连接pandas中连续行的简明方法

import pandas as pd

# 创建示例数据
data = {
    'col1': [1, 2, 3, 4, 6, 7, 9, 10, 12],
    'col2': ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I']
}
df = pd.DataFrame(data)

# 计算每行与上一行的差异
df['diff'] = df['col1'].diff()

# 筛选出差异为1的行
continuous_rows = df[df['diff'] == 1]

# 计算连续行的分组编号
continuous_rows['group'] = continuous_rows['diff'].cumsum()

# 分组连接连续行
result = continuous_rows.groupby('group').agg({'col1': lambda x: '-'.join(map(str, x)), 'col2': 'first'})

# 重置索引
result.reset_index(drop=True, inplace=True)

print(result)

   col1 col2
0   1-2    A
1   3-4    C
2   6-7    E
3  9-10    G