如何使用python删除-9999作为列值的数据帧的行？

Question

for column in df:
print(column)
print((df[column] == -9999).value_counts()) 
print("")

modelMag_u
False    49977
True        23
Name: modelMag_u, dtype: int64

modelMag_i
False    49985
True        15
Name: modelMag_i, dtype: int64

Can anyone please help me how to remove the rows with -9999 as column value?

Answer 1

Assuming the following: You want to delete a row if there is any column in that row that has value -9999

df[~(df == -9999).any(axis='columns')]

How does this work?

df == -9999 is run first. This compares every cell to -9999, creating a dataframe of True and False values depending on content of each cell. We then use .any(axis='columns') on this dataframe. any (with the axis=columns parameter) returns True for each row that has any true value. This is exactly those rows you want to remove. The df[~ ... ] construct keeps any rows that is not True in the ... . What remains is any row that has no cell with value -9999.