i have a question in your Exercises

[navgod]

1. in your SVD exercise, you said Left-singular vectors of P = eigenvectors of PPT . when i try ,
   import torch P = torch.tensor([[25, 2, -5], [3, -2, 1], [5, 7, 4.]]) E_cplx,V_cplx=torch.linalg.eig(torch.matmul(P,P.T)) V = V_cplx.float() E = E_cplx.float() torch.linalg.svd(P) V
   and then when i compared with them, the Left-singular vectors of P is
   tensor([[-0.9757, 0.1823, 0.1214],
   [-0.0975, 0.1350, -0.9860],
   [-0.1961, -0.9739, -0.1140]])
   and eigenvectors of PP.T is
   tensor([[ 0.9757, 0.1823, 0.1214],
   [ 0.0975, 0.1350, -0.9860],
   [ 0.1961, -0.9739, -0.1140]])
   there are some different

2. in your the Moore-Penrose Exercise, you say use torch.svd() but when i use torch.svd() U is differnt from example and got some error but when i use torch.linalg.svd() it works well what is difference between them?
   i don't wanna bother you but in my udemy's korean subtitle version of your lesson there's no reply