# pip install pandas numpy matplotlib seaborn scipy statsmodelsimport osimport numpy as npimport pandas as pdimport matplotlib.pyplot as pltimport seaborn as snsfrom datetime import datetime, timedeltaimport warningswarnings.filterwarnings('ignore')# 设置中文字体plt.rcParams['font.sans-serif'] = ['SimHei'] # 用来正常显示中文标签plt.rcParams['axes.unicode_minus'] = False # 用来正常显示负号# 获取桌面路径并创建结果文件夹desktop_path = os.path.join(os.path.expanduser("~"), "Desktop")results_path = os.path.join(desktop_path, "Results时间")# 创建结果文件夹if not os.path.exists(results_path): os.makedirs(results_path)def generate_weather_disease_data():"""生成气象和疾病数据"""print("开始生成数据...")
# 温度 (°C) temperature = (10 + 15 * np.sin(2 * np.pi * (day_of_year - 80) / 365) + 8 * np.sin(2 * np.pi * (hour - 6) / 24) + np.random.normal(0, 3, n))# 相对湿度 (%) humidity = (60 + 20 * np.sin(2 * np.pi * (day_of_year - 100) / 365) - 0.5 * (temperature - 15) + np.random.normal(0, 10, n)) humidity = np.clip(humidity, 20, 100)# 风速 (m/s) wind_speed = (3 + 2 * np.sin(2 * np.pi * day_of_year / 365) + np.abs(np.random.normal(0, 1.5, n)))# 气压 (hPa) pressure = (1013 + 10 * np.sin(2 * np.pi * (day_of_year - 180) / 365) + np.random.normal(0, 2, n))# 降水量 (mm/h) precipitation = np.where(np.random.uniform(0, 1, n) < 0.1, np.random.exponential(2, n), 0)# 日照时数 sunshine = np.where((hour >= 6) & (hour <= 18) & (precipitation == 0) & (np.random.uniform(0, 1, n) > 0.3), 1, 0)# 添加气象变量到DataFrame hourly_weather['temperature'] = temperature hourly_weather['humidity'] = humidity hourly_weather['wind_speed'] = wind_speed hourly_weather['pressure'] = pressure hourly_weather['precipitation'] = precipitation hourly_weather['sunshine'] = sunshine# 保存逐小时气象数据 hourly_weather.to_csv(os.path.join(results_path, 'hourly_weather_data.csv'), index=False)print(f"逐小时气象数据已保存到: {os.path.join(results_path, 'hourly_weather_data.csv')}")# 2. 按日汇总气象数据print("正在生成逐日气象汇总数据...") hourly_weather['date'] = pd.to_datetime(hourly_weather['timestamp'].dt.date) daily_weather = hourly_weather.groupby('date').agg({'temperature': ['max', 'min', 'mean', 'median'],'humidity': ['max', 'min', 'mean', 'median'],'wind_speed': ['max', 'mean'],'pressure': ['max', 'min', 'mean'],'precipitation': 'sum','sunshine': 'sum' }).reset_index()# 重命名列 daily_weather.columns = ['date', 'temp_max', 'temp_min', 'temp_mean', 'temp_median','humidity_max', 'humidity_min', 'humidity_mean', 'humidity_median','wind_max', 'wind_mean', 'pressure_max', 'pressure_min', 'pressure_mean','precipitation_total', 'sunshine_hours']# 添加时间特征 daily_weather['year'] = daily_weather['date'].dt.year daily_weather['month'] = daily_weather['date'].dt.month daily_weather['day'] = daily_weather['date'].dt.day daily_weather['day_of_year'] = daily_weather['date'].dt.dayofyear daily_weather['week'] = daily_weather['date'].dt.isocalendar().week# 保存逐日气象汇总数据 daily_weather.to_csv(os.path.join(results_path, 'daily_weather_summary.csv'), index=False)print(f"逐日气象汇总数据已保存到: {os.path.join(results_path, 'daily_weather_summary.csv')}")# 3. 生成逐日疾病数据print("正在生成逐日疾病数据...")