projects / platform / upstream / pytorch.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 4047cdc) | patch
Concatenate directly into shared memory when constructing batches for numpy (#14534)
authorChristoph <cbj@mail.uni-paderborn.de>
Sun, 30 Dec 2018 01:48:36 +0000 (17:48 -0800)
committerFacebook Github Bot <facebook-github-bot@users.noreply.github.com>
Sun, 30 Dec 2018 01:51:02 +0000 (17:51 -0800)
commit2a45050fdcf1634a3a931390c0476b127e63a892
tree750a0d2f63877b244ec6afd34124bcfd1a58ce1btree | snapshot
parent4047cdc6906c6e5c5e1fb837f7f479e0af149fa7commit | diff
Concatenate directly into shared memory when constructing batches for numpy (#14534)

Summary:
Since #1323 tensors are shared with shared memory, but this feature is not active for numpy.
This PR fix this.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14534

Differential Revision: D13561649

Pulled By: soumith

fbshipit-source-id: b6bc9e99fb91e8b675c2ef131fba9fa11c1647c0
test/test_dataloader.py diff | blob | history
torch/utils/data/_utils/collate.py diff | blob | history
Domain: Machine Learning / ML Framework;