strisunshinewentingwang/pytorch

Pull Requests Commits

skip test that is broken in head (#88759)

mikey dagitses committed 1 year ago

67af734a

print xpass (#89020)

clee2000 committed 1 year ago

175b7e1c

add `to(dtype)` support for all sparse compressed formats (#89055)

nikitaved committed 1 year ago

8dc3353b

Add test for out-of-bounds Tensor access on GPU (#39211)

malfet committed 1 year ago

da2afcb1

[inductor] Added bucketize to decomp table (#88348)

Fabio Rocha committed 1 year ago

d47b94fa

[inductor] Introduce CSEVariable type and use it to track if Triton variables are scalar (#88347)

Fabio Rocha committed 1 year ago

9262d18e

[torch] [analytics] add dynamo to analytics (#88915)

colin2328 committed 1 year ago

edd2dea8

[torch] [analytics] add pytorch event logger callsites to torch.save and torch.load (#89003)

colin2328 committed 1 year ago

3e2ba60a

Add range check to multi margin loss target (#89008)

malfet committed 1 year ago

d8466964

[torch] [analytics] add pytorch event logger callsites to transformers and encoder/decoders (#88896)

colin2328 committed 1 year ago

18c1f2f8

Add mem efficient backward (#88856)

drisspg committed 1 year ago

ff6d2a6d

[Inductor] Build FX Linear + Permute Vertical Fusion in Inductor (#88859)

jiawenliu64 committed 1 year ago

d60abe4b

Enable channels_last_3d on SyncBatchNorm (#88401)

xwang233 committed 1 year ago

f5df6850

[Profiler] Memory profiler part 3: Schema parsing and mutable arguments (#86854)

Taylor Robie committed 1 year ago

8023c9dc

[Profiler] Memory profiler part 2: Config validation (#86853)

Taylor Robie committed 1 year ago

2439bc1e

disable test that fails in fbcode (#88786)

mikey dagitses committed 1 year ago

279dcce7

[Profiler] Account for caching when assigning IDs (#88917)

Taylor Robie committed 1 year ago

1db0f735

Allow ROCm runners to have 2 or more gpus (#89011)

ZainRizvi committed 1 year ago

ee441238

[ROCm] Enable python ref executor UTs for ROCm (#88981)

pruthvistony committed 1 year ago

2819df9a

Rewrite assert statement with torch._assert under config (#88246)

tugsbayasgalan committed 1 year ago

62ba15e1

Symintify view_as_complex and view_as_real (#89052)

anjali411 committed 1 year ago

b815f1fc

[ao] quant_type.py fixing public v private (#87519)

HDCharles committed 1 year ago

b9029fc4

Symintify decomps for split and upsample_bilinear; Fix decomp for _softmax_backward_data and native_dropout_backward (#88761)

SherlockNoMad committed 1 year ago

5faa2792

[c10d] Implement `__instancecheck__` for `c10d::ReduceOp` (#88275)

crcrpar committed 1 year ago

63e16216

Update xnnpack graph schema to use xnode and xvalue (#89036)

cccclai committed 1 year ago

2452e3f9

Add debug handle to xnnpack schema (#89033)

cccclai committed 1 year ago

8c46a5de

Revert "Add mem efficient backward (#88856)"

pytorchmergebot committed 1 year ago

50c18217

Set correct size of `attr::output_layouts` when the graph has multiple outputs in JIT oneDNN fuser (#88496)

wenzhe-nrv committed 1 year ago

5314af53

Fix get_default_qat_qconfig for PT 1.13 (#88876)

peterjc123 committed 1 year ago

60e59c07

enable index_put test (#89019)

ngimel committed 1 year ago

5ed90c40

Newer Older