CODEWITHSUNDEEP
pandas
cosmosa
cosmosa

Reputation: 761

Pandas: append column values as new rows

I have a dataframe that looks like this:

In [32]: df
Out[32]: 
   200003  200003.1
0     123       456

I want to transform it like this:

       200003
    0     123
    1     456

What is the easiest way to accomplish this?

Edit: df.stack() returns

0  200003      123
   200003.1    456
dtype: int64

My goal is to use df.to_dict('records') to be able to transform the df into a format suitable for machine learning like this [{'200003': '123'}, {'200003': '456'}]

Upvotes: 2

Views: 132

Answers (1)

splinter
splinter

Reputation: 3897

I think you are looking for DataFrame.stack().

Answer to your edit:

stacked = df.stack()
stacked.index = [df.columns[0]] * stacked.shape[0]

and then:

out = [ {j[0]:j[1]} for j in list(zip(stacked.index, stacked.values)) ]

gives you the list of dicts you are looking for.

Upvotes: 1

Related Questions