# 创建日期序列 date_seq = pd.date_range(start='1981-01-01', end='2025-12-31', freq='D') n_days = len(date_seq)# 生成基础疾病数据框架 daily_diseases = pd.DataFrame({'date': date_seq,'year': date_seq.year,'month': date_seq.month,'day': date_seq.day,'day_of_year': date_seq.dayofyear,'week': date_seq.isocalendar().week })# 生成呼吸道疾病数据 year_norm = daily_diseases['year'].values - 2000 day_of_year = daily_diseases['day_of_year'].values# 流感 influenza = np.exp(3 + 1.5 * np.sin(2 * np.pi * (day_of_year - 10) / 365) + 0.3 * np.sin(2 * np.pi * (day_of_year - 200) / 365) + 0.1 * year_norm + np.random.normal(0, 0.3, n_days)) * \ (1 + 0.2 * np.sin(2 * np.pi * daily_diseases['year'].values / 3))# 普通感冒 common_cold = np.exp(2.5 + 1.2 * np.sin(2 * np.pi * (day_of_year - 90) / 365) + 1.0 * np.sin(2 * np.pi * (day_of_year - 270) / 365) + np.random.normal(0, 0.4, n_days))# 肺炎 pneumonia = np.exp(2 + 1.8 * np.sin(2 * np.pi * (day_of_year - 15) / 365) + 0.05 * (daily_diseases['year'].values - 1981) + np.random.normal(0, 0.35, n_days)) daily_diseases['influenza'] = np.round(influenza) daily_diseases['common_cold'] = np.round(common_cold) daily_diseases['pneumonia'] = np.round(pneumonia)# 生成肠道传染病数据 year_from_1981 = daily_diseases['year'].values - 1981# 细菌性痢疾 bacillary_dysentery = np.exp(2 + 2.0 * np.sin(2 * np.pi * (day_of_year - 200) / 365) + 0.02 * year_from_1981 + np.random.normal(0, 0.5, n_days)) * \ (1 - 0.005 * year_from_1981)# 手足口病 hand_foot_mouth = np.exp(2.2 + 1.5 * np.sin(2 * np.pi * (day_of_year - 150) / 365) + 0.8 * np.sin(2 * np.pi * (day_of_year - 330) / 365) + np.random.normal(0, 0.4, n_days))# 感染性腹泻 infectious_diarrhea = np.exp(2.8 + 1.0 * np.sin(2 * np.pi * (day_of_year - 180) / 365) + np.random.normal(0, 0.3, n_days)) daily_diseases['bacillary_dysentery'] = np.round(bacillary_dysentery) daily_diseases['hand_foot_mouth'] = np.round(hand_foot_mouth) daily_diseases['infectious_diarrhea'] = np.round(infectious_diarrhea)# 生成自然疫源性疾病数据# 流行性出血热 hemorrhagic_fever = np.exp(1.5 + 1.2 * np.sin(2 * np.pi * (day_of_year - 80) / 365) + 1.0 * np.sin(2 * np.pi * (day_of_year - 280) / 365) + 0.01 * year_from_1981 + np.random.normal(0, 0.6, n_days))# 莱姆病 lyme_disease = np.exp(1.2 + 1.5 * np.sin(2 * np.pi * (day_of_year - 180) / 365) + np.random.normal(0, 0.5, n_days))# 布鲁氏菌病 brucellosis = np.exp(1.8 + 1.0 * np.sin(2 * np.pi * (day_of_year - 100) / 365) + 0.8 * np.sin(2 * np.pi * (day_of_year - 300) / 365) + np.random.normal(0, 0.4, n_days)) daily_diseases['hemorrhagic_fever'] = np.round(hemorrhagic_fever) daily_diseases['lyme_disease'] = np.round(lyme_disease) daily_diseases['brucellosis'] = np.round(brucellosis)# 确保所有疾病计数为非负整数 disease_cols = ['influenza', 'common_cold', 'pneumonia','bacillary_dysentery', 'hand_foot_mouth', 'infectious_diarrhea','hemorrhagic_fever', 'lyme_disease', 'brucellosis']for col in disease_cols: daily_diseases[col] = np.maximum(daily_diseases[col], 0)# 保存逐日疾病数据 daily_diseases.to_csv(os.path.join(results_path, 'daily_disease_data.csv'), index=False)print(f"逐日疾病数据已保存到: {os.path.join(results_path, 'daily_disease_data.csv')}")# 4. 