How do I get a summary count of missing/NaN data by column in 'pandas'?

问题

In R I can quickly see a count of missing data using the summary command, but the equivalent pandas DataFrame method, describe does not report these values.

I gather I can do something like

len(mydata.index) - mydata.count()

to compute the number of missing values for each column, but I wonder if there's a better idiom (or if my approach is even right).

回答1:

Both describe and info report the count of non-missing values.

In [1]: df = DataFrame(np.random.randn(10,2))

In [2]: df.iloc[3:6,0] = np.nan

In [3]: df
Out[3]: 
          0         1
0 -0.560342  1.862640
1 -1.237742  0.596384
2  0.603539 -1.561594
3       NaN  3.018954
4       NaN -0.046759
5       NaN  0.480158
6  0.113200 -0.911159
7  0.990895  0.612990
8  0.668534 -0.701769
9 -0.607247 -0.489427

[10 rows x 2 columns]

In [4]: df.describe()
Out[4]: 
              0          1
count  7.000000  10.000000
mean  -0.004166   0.286042
std    0.818586   1.363422
min   -1.237742  -1.561594
25%   -0.583795  -0.648684
50%    0.113200   0.216699
75%    0.636036   0.608839
max    0.990895   3.018954

[8 rows x 2 columns]


In [5]: df.info()
<class 'pandas.core.frame.DataFrame'>
Int64Index: 10 entries, 0 to 9
Data columns (total 2 columns):
0    7 non-null float64
1    10 non-null float64
dtypes: float64(2)

To get a count of missing, your soln is correct

In [20]: len(df.index)-df.count()
Out[20]: 
0    3
1    0
dtype: int64

You could do this too

In [23]: df.isnull().sum()
Out[23]: 
0    3
1    0
dtype: int64

回答2:

As a tiny addition, to get percentage missing by DataFrame column, combining @Jeff and @userS's answers above gets you:

df.isnull().sum()/len(df)*100

回答3:

This isnt quite a full summary, but it will give you a quick sense of your column level data

def getPctMissing(series):
    num = series.isnull().sum()
    den = series.count()
    return 100*(num/den)

回答4:

I can't make comments yet but to add on to Jeff's answer but if you didn't care which columns had Nan's and you just wanted to check overall just add a second .sum() to get a single value.

result = df.isnull().sum().sum()
result > 0

a Series would only need one .sum() and a Panel() would need three

回答5:

Following one will do the trick and will return counts of nulls for every column:

df.isnull().sum(axis=0)

df.isnull() returns a dataframe with True / False values
sum(axis=0) sums the values across all rows for a column

来源：https://stackoverflow.com/questions/22257527/how-do-i-get-a-summary-count-of-missing-nan-data-by-column-in-pandas

标签

pandas

reporting

nan

missing-data