Convert grouped array elements from Pandas index into individual array elements in Python

Question

I have a 1x100 1D array that looks like :

(A1,B1,C1,D1),(A2,B2,C2.D2),(A3,B3,C3,D3) ..... (A100,B100,C100,D100)

I want to convert this array into a 4x100 2D array that looks like:

A1, B1, C1, D1
A2, B2, C2, D2
A3, B3, C3, D3
A4, B4, C4, D4

....

A100, B100, C100, D100

I tried using Convert a 1D array to a 2D array in numpy to reshape the array, but that doesn't solve the problem of turning my grouped array elements into separate array elements.

EDIT

My dataset was created from a normal Pandas dataframe with 6 columns.

df = pd.pivot_table(concord, index = ['Col1','Col2','Col3', 'Col4'], columns 
= 'Col5', values = 'Col6')

This produces an index comprised of 4 columns on a dataframe with as many columns as rows in Col5. I then produce my grouped array with.

df.index.values

Here is an example of what I am dealing with:

df = pd.DataFrame(np.random.randn(6,4),columns=list('ABCD'))
df = pd.pivot_table(df, index = ['A','B'], columns = 'C', values = 'D')
df.index.values

How exactly does your array look like when you print it? What you show is not a valid output from Python. It looks like your "grouped array elements" are tuples. You'd have to convert those into arrays first. — Joooeey, Dec 04 '17 at 20:34
my array looks like array([('A', 'B', 318443.998, 398601.13),('C', 'D', 318443.998, 398601.13) ..... ]). I know it is an array because Python says its type is an array. It has strings inside it though if that is a concern. I got it by using the values function on the index of a pandas dataframe that combined four columns into a single index. — fume_hood_geologist, Dec 04 '17 at 20:43
Coming from `pandas` it is probably an `object` dtype array, one `group` per element. — hpaulj, Dec 04 '17 at 21:17

MaxU - stand with Ukraine · Accepted Answer · 2017-12-04T22:13:32.783

1

UPDATE:

In [156]: df.index.values
Out[156]:
array([(-1.1589142652108702, 0.9058310711560317), (-0.8998389109349003, -1.3963768690741623), (-0.39909436307375407, 0.19524579552938015), (
0.3879397588199493, -1.6231489531567047),
       (0.4850794016257364, 0.9079649929281776), (1.263583531683934, -0.8985146123002069)], dtype=object)

In [157]: np.array(df.index.values.tolist())
Out[157]:
array([[-1.15891427,  0.90583107],
       [-0.89983891, -1.39637687],
       [-0.39909436,  0.1952458 ],
       [ 0.38793976, -1.62314895],
       [ 0.4850794 ,  0.90796499],
       [ 1.26358353, -0.89851461]])

Old answer:

I think you want to use Pandas:

In [89]: a
Out[89]:
array([['A', 'B', '318443.998', '398601.13'],
       ['C', 'D', '318443.998', '398601.13']],
      dtype='<U10')

In [90]: df = pd.DataFrame(a)

In [91]: df
Out[91]:
   0  1           2          3
0  A  B  318443.998  398601.13
1  C  D  318443.998  398601.13

edited Dec 04 '17 at 22:13

answered Dec 04 '17 at 20:52

MaxU - stand with Ukraine

205,989
36
386
419

The array I have was originally produced in Pandas. Converting it into a data frame just gives me a dataframe with a single column where the elements are each grouping of items i want to separate, leaving me back where i started – fume_hood_geologist Dec 04 '17 at 20:54
@LeonardShaw, can you provide a __reproducible__ data sample? When using a sample from your comment i was able to generate a DF of 2x4 shape – MaxU - stand with Ukraine Dec 04 '17 at 20:57
@LeonardShaw, I asked for a sample data set - not for an explanation how did you get it from the DF, which we can't see/access... – MaxU - stand with Ukraine Dec 04 '17 at 21:13
df = pd.DataFrame(np.random.randn(6,4),columns=list('ABCD')) df = pd.pivot_table(df, index = ['A','B'], columns = 'C', values = 'D') df.index.values – fume_hood_geologist Dec 04 '17 at 21:19
I added a sample data set – fume_hood_geologist Dec 04 '17 at 21:21
@LeonardShaw, please see UPDATE section – MaxU - stand with Ukraine Dec 04 '17 at 22:13

Convert grouped array elements from Pandas index into individual array elements in Python

1 Answers1