List of values from dataframe column where each row has muliple values-CodePudding

I have a dataframe that looks like this

index     column

A         41 13 4 61 12 35

B         16 35 56 24

C         12

And I want to end up with a long list of all values like [41, 13, 4, 61, 12, 35, 16, 35, 56, 24, 12]

So first I converted that column into a dict with DataFrame.to_dict()

And then what I was trying to do was to split each dict value into a list rather than a long string:

for key,val in d.items():
    d[key] = val.split[' ']

but it's throwing an error: TypeError: 'builtin_function_or_method' object is not subscriptable

Then I would proceed to append all the values into a long list. But given the error, I'm suspecting that there is a simpler way that I am missing. Does anyone know what that could be?

CodePudding user response：

split the column then use hstack to flatten

np.hstack(df['column'].str.split()).tolist()

Alternatively you can use a pure python approach

from itertools import chain

list(chain(*map(str.split, df['column'])))

['41', '13', '4', '61', '12', '35', '16', '35', '56', '24', '12']