How to select the row having any two columns not null-CodePudding

I have a dataframe like the one shown in the image.

enter image description here

I need to select the rows which have more than 2 columns having non zero values. So the highlighted yellow ones only need to be selected. I tried

df[(df.iloc[:, 1:13] != 0.0).any(1)]

but it's for any 1 column how to do it for more than 1 column.

CodePudding user response：

My suggestion would be to generate a True/False filter as:

index_filter = df.apply(lambda x: True if sum(x != 0) > 1 else False, axis=1)

index_filter will be True at the indices where at least two of the dataframe columns are not 0.

Now you can filter the dataframe as df[index_filter].

Another alternative would be:

num_cols_not_zero = df.ne(0).sum(axis=1)
df[num_cols_not_zero.gt(1)]

CodePudding user response：

Try this: main_df.loc[:,main_df.notnull().sum() > 1]