pandas fill a dataframe according to column and row value operations

Question

Let's say that I have this dataframe:

,,,,,,
,,2.0,,,,
,2.0,,2.23606797749979,,,
,,2.23606797749979,,,2.0,
,,,,,2.23606797749979,
,,,2.0,2.23606797749979,,
,,,,,,

I would like to get a two dimensional vector with values of the indexes and the columns of each element which is not nan.

For example, in this case, I am expecting:

[[2,1],[1,2],[3,2],[2,3],[5,3],[3,5],[4,5],[5,4]].

I am thinking about using iloc and the np.where functions but I am not able to merge the two concepts.

jezrael · Accepted Answer · 2023-02-10 14:16:04Z

2

Use DataFrame.stack for remove missing values, if necessary add Series.swaplevel and in list comprehension convert nested tuples to lists:

L = [list(y) for y in df.stack().swaplevel().index]
print (L)
[[2, 1], [1, 2], [3, 2], [2, 3], [5, 3], [5, 4], [3, 5], [4, 5]]

Or if use indices after np.where solution is similar:

r, c = np.where(df.notna())
L = [list(x) for x in zip(c, r)]
print (L)
[[2, 1], [1, 2], [3, 2], [2, 3], [5, 3], [5, 4], [3, 5], [4, 5]]

edited Feb 10, 2023 at 14:16

answered Feb 10, 2023 at 13:33

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

diedro Over a year ago

Is it possible to eliminate the pairs made by the same couple. I mean, instead of having [2,1] and [1,2] just [1,2]. Maybe another post is more appropriate. What do you think?

jezrael Over a year ago

@diedro sorry, I am on phone only, possible solution is https://stackoverflow.com/a/15037238

Collectives™ on Stack Overflow

pandas fill a dataframe according to column and row value operations

1 Answer 1

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related