strisunshinewentingwang/pytorch

Pull Requests Commits

Create prototype for AST rewriter (#47216)

Ansley Ussery committed 3 years ago

dec1c364

[Gradient Compression] Surface C++ comm hooks to Python API as built-in comm hooks (#47270)

Yi Wang committed 3 years ago

f91fcefc

Optimize arguments checks (#46661)

Iurii Zdebskyi committed 3 years ago

2652f2e3

Inlining all non-output buffers, including intermediate buffers. (#47258)

navahgar committed 3 years ago

2caa3bd4

[vulkan] Add mean.dim op for vulkan (#47312)

SS-JIA committed 3 years ago

464c569d

[TensorExpr] Pick meaningful names for functions in TE codegen. (#47255)

Mikhail Zolotukhin committed 3 years ago

9b168a1f

[TensorExpr] CudaCodegen: restart counter for function names unique ID inside each codegen instantiation. (#47254)

Mikhail Zolotukhin committed 3 years ago

a65e7570

[JIT] SubgraphUtils: add a function for generating a string name for a given graph. (#47253)

Mikhail Zolotukhin committed 3 years ago

3161fe6d

Codegen - error when an argument that looks like an out argument isn't a kwarg (fix #43273) (#47284)

bdhirsh committed 3 years ago

7a0f0d24

Provide 'out' parameter for 'tensordot' (#47278)

H-Huang committed 3 years ago

a8ef4d3f

[quant] Quantized flip dispatch (#46235)

z-a-f committed 3 years ago

31ebac3e

Implement bicubic grid sampler (#44780)

pomelyu committed 3 years ago

f41f3e3c

[numpy] `torch.a{cosh, sinh}` : promote integer inputs to float (#47152)

kshitij12345 committed 3 years ago

63978556

[Pytorch][Annotation] Update inlined callstack with module instance info (#46729)

cccclai committed 3 years ago

2b5433de

[NNC] Implement Cond in LLVM codegen (#47256)

Cheng Chang committed 3 years ago

f730f259

Event Logging for NCCL Async Error Handling Process Crash (#47244)

osalpekar committed 3 years ago

8b13ab93

Update minimum supported Python version to 3.6.2 (#47314)

malfet committed 3 years ago

ca61b061

Add comment explaining purpose of the accumulate_grad argument (#47266)

soulitzer committed 3 years ago

ea93bdc2

[JIT] Print out interface mismatch for prim::ModuleDictIndex (#47300)

Meghan Lele committed 3 years ago

dc0d68a1

Embed `libiomp5.dylib` into wheel package (#47262)

malfet committed 3 years ago

14194e4f

[numpy] `torch.a{cos, tan}` : promote integer inputs to float (#47005)

kshitij12345 committed 3 years ago

c424d938

[numpy] `torch.{a}tanh` : promote integer inputs to float (#47064)

kshitij12345 committed 3 years ago

0d00724e

[fix] nn.Embedding.from_pretrained : honour `padding_idx` argument (#47184)

kshitij12345 committed 3 years ago

c68c3d0a

Added Kronecker product of tensors (torch.kron) (#45358)

IvanYashchuk committed 3 years ago

f276ab55

reorganize sparse_nn_partition (#47283)

scottxu0730 committed 3 years ago

32b66b08

Change largeCUDATensorTest to largeTensorTest+onlyCUDA; add a buffer to large cuda tensor test (#45332)

xwang233 committed 3 years ago

774b638e

Optimize backward for torch.repeat (#46726)

ejguan committed 3 years ago

4e6f2440

Update doc to reflect current behavior (#46937)

albanD committed 3 years ago

9c3a7552

fix windows CI passed incorrectly (#47105)

mszhanyi committed 3 years ago

782f92b5

Automated submodule update: FBGEMM (#47263)

facebook-github-bot committed 3 years ago

8c865493

Newer Older