n                    &lt;li&gt;&lt;strong&gt;\u9500\u552e\u9884\u6d4b\uff1a&lt;\/strong&gt;\u7ed3\u5408\u5b63\u8282\u6027\u8c03\u6574\u9500\u552e\u76ee\u6807&lt;\/li&gt;\n                    &lt;li&gt;&lt;strong&gt;\u6210\u672c\u4f18\u5316\uff1a&lt;\/strong&gt;\u5b9a\u671f\u8bc4\u4f30\u8ba2\u8d27\u6210\u672c\u548c\u6301\u6709\u6210\u672c&lt;\/li&gt;\n                &lt;\/ol&gt;\n            &lt;\/div&gt;\n        &lt;\/body&gt;\n        &lt;\/html&gt;\n        \"\"\"\n        \n        with open(filename, 'w', encoding='utf-8') as f:\n            f.write(html_content)\n        \n        print(f\"\u2705 HTML\u62a5\u544a\u5df2\u751f\u6210: {filename}\")\n        return filename\n    \n    def generate_excel_report(self, filename='sales_forecast_report.xlsx'):\n        \"\"\"\u751f\u6210Excel\u683c\u5f0f\u62a5\u544a\"\"\"\n        with pd.ExcelWriter(filename, engine='openpyxl') as writer:\n            # \u9884\u6d4b\u7ed3\u679c\n            for model_name, result in self.forecast_results.items():\n                if 'forecast' in result:\n                    result['forecast'].to_excel(writer, sheet_name=f'{model_name}_\u9884\u6d4b', index=False)\n            \n            # \u5e93\u5b58\u7b56\u7565\n            strategies_df = pd.DataFrame(self.inventory_strategies)\n            strategies_df.to_excel(writer, sheet_name='\u5e93\u5b58\u7b56\u7565', index=False)\n            \n            # \u6c47\u603b\u8868\n            summary_data = {\n                '\u6307\u6807': ['\u603b\u9884\u6d4b\u9500\u552e\u989d', '\u5e73\u5747EOQ', '\u5e73\u5747\u5b89\u5168\u5e93\u5b58', '\u5e73\u5747\u518d\u8ba2\u8d27\u70b9'],\n                '\u6570\u503c': [\n                    sum(r['forecast']['forecast'].sum() for r in self.forecast_results.values() if 'forecast' in r),\n                    strategies_df['eoq'].mean(),\n                    strategies_df['safety_stock'].mean(),\n                    strategies_df['reorder_point'].mean()\n                ],\n                '\u5355\u4f4d': ['\u5143', '\u4ef6', '\u4ef6', '\u4ef6']\n            }\n            summary_df = pd.DataFrame(summary_data)\n            summary_df.to_excel(writer, sheet_name='\u6c47\u603b', index=False)\n        \n        print(f\"\u2705 Excel\u62a5\u544a\u5df2\u751f\u6210: {filename}\")\n        return filename\n\n# \u8fd0\u884c\u793a\u4f8b\nif __name__ == \"__main__\":\n    print(\"=\"*60)\n    print(\"Day 6: \u9500\u552e\u9884\u6d4b\u4e0e\u5e93\u5b58\u4f18\u5316\")\n    print(\"=\"*60)\n    \n    # 1. \u751f\u6210\u793a\u4f8b\u6570\u636e\n    print(\"\\n\u6b65\u9aa41: \u751f\u6210\u9884\u6d4b\u5206\u6790\u793a\u4f8b\u6570\u636e...