strisunshinewentingwang/pytorch

Pull Requests Commits

Enable mkldnn on windows (#31355)

XiaobingSuper committed 4 years ago

19bb496a

[ROCm] Enable Caffe2 video operators for ROCm

lcskrishna committed 4 years ago

957a07ff

[rpc] make handling of FORWARD_AUTOGRAD_REQ in request_callback_impl (#32476)

rohan-varma committed 4 years ago

5b321a09

Make cuda search process of cpp extension quiet (#32620)

peterjc123 committed 4 years ago

1e5aead3

faster bailout tests (#32266)

Krovatkin committed 4 years ago

8fbe1ccd

Bug fix of norm minimization for dev mode (#31462)

csummersea committed 4 years ago

12d59339

Add warning regarding pickle insecurity on torch.load documentation (#32593)

andfoy committed 4 years ago

90a259e1

Update linspace types (#32218)

Enealor committed 4 years ago

3bbb36e0

Fix MagmaInitializesCorrectly_CUDA by using an invertible matrix (#32547)

Charles Hofer committed 4 years ago

5fd037ce

Fix wrong typing (torch/nn/parameter.pyi) (#32617)

jeongukjae committed 4 years ago

320d1a15

[pytorch] codegen flags to whitelist op registrations / generate to separate files (#32451)

ljk53 committed 4 years ago

69283388

[pytorch] move type_derived_methods out of anonymous namespace (#32275)

ljk53 committed 4 years ago

0afe1950

[caffe2] use JIT'ed fp32 SLS (#32413)

jspark1105 committed 4 years ago

bd20274e

Support TorchScript call over remote API (RRef) (#32466)

xush6528 committed 4 years ago

6ad9e5c7

[aten] fix shadowing variable warning (#32573)

jspark1105 committed 4 years ago

e0ffe726

Add operator support for dynamic quant on mobile (#32479)

supriyar committed 4 years ago

16954187

Fix iterator for ncclCommWatchdog. (#32571)

pritamdamania committed 4 years ago

59dbece3

[pytorch][refactor] Explicitly use auto* for pointers (#32548)

jianyuh committed 4 years ago

1218a16a

[jit] Cloning constants in ClassType (#32371)

jerryzh168 committed 4 years ago

e7edc5f2

[docs] Change fut.wait() to torch.jit._wait(fut) in jit overview docs (#32336)

rohan-varma committed 4 years ago

666472a3

duplicate symbols with AT_PARALLEL_OPENMP=0 (#32568)

EscapeZero committed 4 years ago

6412ca3c

[quant][graphmode][refactor] Better API for fold_convbn (#32380)

jerryzh168 committed 4 years ago

91f10a1d

add diag into pt operator microbenchmark (#32597)

hl475 committed 4 years ago

52f8f031

[pytorch] change op dependency output to use double-quoted strings (#32464)

ljk53 committed 4 years ago

9e0ce72e

[refactor] Adding FoldConvBatchNorm2dHelper (#32374)

jerryzh168 committed 4 years ago

2bfd33b4

[pytorch] Minor: boilerplate to propagate errors in request_callback_impl (#32556)

jjlilley committed 4 years ago

573a3027

Make autogen functions correct for multiple outputs and views (#31990)

albanD committed 4 years ago

3ab30753

fix view listing in autograd codegen (#32044)

albanD committed 4 years ago

9e59244b

[quant][graphmode] Call _jit_pass_dedup_module_ueses in quantize_script (#32303)

jerryzh168 committed 4 years ago

d2bda53f

[quant] Re-enable fold_convbn in quantize_script (#32302)

jerryzh168 committed 4 years ago

fe3eb09d

Newer Older