合并气象和疾病数据print("正在合并气象和疾病数据...") combined_data = daily_weather.merge(daily_diseases, on='date', suffixes=('', '_y'))# 选择需要的列并重命名 combined_data = combined_data[['date', 'year', 'month', 'day', 'day_of_year', 'week','temp_max', 'temp_min', 'temp_mean', 'temp_median','humidity_max', 'humidity_min', 'humidity_mean', 'humidity_median','wind_max', 'wind_mean', 'pressure_max', 'pressure_min', 'pressure_mean','precipitation_total', 'sunshine_hours','influenza', 'common_cold', 'pneumonia','bacillary_dysentery', 'hand_foot_mouth', 'infectious_diarrhea','hemorrhagic_fever', 'lyme_disease', 'brucellosis']] combined_data.rename(columns={'date': 'timestamp'}, inplace=True)# 保存合并后的数据 combined_data.to_csv(os.path.join(results_path, 'combined_weather_disease_data.csv'), index=False)print(f"合并数据已保存到: {os.path.join(results_path, 'combined_weather_disease_data.csv')}")# 5. 生成数据摘要报告print("\n=== 数据生成完成 ===")print(f"数据时间范围: {combined_data['timestamp'].min()} 至 {combined_data['timestamp'].max()}")print(f"总记录数: {len(combined_data)} 天")print(f"气象变量数: 15 个")print(f"疾病变量数: 9 个")print(f"总变量数: {len(combined_data.columns)} 个\n")print("合并数据的前6行:")print(combined_data.head(6))print("\n各疾病年度平均报告数:") annual_summary = combined_data.groupby('year').agg({'influenza': 'mean','common_cold': 'mean','pneumonia': 'mean','bacillary_dysentery': 'mean','hand_foot_mouth': 'mean','hemorrhagic_fever': 'mean' }).round(2)print(annual_summary.tail(10))print("\n=== 数据质量检查 ===")print("缺失值检查:")print(combined_data.isnull().sum())print("\n气象变量范围:") temp_range = f"{combined_data['temp_min'].min():.1f} to {combined_data['temp_max'].max():.1f}°C" humidity_range = f"{combined_data['humidity_min'].min():.1f} to {combined_data['humidity_max'].max():.1f}%" max_precipitation = f"{combined_data['precipitation_total'].max():.1f}"print(f"温度范围: {temp_range}")print(f"湿度范围: {humidity_range}")print(f"最大降水量: {max_precipitation}")print("\n疾病报告范围:") influenza_range = f"{combined_data['influenza'].min()} to {combined_data['influenza'].max()}" dysentery_range = f"{combined_data['bacillary_dysentery'].min()} to {combined_data['bacillary_dysentery'].max()}" hemorrhagic_range = f"{combined_data['hemorrhagic_fever'].min()} to {combined_data['hemorrhagic_fever'].max()}"print(f"流感范围: {influenza_range}")print(f"痢疾范围: {dysentery_range}")print(f"出血热范围: {hemorrhagic_range}")print(f"\n所有数据文件已保存到桌面 {results_path} 文件夹中。")print("文件列表:")print("1. hourly_weather_data.csv - 逐小时气象数据")print("2. daily_weather_summary.csv - 逐日气象汇总数据")print("3. daily_disease_data.csv - 逐日疾病数据")print("4. combined_weather_disease_data.csv - 合并的最终数据库")return combined_datadef create_time_series_plots(combined_data):"""创建各种时间序列图"""# 创建输出目录 plot_dirs = ['Daily_Original', 'Weekly_Aggregated', 'Monthly_Aggregated','Quarterly_Aggregated', 'Yearly_Individual', 'Facet_Matrix','Overlay_Plots', 'Seasonal_Decomposition']for dir_name in plot_dirs: os.makedirs(os.path.join(results_path, 'TimeSeries_Plots', dir_name), exist_ok=True)# 定义变量分组 