使用正则表达式过滤 DataFrame 列

Question

I count statistics for the dataset, and I want to filter columns that contain specific strings. How I could do it with regex?

Here in volumes_c I filtered some structures, that have Volume in there names

Select_list = ["Amygdala", "Hippocampus", "Lateral-Ventricle", "Pallidum", "Putamen", "Thalamus", "Caudate"]
Side = ["Left", "Right"]
#Selected columns
if(Select_list):
    for s in Side:
        for struct in Select_list:
            volumes_c = group_c.filter(regex="^(?=.*"+s+")(?=.*"+struct+")(?=.*Volume)")

Now i want to filter columns that contain SurfArea in:

Answer 1

Suppose DataFrame data is in the variable df , so the filter will be:

 df.filter(like="SurfArea", axis=1)

Actually, 'axis' arg has a default value 1 and you can omit it, but if you want to filter by rows set it up as 0.