How to slice a python pandas dataframe with 1200 rows into 12 equal parts?

Question

How to slice a python pandas data-frame with 1200 rows into 12 equal parts? either in Python2 or Python 3

use `np.array_split(df, 100)` – akash karothiya Jan 13 '17 at 06:16 — akash karothiya, Jan 13 '17 at 06:16
change 100 to 12, try this `np.array_split(df, 12)` – akash karothiya Jan 13 '17 at 06:28 — akash karothiya, Jan 13 '17 at 06:28
Love this comment thread :) – Paula Livingstone Nov 08 '17 at 23:36 — Paula Livingstone, Nov 08 '17 at 23:36
Love this comment thread :) too – You Gakukou Nov 24 '17 at 01:38 — You Gakukou, Nov 24 '17 at 01:38

score 2 · Answer 1 · answered Jan 13 '17 at 08:36

You can do it several ways. I'd use groupby and a dictionary comprehension. Even then, there are two obvious and distinct ways to split it.

Consider the dataframe df

df = pd.DataFrame(dict(A=np.arange(1200)))

contiguous
Meaning, grab the first 100, then the next, so on and so forth

twelve_equal_dfs_contiguous = \
    {name: group for name, group in df.groupby(np.arange(1200) // 100)}

stratified
Meaning, grab every other 100 starting with the first. Then repeat starting with the second, so on and so forth

twelve_equal_dfs_stratified = \
    {name: group for name, group in df.groupby(np.arange(1200) % 100)}

There are too many ways to do this. Hopefully this is some guidance on where to start.

1 Answers1