meteorological_vars = ['temp_max', 'temp_min', 'temp_mean', 'temp_median','humidity_max', 'humidity_min', 'humidity_mean', 'humidity_median','wind_max', 'wind_mean', 'pressure_max', 'pressure_min', 'pressure_mean','precipitation_total', 'sunshine_hours' ] disease_vars = ['influenza', 'common_cold', 'pneumonia','bacillary_dysentery', 'hand_foot_mouth', 'infectious_diarrhea','hemorrhagic_fever', 'lyme_disease', 'brucellosis' ] all_vars = meteorological_vars + disease_vars# 确保timestamp是datetime类型 combined_data['timestamp'] = pd.to_datetime(combined_data['timestamp'])# 添加时间聚合变量 combined_data['year_month'] = combined_data['timestamp'].dt.to_period('M').astype(str) combined_data['quarter'] = combined_data['timestamp'].dt.quarter combined_data['year_quarter'] = combined_data['year'].astype(str) + '-Q' + combined_data['quarter'].astype(str)# 1. 每日原始序列图print("生成每日原始序列图...")for var in all_vars: plt.figure(figsize=(12, 6)) plt.plot(combined_data['timestamp'], combined_data[var], linewidth=0.3, alpha=0.7, color='steelblue') plt.title(f'每日 {var} 时间序列\n1981-2025 原始日数据') plt.xlabel('日期') plt.ylabel(var) plt.xticks(rotation=45) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Daily_Original', f'Daily_{var}.png'), dpi=300, bbox_inches='tight') plt.close()
# 2. 每周统计图print("生成每周统计图...")# 按周聚合 weekly_data = combined_data.copy() weekly_data['year_week'] = weekly_data['timestamp'].dt.isocalendar().year.astype(str) + '-W' + \ weekly_data['timestamp'].dt.isocalendar().week.astype(str).str.zfill(2) weekly_agg = weekly_data.groupby('year_week').agg({var: 'mean'for var in all_vars}).reset_index() weekly_agg['date'] = pd.to_datetime(weekly_agg['year_week'] + '-1', format='%Y-W%W-%w')for var in all_vars: plt.figure(figsize=(12, 6)) plt.plot(weekly_agg['date'], weekly_agg[var], linewidth=0.5, color='darkorange', label='周均值')# 添加平滑曲线 from scipy.ndimage import gaussian_filter1d smoothed = gaussian_filter1d(weekly_agg[var].values, sigma=2) plt.plot(weekly_agg['date'], smoothed, linewidth=1, color='red', label='平滑曲线') plt.title(f'每周平均 {var} 时间序列\n按周聚合数据') plt.xlabel('日期') plt.ylabel(f'平均 {var}') plt.legend() plt.xticks(rotation=45) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Weekly_Aggregated', f'Weekly_Mean_{var}.png'), dpi=300, bbox_inches='tight') plt.close()# 3. 每月统计图print("生成每月统计图...") monthly_agg = combined_data.groupby('year_month').agg({var: 'mean'for var in all_vars}).reset_index() monthly_agg['date'] = pd.to_datetime(monthly_agg['year_month'] + '-01')for var in all_vars: plt.figure(figsize=(12, 6)) plt.plot(monthly_agg['date'], monthly_agg[var], linewidth=0.6, color='darkgreen', label='月均值')# 添加平滑曲线 smoothed = gaussian_filter1d(monthly_agg[var].values, sigma=1) plt.plot(monthly_agg['date'], smoothed, linewidth=1, color='blue', label='平滑曲线') plt.title(f'每月平均 {var} 时间序列\n按月聚合数据') plt.xlabel('日期') plt.ylabel(f'平均 {var}') plt.legend() plt.xticks(rotation=45) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Monthly_Aggregated', f'Monthly_Mean_{var}.png'), dpi=300, bbox_inches='tight') plt.close()# 4. 