import pandas as pd

## the time column doesn't matter in your problem
df = pd.DataFrame({
    'time':['2021-3-19','2021-3-20','2021-3-21','2021-3-22',
    '2021-3-23','2021-3-24','2021-3-25','2021-3-26','2021-3-27'],
    'value':[10,11,9,5,4,2,4,9,5],
    'status':['X']*3+['Y']+['X']+['Y']*2+['X']*2
})

df_new = pd.DataFrame(columns=df.columns)

## perform a groupby on consecutive values
for _, g in df.groupby([(df.status != df.status.shift()).cumsum()]):
    g = g.sort_values(by='value')
    ## keep the highest value for X
    if g.status.values[0] == 'X':
        g = g.drop_duplicates(subset=['status'], keep='last')

    ## keep the lowest value for Y
    elif g.status.values[0] == 'Y':
        g = g.drop_duplicates(subset=['status'], keep='first')

    else:
        pass
    df_new = pd.concat([df_new, g])
df_new = df_new.reset_index(drop=True)
print(df_new)
————————
程序输出
        time value status
0  2021-3-20    11      X
1  2021-3-22     5      Y
2  2021-3-23     4      X
3  2021-3-24     2      Y
4  2021-3-26     9      X
怎么能在原表中实现如下效果？==x 取较大值  ==y 取较小值
    time    value    status  value_1
0    2021-3-19    10    X        10
1    2021-3-20    11    X        11
2    2021-3-21    9    X        11
3    2021-3-22    5    Y         5
4    2021-3-23    4    X         4
5    2021-3-24    2    Y         2
6    2021-3-25    4    Y         2
7    2021-3-26    9    X         9
8    2021-3-27    5    X         9

谢谢