CODEWITHSUNDEEP
pythonpandasnumpy
Martingale
Martingale

Reputation: 523

How to reshape an array of arrays in Python using Numpy

As you can see below I have created three arrays that contain different random numbers:

np.random.seed(200)
Array1  = np.random.randn(300)
Array2 = Array1 + np.random.randn(300) * 2
Array3 = Array1 + np.random.randn(300) * 2

data = np.array([Array1, Array2 , Array3])
#data.reshape(data, (Array3, Array1)
mydf = pd.DataFrame(data)
mydf.tail()

My objective is to build a DataFrame with those three arrays. Each array should show its values in a different column. The DataFrame should have three columns and the index. My problem with the above code is that the Dataframe is built in horizontal position instead of vertical position. The DataFrame looks like this:

enter image description here

I have tried to use the reshape function to reshape the numpy array called ”data” but I couldn’t make it work. Any help would be more than welcome. Thanks!

Upvotes: 1

Views: 164

Answers (1)

mcsoini
mcsoini

Reputation: 6642

You can use .T to transpose either the data data = np.array([Array1, Array2 , Array3]).T or the dataframe mydf = pd.DataFrame(data).T.

Output:

            0         1         2
295 -0.126758  1.697413  0.399351
296  0.548405  1.402154 -4.396156
297 -1.063243  0.279774 -0.636649
298 -0.678952 -2.061554  0.244339
299 -0.527970 -0.290680 -0.930381

Or build a 2D array right away

arr = np.random.randn(300, 3)
arr[:, 1:] *= 2

mydf = pd.DataFrame(arr)

Upvotes: 6

Related Questions