df中所有列的值计数？

Question

I have a df such as:

Monday | Tuesday | Friday | January  | Weekday 
True      False     True     False      True
False     False     False    False      False
True      True      False    False      True
False     False     False    False      False

I want to calculate the % that each column has True and False, typically I would use a value counts with a normalize parameter below as:

df['Monday'].value_counts(normalize = True) 

Output:

Monday 
True   .5
False  .5

But how can I run the same code where it would give me that % for all columns in the df, instead of doing it manually per column?

Thanks!

Answer 1

apply value value_counts to the dataframe. Groupby the resulting dataframe's index and sum

g=df.apply(lambda x: x.astype(str).value_counts(normalize=True))
g.groupby(g.index).sum()



     Monday  Tuesday  Friday  January  Weekday
False     0.5     0.75    0.75      1.0      0.5
True      0.5     0.25    0.25      0.0      0.5

Answer 2

Try this, use pd.dataframe.apply to apply the pd.Series.value_counts function passing normalize=True as a parameter to the apply function:

df.apply(pd.Series.value_counts, normalize=True).fillna(0)

Output:

       Monday  Tuesday  Friday  January  Weekday
False     0.5     0.75    0.75      1.0      0.5
True      0.5     0.25    0.25      0.0      0.5

Answer 3

You could apply it to each column and then concat (with pd.concat ) the results:

res = pd.concat((df[col].value_counts(normalize=True) for col in df.columns), axis=1).fillna(0)
print(res)

Output

       Monday  Tuesday  Friday  January  Weekday
True      0.5     0.25    0.25      0.0      0.5
False     0.5     0.75    0.75      1.0      0.5

Use fillna , to fill the cells of columns where some values are missing, for example, the True value is missing from the January column.

Answer 4

df.apply(pd.Series.value_counts, normalize=True).fillna(0)

This should work.