Say I have two pandas dataframe like
data1 = [[s1, 0], [s2, 0], [s3, 1], [s4, 1], [s5, 0], [s6, 1], [s7, 0], [s8, 1]]
df1 = pd.DataFrame(data1, columns = ['s_no', 'values'])
data2 = [[s1, 0], [s2, 1], [s3, 1], [s4, 0], [s5, 0], [s6, 1], [s7, 1], [s8, 0]]
df2 = pd.DataFrame(data2, columns = ['s_no2', 'values2'])
How to find the value_counts of
df2['values2'].value_counts() when df1['values'] == 1 and
df2['values2'].value_counts() when df1['values'] == 0
First is necessary same index values and size in both DataFrames, so possible join togehter, and then is possible use boolean indexing.
Advatage is possible check if correct aligned, if data for next processing are like you need.
df = pd.concat([df1, df2], axis=1)
print (df)
s_no values s_no2 values2
0 s1 0 s1 0
1 s2 0 s2 0
2 s3 1 s3 1
3 s4 1 s4 1
4 s5 0 s5 0
5 s6 1 s6 1
6 s7 0 s7 0
7 s8 1 s8 1
print (df.loc[df['values'] == 1, 'values2'].value_counts())
1 4
Name: values2, dtype: int64
print (df.loc[df['values'] == 0, 'values2'].value_counts())
0 4
Name: values2, dtype: int64
print (df.groupby('values')['values2'].value_counts())
values values2
0 0 4
1 1 4
Name: values2, dtype: int64
Or is possible filter with df1, df2 - but here is not so easy checking, if something wrong with data:
print (df2.loc[df1['values'] == 1, 'values2'].value_counts())
1 4
Name: values2, dtype: int64
print (df2.loc[df1['values'] == 0, 'values2'].value_counts())
0 4
Name: values2, dtype: int64
print (df2['values2'].groupby(df1['values']).value_counts())
values values2
0 0 4
1 1 4
Name: values2, dtype: int64
