strisunshinewentingwang/pytorch

Pull Requests Commits

Name refactoring to make MemoryPlanning more readable

Peng Wu committed 3 years ago

28a3a7c7

[iOS GPU] Add the reset of binary ops (#53950)

xta0 committed 3 years ago

bb21aea3

[iOS GPU] Support element-wise broadcasting for binary ops in shaders (#53949)

xta0 committed 3 years ago

530dc828

[testing] assert no duplicate in method_tests for an OpInfo entry (#53492)

kshitij12345 committed 3 years ago

df7c0a06

Fix restriding logic for structured kernels (#53759)

ezyang committed 3 years ago

547f4357

Add meta device to generic device testing framework, skip NotImplementedError (#53682)

ezyang committed 3 years ago

c2f41b6b

Add 'noarch' tests which only run in one CI config (#53747)

ezyang committed 3 years ago

d47d2462

Clean up future imports for Python 2 (#53349)

skyline75489 committed 3 years ago

f6df18f6

Skips test_linalg_lstsq on ROCm (#53977)

Mike Ruberry committed 3 years ago

319ab58e

Fixed the size of the workspace array in functions calling LAPACK (#53909)

IvanYashchuk committed 3 years ago

790326d4

Added OpInfo-based testing of some linalg functions (#51107)

IvanYashchuk committed 3 years ago

7df176b1

Refines test_orgqr_* skip (#53975)

Mike Ruberry committed 3 years ago

d46978cc

matmul performance benchmarks (#51647)

aocsa committed 3 years ago

39f50f46

increase timeout for test_op_nnpi_fp16

hl475 committed 3 years ago

142c6b0e

Refactor ForeachUtils.h (#51131)

Iurii Zdebskyi committed 3 years ago

84af0c7a

Make ideep honor `torch.set_num_thread` changes (#53871)

malfet committed 3 years ago

f2689b1e

Clang format default_hooks.py (#53956)

Yi Wang committed 3 years ago

de70cdb6

[Gradient Compression] Update the docstring of fp16_compress_wrapper (#53955)

Yi Wang committed 3 years ago

ca4aae85

remove th_fill and th_mul dead code (#52546)

XiaobingSuper committed 3 years ago

3ce51fd5

[TB] Support writing new style scalar (#53496)

cryptopic committed 3 years ago

317ff429

[NNC] New APIs to get loops corresponding to a Buf (#53778)

navahgar committed 3 years ago

ef07a040

Implemented embedding_bag for SR (#52429)

Edvard Ghazaryan committed 3 years ago

ce0fd095

[Gradient Compression] Make FP16 compression as a wrapper that can be combined with other communication hooks (#53808)

iseessel committed 3 years ago

3078233e

[caffe2] Don't call TensorImpl::size() in dim32() (#53852)

swolchok committed 3 years ago

8a5b946f

Revert D26922420: [ONNX] fix export of embedding with padding_idx (#53053)

malfet committed 3 years ago

5b648ef9

[reland] Add OpInfo for `bitwise_not` (#53181)

kshitij12345 committed 3 years ago

00771eff

Added cuBLAS path for torch.triangular_solve (#53147)

IvanYashchuk committed 3 years ago

fe086717

Implements `torch.linalg.lstsq` (#49093)

nikitaved committed 3 years ago

afa1ff8e

[Static Runtime] Fix bug in ClipRangesGatherRangesX2SigridHash (#53799)

Hao Lu committed 3 years ago

49323423

Revert D26993790: [pytorch][PR] [CUDA graphs] Private mempools for CUDA graphs

ngimel committed 3 years ago

76129c7c

Older