Filtering pandas dataframe column names that are not in a list

Question

I have the following dataframe:

df = pd.DataFrame(
{'a': [-0.558, -0.5584, -0.5583, -0.5583, -0.5581], 
 'b': [0.5324, 0.5324, 0.5324, 0.5324, 0.5323], 
 'c': [-0.2803, -0.2803, 0.2803, -0.2803, -0.2851],
 'd': [0.2359, 0.6532, 0.1254, 0.1231, 0.1785]})

    a        b       c      d
0   -0.5580 0.5324  -0.2803 0.2359
1   -0.5584 0.5324  -0.2803 0.6532
2   -0.5583 0.5324  -0.2803 0.1254
3   -0.5583 0.5324  -0.2803 0.1231
4   -0.5581 0.5323  -0.2851 0.1785

I want to create two lists with column names where one list contains column names with negative values and another one with column names with positive values.

negative = df.columns[(df < 0).any()]

I'm struggling to get something like:

positive = df.columns NOT IN NEGATIVE

This code return error:

list(df[[column for column in df.columns if column != negative]])

>>> ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

Any suggestions are much appreciated! Thank you!

Answer 1

You can use boolean indexing:

mask = (df < 0).all()

negative = df.columns[mask].tolist()
not_in_negative = df.columns[~mask].tolist()

print(negative)
print(not_in_negative)

Prints:

['a', 'c']
['b', 'd']