How to convert the list of Dataframes in this required shape of array?

Question

I have a list of dataframes which looks like shown in image.

The list length is 200 where it has 200 dataframes, each dataframe has length of 205 and has 6 columns.

I want to bring this list in the array of shape (200,205,6)

I tried it doing follwing method but I keep getting the error given below.

np.array(samples).reshape(200,205,6)

For this code I am getting an error:

ValueError: cannot copy sequence with size 205 to array axis with dimension 1

I also removed 5 columns and kept only one column in all datarames but yet am getting same error.

Is there any way to bring it in the required shape..enter image description here

still not very clear what you want to do - do you want to take all the values in the dataframes and create a nested array from the dataframes? — Chinny84, Jun 25 '19 at 14:44

score 0 · Answer 1 · answered Jun 25 '19 at 14:56

You're getting that error because numpy doesn't know how to convert a pandas dataframe into an np.array, but you can solve this pretty quickly:

np.array([df.values for for df in list_of_dataframes])

I tried it with this sample data:

data = [pd.DataFrame(np.random.random((205,6))) for i in range(200)]

np.array([df.values for df in data]).shape

(200, 205, 6)

1 Answers1