Let’s say we have the following example pandas dataframe:

import pandas as pd

data = {'A': [1, 2, 3, 4], 'B': [None, 'x', 'y', None]}
df = pd.DataFrame(data)
print(df)

# we get the following
#    A     B
# 0  1  None
# 1  2     x
# 2  3     y
# 3  4  None

However, problems occur when attempting to filter using df['column'] != None

print(df[df['B'] != None])

# we get the following
#    A     B
# 0  1  None
# 1  2     x
# 2  3     y
# 3  4  None

It only functions correctly when we use the specific method df['column'].notna()

print(df[df['B'].notna()])

# we get the following
#    A  B
# 1  2  x
# 2  3  y

Why? Can someone explain this behavior?