季度统计图print("生成季度统计图...") quarterly_agg = combined_data.groupby('year_quarter').agg({var: 'mean'for var in all_vars}).reset_index() quarterly_agg['date'] = pd.to_datetime(quarterly_agg['year_quarter'].str.replace('Q', ''), format='%Y-%m') + pd.offsets.MonthBegin(1)for var in all_vars: plt.figure(figsize=(12, 6)) plt.plot(quarterly_agg['date'], quarterly_agg[var], linewidth=0.8, color='purple', marker='o', markersize=3, label='季度均值')# 添加线性趋势线 x_numeric = np.arange(len(quarterly_agg)) z = np.polyfit(x_numeric, quarterly_agg[var], 1) p = np.poly1d(z) plt.plot(quarterly_agg['date'], p(x_numeric), linewidth=1, color='red', linestyle='--', alpha=0.7, label='线性趋势') plt.title(f'季度平均 {var} 时间序列\n按季度聚合数据，包含线性趋势') plt.xlabel('日期') plt.ylabel(f'平均 {var}') plt.legend() plt.xticks(rotation=45) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Quarterly_Aggregated', f'Quarterly_Mean_{var}.png'), dpi=300, bbox_inches='tight') plt.close()# 5. 按年份的单独时间序列图print("生成按年份的单独时间序列图...") recent_years = range(2016, 2026)for var in all_vars: yearly_data = combined_data[combined_data['year'].isin(recent_years)].copy() yearly_agg = yearly_data.groupby(['year', 'day_of_year'])[var].mean().reset_index() plt.figure(figsize=(12, 8))for year in recent_years: year_data = yearly_agg[yearly_agg['year'] == year] plt.plot(year_data['day_of_year'], year_data[var], linewidth=0.6, label=str(year)) plt.title(f'按年份的 {var} 时间序列\n年份: 2016-2025') plt.xlabel('一年中的第几天') plt.ylabel(var) plt.legend(bbox_to_anchor=(0.5, -0.2), loc='upper center', ncol=5) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Yearly_Individual', f'Yearly_Individual_{var}.png'), dpi=300, bbox_inches='tight') plt.close()# 6. 分面矩阵排列图print("生成分面矩阵排列图...")# 气象变量分面 meteo_vars_to_plot = meteorological_vars[:8] fig, axes = plt.subplots(len(meteo_vars_to_plot), 1, figsize=(14, 16))for i, var in enumerate(meteo_vars_to_plot): axes[i].plot(monthly_agg['date'], monthly_agg[var], linewidth=0.3, color='steelblue') axes[i].set_title(var, fontweight='bold') axes[i].set_ylabel(var)if i == len(meteo_vars_to_plot) - 1: axes[i].set_xlabel('日期') axes[i].tick_params(axis='x', rotation=45) plt.suptitle('气象变量分面时间序列图\n按月聚合数据', fontsize=16) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Facet_Matrix', 'Facet_Meteorological_Vars.png'), dpi=300, bbox_inches='tight') plt.close()# 疾病变量分面 fig, axes = plt.subplots(len(disease_vars), 1, figsize=(14, 16))for i, var in enumerate(disease_vars): axes[i].plot(monthly_agg['date'], monthly_agg[var], linewidth=0.3, color='firebrick') axes[i].set_title(var, fontweight='bold') axes[i].set_ylabel('病例数')if i == len(disease_vars) - 1: axes[i].set_xlabel('日期') axes[i].tick_params(axis='x', rotation=45) plt.suptitle('疾病变量分面时间序列图\n按月聚合数据', fontsize=16) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Facet_Matrix', 'Facet_Disease_Vars.png'), dpi=300, bbox_inches='tight') plt.close()# 7. 叠加时间序列图print("生成叠加时间序列图...")