CODEWITHSUNDEEP
pythonpandasdataframe
stomar02
stomar02

Reputation: 97

pandas dataframe: identifiy NaN and zero values in one statement

Is there any way to combine the two statements df.isnull().sum() and (df == 0).sum() to get the following overview?

Demo:

df = pd.DataFrame({'a':[1,0,0,1,3], 'b':[0,NaN,1,NaN,1], 'c':[0,0,0,0,NaN]})

df

    a   b       c
0   1   0.0     0.0
1   0   NaN     0.0
2   0   1.0     0.0
3   1   NaN     0.0
4   3   1.0     NaN

Expected result:

a    2
b    3
c    5

Probably very simple, but I can't find the solution... Thank's for your help

Upvotes: 2

Views: 45

Answers (3)

EdChum
EdChum

Reputation: 394279

You mean just this:

In[27]:

(df==0).sum() + df.isnull().sum()
Out[27]: 
a    2
b    3
c    5
dtype: int64

EDIT

Thanks to @coldpseed for the suggestion, you can also do the following:

In[28]:
df[df!=0].isnull().sum()

Out[28]: 
a    2
b    3
c    5
dtype: int64

which is more succinct, I've always been more in favour of clarity but shorter code sometimes wins.

Upvotes: 5

BENY
BENY

Reputation: 323366

With fillna

df.fillna(0).eq(0).sum()
Out[8]: 
a    2
b    3
c    5
dtype: int64

Upvotes: 5

pault
pault

Reputation: 43534

Another option:

>>> ((df == 0) | df.isnull()).sum()
a    2
b    3
c    5
dtype: int64
>>>

Also:

>>> (df.eq(0) | df.isnull()).sum()
a    2
b    3
c    5
dtype: int64
>>>

Upvotes: 2

Related Questions