Find Values In Indexed Columns Of Pandas Which Is A List

Let's say I have a dataframe like following d = {'col1': [1, 2,4,5,6], 'col2': [3, 4,5,6,7], 'col3': ['foo', 'bar', 'baz','biz', 'boo'], 'col4': [['foo', 'bar'], ['baz'], ['foo

Solution 1:

Use Index.map with in statement:

df1 = indexed_df[indexed_df.index.map(lambda x: 'baz'in x)]
print (df1)
            col1  col2 col3
col4                       
[baz]          24  bar
[baz, foo]     67  boo

Alternative with list comprehension:

df1 = indexed_df[['baz' in x for x in indexed_df.index]]

Solution 2:

Let us just try explode

out= indexed_df[pd.Series(indexed_df.index).explode().isin(['baz']).any(level=0).values]
Out[241]: 
            col1  col2 col3
col4                       
[baz]          24  bar
[baz, foo]     67  boo

Solution 3:

You can take advantage of using MultiIndex rather than a list as index:

>>> df.set_index(pd.MultiIndex.from_tuples(d["col4"]))
         col1  col2 col3        col4
foo bar     13  foo  [foo, bar]
baz NaN     24  bar       [baz]
foo NaN     45  baz       [foo]
biz NaN     56  biz       [biz]
baz foo     67  boo  [baz, foo]