\")\n    sales_data, product_sales = create_forecast_sample_data()\n    \n    # 2. \u9500\u552e\u9884\u6d4b\n    print(\"\\n\u6b65\u9aa42: \u9500\u552e\u9884\u6d4b\u5206\u6790...\")\n    forecaster = SalesForecaster(sales_data)\n    \n    # \u51c6\u5907\u65f6\u95f4\u5e8f\u5217\n    ts_data = forecaster.prepare_time_series(freq='D')\n    print(f\"\u65f6\u95f4\u5e8f\u5217\u6570\u636e\u5f62\u72b6: {ts_data.shape}\")\n    \n    # \u591a\u79cd\u9884\u6d4b\u65b9\u6cd5\n    print(\"\\n\u4f7f\u7528\u591a\u79cd\u65b9\u6cd5\u8fdb\u884c\u9884\u6d4b:\")\n    \n    # \u79fb\u52a8\u5e73\u5747\n    ma_forecast = forecaster.moving_average_forecast(ts_data, window=30, forecast_days=30)\n    print(f\"  \u2713 \u79fb\u52a8\u5e73\u5747\u9884\u6d4b\u5b8c\u6210\")\n    \n    # \u6307\u6570\u5e73\u6ed1\n    es_forecast = forecaster.exponential_smoothing_forecast(ts_data, alpha=0.3, forecast_days=30)\n    print(f\"  \u2713 \u6307\u6570\u5e73\u6ed1\u9884\u6d4b\u5b8c\u6210\")\n    \n    # ARIMA\n    arima_forecast = forecaster.arima_forecast(ts_data, order=(1,1,1), forecast_days=30)\n    if arima_forecast is not None:\n        print(f\"  \u2713 ARIMA\u9884\u6d4b\u5b8c\u6210\")\n    \n    # Prophet\uff08\u53ef\u9009\uff09\n    prophet_forecast = forecaster.prophet_forecast(ts_data, forecast_days=30)\n    if prophet_forecast is not None:\n        print(f\"  \u2713 Prophet\u9884\u6d4b\u5b8c\u6210\")\n    \n    # 3. \u53ef\u89c6\u5316\u9884\u6d4b\u7ed3\u679c\n    print(\"\\n\u6b65\u9aa43: \u751f\u6210\u9884\u6d4b\u53ef\u89c6\u5316...\")\n    forecaster.visualize_forecasts(ts_data, forecast_days=30)\n    \n    # 4. \u5e93\u5b58\u4f18\u5316\n    print(\"\\n\u6b65\u9aa44: \u5e93\u5b58\u4f18\u5316\u5206\u6790...\")\n    optimizer = InventoryOptimizer(product_sales, lead_time=7, service_level=0.95)\n    \n    # \u5206\u6790\u524d5\u4e2a\u4ea7\u54c1\n    products_to_analyze = ['P001', 'P002', 'P003', 'P004', 'P005']\n    inventory_strategies = []\n    \n    for product_id in products_to_analyze:\n        print(f\"\\n\u5206\u6790\u4ea7\u54c1: {product_id}\")\n        strategy = optimizer.optimize_inventory(\n            product_id, \n            ordering_cost=50, \n            holding_cost_rate=0.2, \n            unit_cost=100\n        )\n        inventory_strategies.append(strategy)\n        \n        print(f\"  \u5e74\u9700\u6c42\u91cf: {strategy['annual_demand']:,.0f}\")\n        print(f\"  EOQ: {strategy['eoq']:,.1f}\")\n        print(f\"  \u5b89\u5168\u5e93\u5b58: {strategy['safety_stock']:,.1f}\")\n        print(f\"  \u518d\u8ba2\u8d27\u70b9: {strategy['reorder_point']:,.1f}\")\n    \n    # 5. \u53ef\u89c6\u5316\u5e93\u5b58\u5206\u6790\n    print(\"\\n\u6b65\u9aa45: \u751f\u6210\u5e93\u5b58\u5206\u6790\u53ef\u89c6\u5316...