CODEWITHSUNDEEP
pythonpandas
Daniel
Daniel

Reputation: 5391

Get index and columns based on conditional

Having a DF as the example below, how can i get list with the index and column name of the values that arent nan?

                  es  ms        hs
subdist_id                        
1                NaN NaN       NaN
2                NaN NaN       NaN
3          -0.218066 NaN -0.309002
4                NaN NaN       NaN
5                NaN NaN       NaN
6                NaN NaN       NaN
7                NaN NaN       NaN
9                NaN NaN       NaN
10               NaN NaN       NaN
11         -0.385217 NaN       NaN

so i can get a list [[3,'es]', [3,'hs'], [11,'es']]

Upvotes: 1

Views: 207

Answers (1)

jezrael
jezrael

Reputation: 863731

Use DataFrame.stack for remove NaNs with convert MultiIndex to list of tuples and then convert it to list of lists:

L = list(map(list, df.stack().index.tolist()))
print (L)
[[3, 'es'], [3, 'hs'], [11, 'es']]

Alternative:

L = [list(x) for x in df.stack().index.tolist()]
print (L)
[[3, 'es'], [3, 'hs'], [11, 'es']]

If need performance,get indices by condition by DataFrame.isna and numpy.where and get values by zip with indexing for values of columns and index:

i,c = np.where(df.notna())
L = list(map(list, zip(df.index[i],df.columns[c])))
print (L)
[[3, 'es'], [3, 'hs'], [11, 'es']]

i,c = np.where(df.notna())
L = [list(x) for x in zip(df.index[i],df.columns[c])]

Upvotes: 3

Related Questions