如何基于特定值标记DataFrame列中的降雨事件

现有一个包含多列的DataFrame，重点处理Rain_Stream['rain_event']列，原始数据示例如下（实际数据更长）：

Rain_Stream['rain_event'] = [nan, nan, nan, nan, nan, nan, nan, 1, nan, nan, nan, nan, nan, nan, nan, nan, 2, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, 1, nan, nan, nan, nan, nan, nan, 2, nan, nan, nan, nan, nan, nan, nan, 1, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, 2, nan, nan, nan, nan, nan, 1, nan, nan, nan, nan, nan, nan, 2, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan]

其中1标记降雨事件的开始，2标记降雨事件的结束。期望处理后，该列变为：

Rain_Stream['rain_event'] = [nan, nan, nan, nan, nan, nan, nan, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, 2, 2, 2, 2, 2, 2, 2, 2, nan, nan, nan, nan, nan, nan, nan, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, nan, nan, nan, nan, nan, 4, 4, 4, 4, 4, 4, 4, 4, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan]

要求：每个降雨事件按出现顺序用对应序号标记，从开始的1到结束的2之间的所有值都替换为该事件的序号（原1和2也需替换），事件之间保留nan或0均可。

已尝试掩码、向前填充（ffill）等方法但未成功，尝试的代码如下：

# 尝试1
mask = Rain_Stream['begin_rain_checked'] == 1
Rain_Stream['rain_event'] = mask.cumsum()
mask = Rain_Stream['end_rain_checked'] == 2
Rain_Stream['rain_event'] = mask.nan

# 尝试2
Rain_Stream['rain_event'] = Rain_Stream['rain_event'].ffill()
Rain_Stream['rain_event'] = Rain_Stream['rain_event'].mask(Rain_Stream['rain_event'].duplicated())
Rain_Stream['rain_event'] = Rain_Stream['rain_event'].ffill()

可以通过以下步骤实现需求：

• 标记降雨事件的起始点：统计所有1出现的次数，生成事件序号的基础
• 识别事件区间：判断每行是否处于某个降雨事件的区间内（起始后、结束前）
• 填充事件序号：将事件区间内的行统一赋值为对应事件的序号
• 处理结束点：确保结束点2也被替换为对应事件的序号

具体代码实现：

import pandas as pd
import numpy as np

# 复制原列避免修改原始数据，若需直接覆盖原列，替换为'rain_event'即可
Rain_Stream['rain_event_processed'] = Rain_Stream['rain_event'].copy()

# 生成每个事件的序号：每遇到1就递增1
event_num = Rain_Stream['rain_event'].eq(1).cumsum()

# 标记哪些行处于事件区间内（起始累计数≠结束累计数）
in_event = Rain_Stream['rain_event'].eq(1).cumsum() != Rain_Stream['rain_event'].eq(2).cumsum()

# 给事件区间内的行填充对应序号，非区间行保留nan
Rain_Stream['rain_event_processed'] = np.where(in_event, event_num, np.nan)

# 处理结束点2，替换为对应事件序号
end_positions = Rain_Stream['rain_event'] == 2
Rain_Stream.loc[end_positions, 'rain_event_processed'] = event_num[end_positions]

代码说明 #

• event_num：通过eq(1).cumsum()统计起始点1的出现次数，得到每个事件的唯一序号
• in_event：对比起始点和结束点的累计次数，精准锁定每个降雨事件的区间范围
• np.where：快速实现区间内赋值、非区间保留nan的逻辑
• 单独处理结束点，确保2被正确替换为对应事件序号，避免遗漏

内容的提问来源于stack exchange，提问作者Camilla T