\")\n    optimizer.visualize_inventory_analysis('P001')\n    \n    # 6. \u751f\u6210\u81ea\u52a8\u62a5\u8868\n    print(\"\\n\u6b65\u9aa46: \u751f\u6210\u81ea\u52a8\u62a5\u8868...\")\n    report_generator = AutoReportGenerator(\n        forecaster.forecast_results,\n        inventory_strategies\n    )\n    \n    # \u751f\u6210HTML\u62a5\u544a\n    html_report = report_generator.generate_html_report()\n    \n    # \u751f\u6210Excel\u62a5\u544a\n    excel_report = report_generator.generate_excel_report()\n    \n    print(\"\\n\" + \"=\"*60)\n    print(\"\u2705 \u5206\u6790\u5b8c\u6210\uff01\")\n    print(\"=\"*60)\n    print(\"\u751f\u6210\u7684\u6587\u4ef6:\")\n    print(f\"  1. HTML\u62a5\u544a: {html_report}\")\n    print(f\"  2. Excel\u62a5\u544a: {excel_report}\")\n    print(f\"  3. \u539f\u59cb\u6570\u636e: daily_sales_forecast.csv\")\n    print(f\"  4. \u4ea7\u54c1\u660e\u7ec6: product_sales_detail.csv\")\n    print(\"\\n\u5efa\u8bae\u4e0b\u4e00\u6b65:\")\n    print(\"  1. \u6839\u636e\u9884\u6d4b\u7ed3\u679c\u8c03\u6574\u9500\u552e\u76ee\u6807\")\n    print(\"  2. \u6309\u7167\u5e93\u5b58\u7b56\u7565\u4f18\u5316\u91c7\u8d2d\u8ba1\u5212\")\n    print(\"  3. \u8bbe\u7f6e\u5e93\u5b58\u9884\u8b66\u76d1\u63a7\")\n<\/code><\/pre>\n\n\n\n<h3 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"%E5%AE%89%E8%A3%85%E8%A6%81%E6%B1%82\"><\/span><strong>\u5b89\u88c5\u8981\u6c42<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h3>\n\n\n\n<pre class=\"wp-block-preformatted\">bash<code># \u57fa\u7840\u4f9d\u8d56\npip install pandas==1.5.3 numpy==1.24.3 matplotlib==3.7.1 seaborn==0.12.2\n\n# \u65f6\u95f4\u5e8f\u5217\u5206\u6790\npip install statsmodels==0.14.0 scikit-learn==1.3.0\n\n# Prophet\u9884\u6d4b\uff08\u53ef\u9009\uff09\npip install prophet==1.1.4\n\n# Excel\u62a5\u544a\u751f\u6210\npip install openpyxl==3.1.2 xlsxwriter==3.1.2\n\n# \u79d1\u5b66\u8ba1\u7b97\npip install scipy==1.11.4\n<\/code><\/pre>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"%E8%AF%BE%E7%A8%8B%E6%80%BB%E7%BB%93\"><\/span><strong>\u8bfe\u7a0b\u603b\u7ed3<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<h3 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"4-6%E5%A4%A9%E6%8A%80%E8%83%BD%E6%8E%8C%E6%8F%A1%E6%B8%85%E5%8D%95\"><\/span><strong>4-6\u5929\u6280\u80fd\u638c\u63e1\u6e05\u5355<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h3>\n\n\n\n<ol>\n<li><strong>Day 4\uff1a\u6570\u636e\u91c7\u96c6\u4e0e\u6e05\u6d17<\/strong>\n<ul>\n<li>\u2705 \u591a\u6e90\u6570\u636e\u81ea\u52a8\u91c7\u96c6\uff08Excel\u3001CSV\u3001MySQL\uff09<\/li>\n\n\n\n<li>\u2705 \u81ea\u52a8\u5316\u6570\u636e\u6e05\u6d17\uff08\u7f3a\u5931\u503c\u3001\u5f02\u5e38\u503c\u3001\u91cd\u590d\u503c\uff09<\/li>\n\n\n\n<li>\u2705 \u6570\u636e\u7c7b\u578b\u6807\u51c6\u5316\u4e0e\u8f6c\u6362<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>Day 5\uff1a\u7535\u5546\u6838\u5fc3\u6307\u6807\u4e0eRFM\u5206\u6790<\/strong>\n<ul>\n<li>\u2705 \u7535\u5546\u6838\u5fc3KPI\u8ba1\u7b97\uff08GMV\u3001\u8f6c\u5316\u7387\u3001\u5ba2\u5355\u4ef7\u7b49\uff09<\/li>\n\n\n\n<li>\u2705 RFM\u7528\u6237\u4ef7\u503c\u5206\u5c42\u6a21\u578b<\/li>\n\n\n\n<li>\u2705 \u9500\u552e\u6f0f\u6597\u5206\u6790\u4e0e\u53ef\u89c6\u5316<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>Day 6\uff1a\u9500\u552e\u9884\u6d4b\u4e0e\u5e93\u5b58\u4f18\u5316<\/strong>\n<ul>\n<li>\u2705 \u65f6\u95f4\u5e8f\u5217\u9884\u6d4b\uff08\u79fb\u52a8\u5e73\u5747\u3001\u6307\u6570\u5e73\u6ed1\u3001ARIMA\uff09<\/li>\n\n\n\n<li>\u2705 \u5e93\u5b58\u4f18\u5316\u6a21\u578b\uff08EOQ\u3001\u5b89\u5168\u5e93\u5b58\u3001\u518d\u8ba2\u8d27\u70b9\uff09<\/li>\n\n\n\n<li>\u2705 \u81ea\u52a8\u5316\u62a5\u8868\u751f\u6210\uff08HTML\u3001Excel\uff09<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n\n\n\n<h3 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"%E8%BF%90%E8%A1%8C%E8%AF%B4%E6%98%8E\"><\/span><strong>\u8fd0\u884c\u8bf4\u660e<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h3>\n\n\n\n<ol>\n<li><strong>\u73af\u5883\u51c6\u5907<\/strong>\uff1a bash<code># \u521b\u5efa\u865a\u62df\u73af\u5883 python -m venv ecommerce_env source ecommerce_env\/bin\/activate # Linux\/Mac # \u6216 ecommerce_env\\Scripts\\activate # Windows # \u5b89\u88c5\u4f9d\u8d56 pip install -r requirements.txt<\/code><\/li>\n\n\n\n<li><strong>\u8fd0\u884c\u987a\u5e8f<\/strong>\uff1a\n<ul>\n<li>\u6309Day 4 \u2192 Day 5 \u2192 Day 6\u987a\u5e8f\u8fd0\u884c<\/li>\n\n\n\n<li>\u6bcf\u4e2a\u811a\u672c\u90fd\u662f\u72ec\u7acb\u7684\uff0c\u5305\u542b\u793a\u4f8b\u6570\u636e\u751f\u6210<\/li>\n\n\n\n<li>\u6240\u6709\u8f93\u51fa\u6587\u4ef6\u4f1a\u81ea\u52a8\u4fdd\u5b58\u5230\u5f53\u524d\u76ee\u5f55<\/li>\n<\/ul>\n<\/li>\n\n\n\n<li><strong>\u81ea\u5b9a\u4e49\u6570\u636e<\/strong>\uff1a\n<ul>\n<li>\u66ff\u6362\u793a\u4f8b\u6570\u636e\u6587\u4ef6\u4e3a\u81ea\u5df1\u7684\u7535\u5546\u6570\u636e<\/li>\n\n\n\n<li>\u8c03\u6574\u53c2\u6570\u4ee5\u9002\u5e94\u5b9e\u9645\u4e1a\u52a1\u573a\u666f<\/li>\n\n\n\n<li>\u6269\u5c55\u6a21\u578b\u4ee5\u5904\u7406\u66f4\u590d\u6742\u7684\u9700\u6c42<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n\n\n\n<h3 