fixes printing non-contiguous tensors #10405
Conversation
ptrblck
requested review from
apaszke,
colesbury,
ezyang,
gchanan,
soumith and
zdevito
as code owners
|
Link to issue: #10402 |
|
Can we get a test? |
|
@ezyang Do you mean a specific test in import torch
import torch.nn as nn
batch_size = 10
channels = 3
h, w = 16, 16
x = torch.randn(batch_size, channels, h, w)
pool = nn.AvgPool2d((1, 2), (1, 2))
output = pool(x)
output = output.view(batch_size, channels, h//2, 2, w//2)
print(output.select(3, 1).mul_(-1))doesn't throw an error in the last line. If the former, sure, where should I add it? |
torch/_tensor_str.py
Outdated
| else: | ||
| return self | ||
| if self.size(0) > 2 * PRINT_OPTS.edgeitems: | ||
| self = self.contiguous() |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
|
@ptrblck For test, you can manually create a tensor and use |
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
colesbury has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
soumith is landing this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
Summary: Pull Request resolved: pytorch#10405 Differential Revision: D9302794 Pulled By: soumith fbshipit-source-id: e4a7db8d33400a5a050d05fd1679de8bc3cbcf30
6 participants
No description provided.