# 温度变量叠加 temp_vars = ['temp_max', 'temp_min', 'temp_mean'] temp_data = monthly_agg[['date'] + temp_vars].melt(id_vars=['date'], value_vars=temp_vars, var_name='temperature_type', value_name='value') plt.figure(figsize=(12, 6))for temp_type in temp_vars: data = temp_data[temp_data['temperature_type'] == temp_type] plt.plot(data['date'], data['value'], label=temp_type, linewidth=0.6) plt.title('温度变量叠加时间序列\n最大、最小、平均温度对比') plt.xlabel('日期') plt.ylabel('温度 (°C)') plt.legend(bbox_to_anchor=(0.5, -0.15), loc='upper center', ncol=3) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Overlay_Plots', 'Overlay_Temperature.png'), dpi=300, bbox_inches='tight') plt.close()# 湿度变量叠加 humidity_vars = ['humidity_max', 'humidity_min', 'humidity_mean'] humidity_data = monthly_agg[['date'] + humidity_vars].melt(id_vars=['date'], value_vars=humidity_vars, var_name='humidity_type', value_name='value') plt.figure(figsize=(12, 6))for humidity_type in humidity_vars: data = humidity_data[humidity_data['humidity_type'] == humidity_type] plt.plot(data['date'], data['value'], label=humidity_type, linewidth=0.6) plt.title('湿度变量叠加时间序列\n最大、最小、平均湿度对比') plt.xlabel('日期') plt.ylabel('相对湿度 (%)') plt.legend(bbox_to_anchor=(0.5, -0.15), loc='upper center', ncol=3) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Overlay_Plots', 'Overlay_Humidity.png'), dpi=300, bbox_inches='tight') plt.close()# 呼吸道疾病叠加 respiratory_vars = ['influenza', 'common_cold', 'pneumonia'] respiratory_data = monthly_agg[['date'] + respiratory_vars].melt(id_vars=['date'], value_vars=respiratory_vars, var_name='disease', value_name='cases') plt.figure(figsize=(12, 6))for disease in respiratory_vars: data = respiratory_data[respiratory_data['disease'] == disease] plt.plot(data['date'], data['cases'], label=disease, linewidth=0.6) plt.title('呼吸道疾病叠加时间序列\n流感、普通感冒、肺炎对比') plt.xlabel('日期') plt.ylabel('病例数') plt.legend(bbox_to_anchor=(0.5, -0.15), loc='upper center', ncol=3) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Overlay_Plots', 'Overlay_Respiratory_Diseases.png'), dpi=300, bbox_inches='tight') plt.close()# 肠道疾病叠加 intestinal_vars = ['bacillary_dysentery', 'hand_foot_mouth', 'infectious_diarrhea'] intestinal_data = monthly_agg[['date'] + intestinal_vars].melt(id_vars=['date'], value_vars=intestinal_vars, var_name='disease', value_name='cases') plt.figure(figsize=(12, 6))for disease in intestinal_vars: data = intestinal_data[intestinal_data['disease'] == disease] plt.plot(data['date'], data['cases'], label=disease, linewidth=0.6) plt.title('肠道传染病叠加时间序列\n细菌性痢疾、手足口病、感染性腹泻对比') plt.xlabel('日期') plt.ylabel('病例数') plt.legend(bbox_to_anchor=(0.5, -0.15), loc='upper center', ncol=3) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Overlay_Plots', 'Overlay_Intestinal_Diseases.png'), dpi=300, bbox_inches='tight') plt.close()
# 8. 