问题
My dataframe looks like this:
datetime1 datetime2 datetime3 datetime4
id
1 5 6 5 5
2 7 2 3 5
3 4 2 3 2
4 6 4 4 7
5 7 3 8 9
and I have a numpy array like this:
index_arr = [3, 2, 0, 1, 2]
This numpy array refers to the index in each row, respectively, that I want to replace. The values I want to use in the replacement are in another numpy array:
replace_arr = [14, 12, 23, 17, 15]
so that the updated dataframe looks like this:
datetime1 datetime2 datetime3 datetime4
id
1 5 6 5 14
2 7 2 12 5
3 23 2 3 2
4 6 17 4 7
5 7 3 15 9
What is the best way to go about doing this replacement quickly? I've tried using enumerate and iterrows but couldn't get the syntax to work. Would appreciate any help - thank you
回答1:
Here's one way with np.put_along_axis -
In [50]: df
Out[50]:
datetime1 datetime2 datetime3 datetime4
1 5 6 5 5
2 7 2 3 5
3 4 2 3 2
4 6 4 4 7
5 7 3 8 9
In [51]: index_arr = np.array([3, 2, 0 ,1 ,2])
In [52]: replace_arr = np.array([14, 12, 23, 17 ,15])
In [53]: np.put_along_axis(df.to_numpy(),index_arr[:,None],replace_arr[:,None],axis=1)
In [54]: df
Out[54]:
datetime1 datetime2 datetime3 datetime4
1 5 6 5 14
2 7 2 12 5
3 23 2 3 2
4 6 17 4 7
5 7 3 15 9
回答2:
IIUC, you can just assign to .values (or .to_numpy(copy=False)):
# <= 0.23
df.values[np.arange(len(df)), index_arr] = replace_arr
# 0.24+
df.to_numpy(copy=False)[np.arange(len(df)), index_arr] = replace_arr
df
datetime1 datetime2 datetime3 datetime4
id
1 5 6 5 14
2 7 2 12 5
3 23 2 3 2
4 6 17 4 7
5 7 3 15 9
回答3:
End up using .iat
for x, y ,z in zip(np.arange(len(df)),index_arr ,replace_arr ):
df.iat[x,y]=z
df
Out[657]:
datetime1 datetime2 datetime3 datetime4
id
1 5 6 5 14
2 7 2 12 5
3 23 2 3 2
4 6 17 4 7
5 7 3 15 9
来源：https://stackoverflow.com/questions/56710785/how-to-replace-value-in-specific-index-in-each-row-with-corresponding-value-in-n