Torch.linalg.lstsq vs Np.linalg.lstsq

Оба Torch.linalg.lstsq и Np.linalg.lstsq используют SVD, и

адресуют вопрос  https://yandex.ru/q/question/kak_ispolzovat_singuliarnoe_razlozhenie_e08cfb19/?answer_id=5988bbfa-d4c5-4926-9e57-b5cfb15814ea#5988bbfa-d4c5-4926-9e57-b5cfb15814ea

(.env)boris@boris-All-Series:~/TORCHLSTSQ$ cat torchLstsq.py

import numpy as np 

import torch

a = torch.tensor([[1., 1, 1],

                  [2, 3, 4],

                  [3, 5, 2],

                  [4, 2, 5],

                  [5, 4, 3]])

b = torch.tensor([[-10., -3],

                  [ 12, 14],

                  [ 14, 12],

                  [ 16, 16],

                  [ 18, 16]])

a1 = a.clone().numpy()

b1 = b.clone().numpy()

x = torch.linalg.lstsq(b, a).solution

x1, res, r1, s = np.linalg.lstsq(b1, a1,rcond = -1)

print(f'torch_x: {x}')

print(f'torch_r1: {r1}\n')

print(f'np_x: {x1}')

print(f'np_res: {res}')

print(f'np_r1(rank): {r1}')

print(f'np_s: {s}')

(.env) boris@boris-All-Series:~/TORCHLSTSQ$ python3 torchLstsq.py

torch_x: tensor([[-0.1145, -0.1047, -0.2863],

        [ 0.3591,  0.3372,  0.5407]])

torch_r1: 2

np_x: [[-0.11452514 -0.10474861 -0.28631285]

 [ 0.35913807  0.33719075  0.54070234]]

np_res: [ 5.4269753 10.197526   1.4185953]

np_r1(rank): 2

np_s: [43.057705  5.199417]

References

https://pytorch.org/docs/stable/generated/torch.linalg.lstsq.html