季节性分解图print("生成季节性分解图...") try: from statsmodels.tsa.seasonal import seasonal_decompose key_vars = ['temp_mean', 'influenza', 'bacillary_dysentery', 'hemorrhagic_fever']for var in key_vars:# 创建月度时间序列 ts_data = monthly_agg.set_index('date')[var]# 季节性分解 decomposition = seasonal_decompose(ts_data, model='additive', period=12)# 绘制分解图 fig, axes = plt.subplots(4, 1, figsize=(12, 8)) decomposition.observed.plot(ax=axes[0], title='原始序列', color='blue') decomposition.trend.plot(ax=axes[1], title='趋势', color='red') decomposition.seasonal.plot(ax=axes[2], title='季节性', color='green') decomposition.resid.plot(ax=axes[3], title='残差', color='black') plt.suptitle(f'{var} 季节性分解', fontsize=16) plt.tight_layout() plt.savefig(os.path.join(results_path, 'TimeSeries_Plots', 'Seasonal_Decomposition', f'Seasonal_Decomposition_{var}.png'), dpi=300, bbox_inches='tight') plt.close() except ImportError:print("警告: 未安装statsmodels库，跳过季节性分解图")# 9. 生成汇总报告print("生成绘图汇总报告...") plot_summary = pd.DataFrame({'图表类型': ['每日原始序列', '每周统计', '每月统计', '季度统计','按年份单独', '分面矩阵', '叠加序列', '季节性分解'],'数量': [len(all_vars), len(all_vars), len(all_vars), len(all_vars), len(all_vars), 4, 5, len(key_vars)],'文件位置': ['Daily_Original', 'Weekly_Aggregated', 'Monthly_Aggregated', 'Quarterly_Aggregated','Yearly_Individual', 'Facet_Matrix', 'Overlay_Plots', 'Seasonal_Decomposition'],'描述': ['原始日数据序列', '按周聚合的均值', '按月聚合的均值', '按季度聚合的均值','按年份分别展示', '变量分面排列', '相关变量叠加', '季节性模式分解'] }) plot_summary.to_csv(os.path.join(results_path, 'TimeSeries_Plots', 'plot_summary.csv'), index=False)# 生成变量统计摘要 variable_summary = pd.DataFrame({'变量名称': all_vars,'变量类型': ['气象'if var in meteorological_vars else'疾病'for var in all_vars],'数据范围': [f"{combined_data[var].min():.2f} - {combined_data[var].max():.2f}"for var in all_vars],'均值': [f"{combined_data[var].mean():.2f}"for var in all_vars] }) variable_summary.to_csv(os.path.join(results_path, 'TimeSeries_Plots', 'variable_summary.csv'), index=False)print("\n=== 时间序列图生成完成 ===")print("总生成图表数量:")print(f"- 每日原始序列图: {len(all_vars)} 个")print(f"- 每周统计图: {len(all_vars)} 个")print(f"- 每月统计图: {len(all_vars)} 个")print(f"- 季度统计图: {len(all_vars)} 个")print(f"- 按年份单独图: {len(all_vars)} 个")print(f"- 分面矩阵图: 4 个")print(f"- 叠加序列图: 5 个")print(f"- 季节性分解图: {len(key_vars)} 个")print(f"总计: {len(all_vars) * 5 + 9} 个图表文件\n")print(f"文件保存位置: {os.path.join(results_path, 'TimeSeries_Plots')}")print("子文件夹:")print("1. Daily_Original - 每日原始序列")print("2. Weekly_Aggregated - 每周聚合序列")print("3. Monthly_Aggregated - 每月聚合序列")print("4. Quarterly_Aggregated - 每季度聚合序列")print("5. Yearly_Individual - 按年份单独序列")print("6. Facet_Matrix - 分面矩阵排列")print("7. Overlay_Plots - 叠加时间序列")print("8. Seasonal_Decomposition - 季节性分解\n")print("汇总文件:")print("- plot_summary.csv - 绘图汇总信息")print("- variable_summary.csv - 变量统计摘要")print("\n示例图表预览:")print( f"气象变量分面图: {os.path.join(results_path, 'TimeSeries_Plots', 'Facet_Matrix', 'Facet_Meteorological_Vars.png')}")print(f"疾病变量分面图: {os.path.join(results_path, 'TimeSeries_Plots', 'Facet_Matrix', 'Facet_Disease_Vars.png')}")print( f"呼吸道疾病叠加图: {os.path.join(results_path, 'TimeSeries_Plots', 'Overlay_Plots', 'Overlay_Respiratory_Diseases.png')}")def main():"""主函数"""print(f"结果将保存到: {results_path}")# 生成数据 combined_data = generate_weather_disease_data()# 创建时间序列图 create_time_series_plots(combined_data)if __name__ == "__main__": main()
