pandas DataFrame如何补全循环规律缺失行并将对应值填充为0

实现思路 #

核心逻辑是先给原数据的循环序列打分组标签，每遇到一次循环起始值0就归为新的一组，再给每组补全0、1、2、3的完整序列，匹配原数据的值，缺失位置填充0即可，该方法同时适配你提到的两种场景。

场景1：列名为df、df1的情况 #

import pandas as pd

# df_raw为你读入的原始DataFrame
# 新增分组列：每遇到df值为0，分组号+1
df_raw['group_id'] = (df_raw['df'] == 0).cumsum()

# 生成单组的完整循环序列
full_seq = pd.DataFrame({'df': [0, 1, 2, 3]})
# 生成所有分组的完整循环序列
full_df = df_raw[['group_id']].drop_duplicates().merge(full_seq, how='cross')

# 合并原数据，缺失的df1填充为0
result = full_df.merge(df_raw, on=['group_id', 'df'], how='left').fillna({'df1': 0})

# 调整列顺序，重置索引得到最终结果
result = result[['df', 'df1']].reset_index(names='index')

场景2：列名为df1、df2的补充场景 #

仅需要替换对应列名即可，逻辑完全一致：

import pandas as pd

# df_raw为你读入的原始DataFrame
df_raw['group_id'] = (df_raw['df1'] == 0).cumsum()

full_seq = pd.DataFrame({'df1': [0, 1, 2, 3]})
full_df = df_raw[['group_id']].drop_duplicates().merge(full_seq, how='cross')

result = full_df.merge(df_raw, on=['group_id', 'df1'], how='left').fillna({'df2': 0})

result = result[['df1', 'df2']].reset_index(names='index')

内容的提问来源于stack exchange，提问作者김수환