8000 Commits · YongLuoCode/pytorch · GitHub

More Web Proxy on the site http://driver.im/

Commits

Commits on Jul 30, 2017

Add ATen overload to AutoGPU. (#2234)

ezyang
authored and
soumith
committed

Commits on Jul 26, 2017

Opt into Trusty builds. (#2214)

ezyang
authored and
soumith
committed
Deduplicate THPUtils_checkLong/THPUtils_unpackLong (#2218)

ezyang
authored and
soumith
committed
Note [Undefined-dim versus 0-dim]
ezyang
committed

Commits on Jul 21, 2017

Warning squash.

ezyang
authored and
apaszke
committed

Commits on Jul 20, 2017

Comment that data of THStorage may be NULL.

ezyang
authored and
soumith
committed
Document how Numpy memory management works.

ezyang
authored and
soumith
committed
Convert Embedding to new style. (#1916)

ezyang
authored and
soumith
committed

Commits on Jul 19, 2017

Squash "macro expansion producing 'defined' has undefined behavior" warnings.

ezyang
authored and
apaszke
committed
Make clang shut up about class/struct mismatch.

ezyang
authored and
soumith
committed

Commits on Jun 30, 2017

Fix unused linker argument warnings. (#1958)

ezyang
authored and
soumith
committed

Commits on Jun 29, 2017

Make cudnn warnings clean. (#1940)

ezyang
authored and
soumith
committed

Commits on Jun 14, 2017

More accurate build instructions based on @apaszke's comments. (#1800)

ezyang
authored and
apaszke
committed
Make csrc -Werror clean. (#1795)

ezyang
authored and
soumith
committed

Commits on Jun 13, 2017

Better document how to rebuild only parts of the project. (#1796)

ezyang
authored and
soumith
committed
Support device on sparse tensor constructor, assert values/indices on same device.

ezyang
authored and
soumith
committed
Enforce storage is not NULL invariant for sparse tensors.

ezyang
authored and
soumith
committed
Add comment stating nDenseTensors != nTensors in checkGPU.

ezyang
authored and
soumith
committed
Do NOT clone indices/values passed to sparse tensor by default.

ezyang
authored and
soumith
committed
Add comment about new implementation being CPU-only.

ezyang
authored and
soumith
committed
Abort on known bug (#1521) for spcadd on non-coalesced.

ezyang
authored and
soumith
committed
More efficient nnz test in spcadd.

ezyang
authored and
soumith
committed
Test clone preserves uncoalescedness if it wasn't coalesced.

ezyang
authored and
soumith
committed
Add some docs about storage->Size.

ezyang
authored and
soumith
committed
Sparse tensor clone() preserves coalescedness.

ezyang
authored and
soumith
committed
Short-circuit copy if src and dest are equal.

ezyang
authored and
soumith
committed
Short-circuit copy if src and dest are equal.

ezyang
authored and
soumith
committed

Commits on Jun 10, 2017

Add support for NVTX functions. (#1748)

ezyang
authored and
apaszke
committed

Commits on Jun 2, 2017

A pile of misc doc fixes. (#1682)

ezyang
authored and
soumith
committed

Commits on May 25, 2017

How to manage multiple build trees of PyTorch. (#1654)

ezyang
authored and
soumith
committed

Commits on May 17, 2017

Rewrite 'How autograd encodes the history' to accurately describe current setup. (#1580)

ezyang
authored and
soumith
committed
Miscellaneous documentation around autograd. (#1577)
< A1F7 div class="pc-AvatarStackBody prc-AvatarStack-AvatarStackBody-mSyoJ" tabindex="0">
ezyang
authored and
soumith
committed

Commits on May 15, 2017

Fix flaxy test_sparse_adagrad (#1562)

ezyang
authored and
apaszke
committed
Fix memory leak in THCSTensor_spcadd. (#1519)

ezyang
authored and
soumith
committed

Commits on May 10, 2017

Reference counting documentation. (#1520)

ezyang
authored and
soumith
committed

0