diff options
| author | Tongzhou Wang <SsnL@users.noreply.github.com> | 2018-05-23 11:03:12 -0400 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2018-05-23 11:03:12 -0400 |
| commit | e3e15b5d9534f1c10170e169f1423e9298648e86 (patch) | |
| tree | d74cb50a982ca753d74e539a27d66731189130e2 /test/common_nn.py | |
| parent | 6a604f16cc452424c58ce456da7f46e1619b9e18 (diff) | |
| download | pytorch-e3e15b5d9534f1c10170e169f1423e9298648e86.tar.gz pytorch-e3e15b5d9534f1c10170e169f1423e9298648e86.tar.bz2 pytorch-e3e15b5d9534f1c10170e169f1423e9298648e86.zip | |
[PyTorch] [gradcheck] change backward() to grad() (#7710)
* Change backward calls to grad to avoid memory leak from #7343; Replace unnecesary create_graph=True with retain_graph=True
* fix gradgradcheck use of make_non_contiguous
* allow non-contguous target
* remove unnecessray .grad.zero_()
* remove contiguous_detach
* fix PReLU double backward always returning ggW as a scalar
* let noncontig gO require grad
* move requires_grad to return
Diffstat (limited to 'test/common_nn.py')
| -rw-r--r-- | test/common_nn.py | 9 |
1 files changed, 4 insertions, 5 deletions
diff --git a/test/common_nn.py b/test/common_nn.py index 51b883ba9e..48a5211296 100644 --- a/test/common_nn.py +++ b/test/common_nn.py @@ -7,7 +7,7 @@ from itertools import product import torch import torch.cuda from common import TestCase, to_gpu, freeze_rng_state, is_iterable -from torch.autograd.gradcheck import get_numerical_jacobian, iter_tensors, contiguous +from torch.autograd.gradcheck import get_numerical_jacobian, iter_tensors import torch.backends.cudnn # tarfile module tries to obtain a file object name in python 3.3 @@ -783,9 +783,8 @@ class NNTestCase(TestCase): return self._forward(module, input).detach() res = tuple() - input = contiguous(input) if jacobian_input: - res += get_numerical_jacobian(fw, input, input, eps=1e-6), + res += get_numerical_jacobian(fw, input, eps=1e-6), if jacobian_parameters: param, _ = self._get_parameters(module) res += torch.cat([get_numerical_jacobian(fw, input, p, eps=1e-6) for p in param], 0), @@ -813,8 +812,8 @@ class NNTestCase(TestCase): input_t = iter_tensors(input) numerical_t = iter_tensors(numerical_d_x) for x, d_x in zip(input_t, numerical_t): - x = x.view(-1) - d_x = d_x.view(-1) + x = x.view(-1).data + d_x = d_x.view(-1).data for i in range(x.nelement()): original = x[i].item() x[i] = original + eps |