Python pandas，如何在列中找到最大值后的最小值

Question

I'd like to find the min value/index in the rows that are after the max value location. For example:

df = pd.DataFrame({"c1": [22, 55, 48, 30, 35], "c2": [6, 2, 33, 22, 17]})
df

    c1  c2
0   22  6
1   55  2
2   48  33
3   30  22
4   35  17

For c1, I need the min after max value 55, for c2, I need the min after max value 33, so the desired result should be

c1 30
c2 17

Obviously I could get idxmax first and pass to an apply function. Is there a simpler/efficient way and avoiding the apply function? I may have thousands of columns. Appreciate the help....

Answer 1

We can use max then do the cummax

s=df.where(df.eq(df.max()).cummax()).min()
Out[243]: 
c1    30.0
c2    17.0
dtype: float64

Answer 2

Does this work?

import pandas as pd

df = pd.DataFrame({"c1": [22, 55, 48, 30, 35], "c2": [6, 2, 33, 22, 17]})

col_name = 'c1'
max_index = df[col_name].argmax()
min_index = max_index + 1 + df[col_name].iloc[max_index + 1:].argmin()
df[col_name].iloc[min_index]  # 30

col_name = 'c2'
max_index = df[col_name].argmax()
min_index = max_index + 1 + df[col_name].iloc[max_index + 1:].argmin()
df[col_name].iloc[min_index]  # 17

Note that pd.Series.argmax and pd.Series.argmin return the index of the first max (or min) only. If there are multiple max or min values, this may need to be adjusted.

Also, this won't work if max is in the last row but we can modify it to return None or let it throw an error if that's expected.