Counting the indices of Pandas DataFrame based on label location

Question

I am trying to count the indices of labels in Pandas DataFrame in each column. Basically I have the following DataFrame:

d = {'col1': ['label1', 'label2', 'label3'], 'col2': ['label2', 'label3', 'label1'], 'col3': ['label2', 'label1', 'label3'], 'col4': ['label3', 'label1', 'label2']} df = pd.DataFrame(data = d)

which formats as:

 col1 col2 col3 col4 0 label1 label2 label2 label3 1 label2 label3 label1 label1 2 label3 label1 label3 label2

The idea would be to the count the indices of each label over all the columns into an array (or dataframe) as follows:

 label1 label2 label3 0 1 2 1 1 2 1 1 2 1 1 2

This tells that, for example, label1 appears once at index 0, twice at index 1 and once at index 2 in the original DataFrame.

I am performing this operation inside a loop so an efficient method would be preferred. Any ideas?

jezrael · Accepted Answer · 2019-01-25 12:39:13Z

3

Use:

df = df.apply(pd.value_counts, axis=1) print (df) label1 label2 label3 0 1 2 1 1 2 1 1 2 1 1 2

answered Jan 25, 2019 at 12:39

jezrael

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

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

Collectives™ on Stack Overflow

Counting the indices of Pandas DataFrame based on label location

1 Answer 1

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Linked

Related