strisunshinewentingwang/pytorch

Pull Requests Commits

pin multipy (#90942)

PaliC committed 1 year ago

d3e0bcc7

Make it easier to find troubleshooting steps (#90948)

ZainAlt committed 1 year ago

d8c1872c

fix segfault for EmbeddingBag on CPU slow path when include_last_offset is true (#90358)

mingfeima committed 1 year ago

9d523616

[Quant] Add fused LinearLeakyReLU module for onednn backend (#88661)

Xia-Weiwen committed 1 year ago

353c2e7d

Revert "Include dispatch key in wrapper symbol name (#90674)"

pytorchmergebot committed 1 year ago

750576a5

Make dynamo.export preserve user input/output format (#90884)

tugsbayasgalan committed 1 year ago

f660d62d

Revert "[JIT] Frozen Graph Linear-BatchNormNd Folding (#86706)"

pytorchmergebot committed 1 year ago

31b8dc75

Suppress RecursionError in sympy; fix logging (#90904)

ezyang committed 1 year ago

535b0e37

Revert "Add macro C10_AS_INTARRAYREF_SLOW (#90675)"

pytorchmergebot committed 1 year ago

140a3139

[ao][fx] fixing public v private prepare.py (#88398)

HDCharles committed 1 year ago

9259933e

Reset rng in hf before loading a model (#90936)

mlazos committed 1 year ago

f3da157c

[FSDP] Fix input grad propagation when using param mixed precision (#90921)

awgu committed 1 year ago

d04e3c99

Revert "[quant][pt2e] Add early prototype top level quantize_pt2e APIs (#90802)"

pytorchmergebot committed 1 year ago

9c912c7d

[inductor] Use --continue_on_fail when installing torchbench (#90922)

desertfire committed 1 year ago

fdc97330

[Dynamo][TIMM][Benchmarks] Fix TIMM `0.8.0dev` breaking the `timm_models.py` script's data config (#90404)

eqy committed 1 year ago

57e2090e

If a torch.* returns non-Tensor, make this unimplemented rather than assert. (#89918)

ezyang committed 1 year ago

e686a442

[quant][pt2e] Add early prototype top level quantize_pt2e APIs (#90802)

jerryzh168 committed 1 year ago

a66af1fe

Hack get_nbytes() to return 0 for sparse tensors as workaround for functionalization (#90702)

jbschlosser committed 1 year ago

201c36d8

Error messages for kernel selection (#90783)

mikekgfb committed 1 year ago

15c9df77

[ao][fx] fixing public v private qconfig_mapping_utils.py (#88399)

HDCharles committed 1 year ago

173accd1

Revert "Revert "[functorch] Refactor life handle storage (#90317)"" (#90856)

zou3519 committed 1 year ago

abc54f93

[inductor] Prevent blowup in inner_fn_str and extract_read_writes (#88933)

peterbell10 committed 1 year ago

81f351ac

[FSDP] Enable mixed hybrid/non-hybrid sharding strategies (#90846)

awgu committed 1 year ago

c4718e9b

[FSDP][Easy] Use `run_subtests` for hybrid shard test (#90859)

awgu committed 1 year ago

2f8c0cb2

replicate state_dict tests (#90868)

rohan-varma committed 1 year ago

b92975a6

[inductor] add conv_transpose2d unary fusion for cpu in inference mode (#90265)

chunyuan-w committed 1 year ago

d6fe9838

add conv_transpose2d pointwise(unary) fusion kernel (#90264)

chunyuan-w committed 1 year ago

85698d0a

[Quant] onednn backend switch to ideep new api without affacting performance (#90354)

Xia-Weiwen committed 1 year ago

9b89ff09

[CUDA 12] Fix the endif guard position for cusparse const descriptors (#90897)

xwang233 committed 1 year ago

79009cbc

[Composable API] `replicate`: cleanup _ddp.py (#90257)

Charlie Yan committed 1 year ago

98799ca0

Newer Older