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"%E4%B8%9A%E5%8A%A1%E5%BA%94%E7%94%A8%E4%BB%B7%E5%80%BC\"><\/span><strong>\u4e1a\u52a1\u5e94\u7528\u4ef7\u503c<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h3>\n\n\n\n<ol>\n<li><strong>\u6570\u636e\u9a71\u52a8\u51b3\u7b56<\/strong>\uff1a\u57fa\u4e8e\u6570\u636e\u5206\u6790\u5236\u5b9a\u9500\u552e\u7b56\u7565<\/li>\n\n\n\n<li><strong>\u5e93\u5b58\u4f18\u5316<\/strong>\uff1a\u964d\u4f4e\u5e93\u5b58\u6210\u672c\uff0c\u63d0\u9ad8\u5468\u8f6c\u7387<\/li>\n\n\n\n<li><strong>\u7528\u6237\u7cbe\u7ec6\u5316\u8fd0\u8425<\/strong>\uff1aRFM\u5206\u7fa4\u5b9e\u73b0\u7cbe\u51c6\u8425\u9500<\/li>\n\n\n\n<li><strong>\u81ea\u52a8\u5316\u62a5\u8868<\/strong>\uff1a\u51cf\u5c11\u4eba\u5de5\u5de5\u4f5c\u91cf\uff0c\u63d0\u9ad8\u6548\u7387<\/li>\n<\/ol>\n\n\n\n<p>\u8fd9\u5957\u8bfe\u7a0b\u4f53\u7cfb\u4ece\u6570\u636e\u91c7\u96c6\u5230\u9884\u6d4b\u4f18\u5316\uff0c\u5f62\u6210\u4e86\u5b8c\u6574\u7684\u7535\u5546\u6570\u636e\u5206\u6790\u95ed\u73af\uff0c\u5b66\u5458\u901a\u8fc74-6\u5929\u7684\u5b66\u4e60\u53ef\u4ee5\u638c\u63e1\u7535\u5546\u5168\u94fe\u8def\u6570\u636e\u5206\u6790\u7684\u6838\u5fc3\u6280\u80fd\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u8bfe\u7a0b\u76ee\u6807 \u638c\u63e1\u65f6\u95f4\u5e8f\u5217\u9884\u6d4b\u3001\u5e93\u5b58\u4f18\u5316\u6a21\u578b\u3001\u81ea\u52a8\u62a5\u8868\u751f\u6210\u3002 \u6838\u5fc3\u6280\u80fd \u8bfe\u4ef6\u5185\u5bb9 1. \u9500\u552e\u9884\u6d4b\u6a21\u578b p&hellip; <a href=\"http:\/\/viplao.com\/index.php\/2026\/03\/01\/%e6%95%b0%e6%8d%ae%e6%8a%80%e8%83%bd%ef%bc%88skills%ef%bc%89-%ef%bc%9aday-6%ef%bc%9a%e9%94%80%e5%94%ae%e9%a2%84%e6%b5%8b%e4%b8%8e%e5%ba%93%e5%ad%98%e4%bc%98%e5%8c%96\/\" class=\"more-link read-more\" rel=\"bookmark\">\u7ee7\u7eed\u9605\u8bfb <span class=\"screen-reader-text\">\u6570\u636e\u6280\u80fd\uff08SKILLS\uff09 \uff1aDay 6\uff1a\u9500\u552e\u9884\u6d4b\u4e0e\u5e93\u5b58\u4f18\u5316<\/span><i class=\"fa fa-arrow-right\"><\/i><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[28],"views":178,"_links":{"self":[{"href":"http:\/\/viplao.com\/index.php\/wp-json\/wp\/v2\/posts\/4225"}],"collection":[{"href":"http:\/\/viplao.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/viplao.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/viplao.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/viplao.com\/index.php\/wp-json\/wp\/v2\/comments?post=4225"}],"version-history":[{"count":1,"href":"http:\/\/viplao.com\/index.php\/wp-json\/wp\/v2\/posts\/4225\/revisions"}],"predecessor-version":[{"id":4226,"href":"http:\/\/viplao.com\/index.php\/wp-json\/wp\/v2\/posts\/4225\/revisions\/4226"}],"wp:attachment":[{"href":"http:\/\/viplao.com\/index.php\/wp-json\/wp\/v2\/media?parent=4225"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/viplao.com\/index.php\/wp-json\/wp\/v2\/categories?post=4225"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/viplao.com\/index.php\/wp-json\/wp\/v2\/tags?post=4225"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}