Data Structure Transformation

When I read a really terrible repository, it uses tensor and sparse matrix simultaneously for the same initial dataset. I’m so confused… and after I transform both of them into Numpy array, they are totally the same.

I tried many times to understand it because every time I got tired before I understood it… and after several days I finally tried some codes and found the tensor list and csr_matrix list are the same thing.

I think the following commands are useful.

csr_matrix to Numpy array

sp_matrixA.toarray()

tensor to Numpy array

tensorB.numpy()

Update on June 1st

I found when I turn another tensor into numpy array, .numpy goes wrong, while .detach().numpy() is OK. TO CLARRIFIED.

Numpy array to csr_matrix

import scipy.sparse as sp
sp.csr_matrix(arrayC)

Numpy array to tensor

import torch
torch.from_numpy(arrayD)

And another wonderful command to find whether 2 arrays/Numpy matrixs a and b are totally the same.

print((a==b).all())

If I want to see whether 2 arrays share some public elements,

print((a==b).any())

• Previous
  python argparse