缺失率

DataMiss/missingRate.py

@@ -0,0 +1,52 @@
+import pandas as pd
+import numpy as np
+
+def calculate_missing_rate(file_path):
+    # 读取CSV文件
+    df = pd.read_csv(file_path)
+
+    # 确保C_TIME列的格式是字符串并转换为datetime对象
+    df['C_TIME'] = pd.to_datetime(df['C_TIME'], format='%Y-%m-%d %H:%M:%S')
+
+    # 提取年月信息作为分组依据
+    df['YearMonth'] = df['C_TIME'].dt.to_period('M')
+
+    # 按月份分组处理数据
+    grouped = df.groupby('YearMonth')
+
+    # 结果字典保存每月缺失率和缺失的时间点
+    missing_rates = {}
+
+    for period, group in grouped:
+        # 获取该月的最小和最大时间
+        start_time = group['C_TIME'].min()
+        end_time = group['C_TIME'].max()
+
+        # 生成该月所有的15分钟时间点
+        full_time_range = pd.date_range(start=start_time, end=end_time, freq='15T')
+
+        # 找到该月的时间点缺失部分
+        missing_times = full_time_range.difference(group['C_TIME'])
+
+        # 计算缺失率：缺失时间点的数量 / 全部应该有的时间点数量
+        missing_rate = len(missing_times) / len(full_time_range)
+
+        # 保存结果
+        missing_rates[period] = {
+            'missing_rate': missing_rate,
+            'missing_times': missing_times
+        }
+
+    # 输出结果
+    for period, data in missing_rates.items():
+        print(f"Month: {period}, Missing Rate: {data['missing_rate']:.2%}")
+        if len(data['missing_times']) > 0:
+            print(f"  Missing Time Points:")
+            for missing_time in data['missing_times']:
+                print(f"    {missing_time}")
+        else:
+            print("  No missing time points")
+
+# 调用函数，假设文件路径是'tower-1-process.csv'
+file_path = 'weather-1-process.csv'
+calculate_missing_rate(file_path)