Is there a way of extracting indices from a pandas DataFrame based on value

user_7337491 0
Viewed 61

I have the following DataFrame:

index col0 col1 col2
0     0    1    0
1     1    0    1
2     0    1    1

I would like to extract the following indices(those that contain ones(or any value)): 

[(0, 1), (1, 0), (1, 2), (2, 1), (2,2))]

Is there a method in pandas that can do this?

pandas python
3 Answers

Use np.where + zip

[*zip(*np.where(df))]


[(0, 1), (1, 0), (1, 2), (2, 1), (2, 2)]
user_34832030

Here is one way 

df.columns=np.arange(df.shape[1])
df.stack().loc[lambda x : x==1].index.tolist()
[(0, 1), (1, 0), (1, 2), (2, 1), (2, 2)]
user_79645270

You could use numpy.nonzero:

result = list(zip(*np.nonzero(df.values)))
print(result)

Output

[(0, 1), (1, 0), (1, 2), (2, 1), (2, 2)]
user_40015920
Related