strisunshinewentingwang/pytorch

Pull Requests Commits

[JIT] add support for lu_unpack (#33736)

Elias Ellison committed 4 years ago

f31b1d34

[JIT] add support for torch.lu to torchscript (#33724)

Elias Ellison committed 4 years ago

4543cf4e

[JIT] fix resolving of functions in torch/functional. fix compilation of torch.stft (#33504)

Elias Ellison committed 4 years ago

fddf7325

add support for _modules, reducing special casing of nn.Sequential (#29495)

Elias Ellison committed 4 years ago

057fd5e1

.circleci: Divert packages to test channel on tag (#33842)

seemethere committed 4 years ago

6eef66e1

port masked_fill from TH to ATen (#33330)

mingfeima committed 4 years ago

cd0acf43

ONNX Error Message on Missing Op (#33593)

lara-hdr committed 4 years ago

a0e90e1b

remove setStorage with null StorageImpl support. (#33735)

gchanan committed 4 years ago

02908dfa

Add partition info message to NetDef (#33616)

Yinghai Lu committed 4 years ago

04f88a3a

Revert D20010383: [jit] Unify augmented assign handling

David Riazati committed 4 years ago

51e40574

[jit] Unify augmented assign handling (#33578)

davidriazati committed 4 years ago

867990dc

Preserve Backward compatibility of models serialized before #31040 (#33796)

jerryzh168 committed 4 years ago

c32fa465

Add assert_tensor_equal and assert_tensor_not_equal to test/cpp/api/support.h (#30426)

Will Feng committed 4 years ago

5c33d98b

build: set -DNDEBUG in Release (#32719)

Baranowski committed 4 years ago

8aa09de1

.circleci: Switch to using robot token for conda uploads (#33786)

seemethere committed 4 years ago

93e30c16

Per channel quantization performance improvement (#33772)

zasdfgbnm committed 4 years ago

45e4b614

Fix grid_sample gradients at image borders (#32829)

bnehoran committed 4 years ago

f597ac6e

Fix examples with updated pruning naming convention (#33144)

mickypaganini committed 4 years ago

b8f0acf5

[pt][quant] Parallelize quantize and dequantize (#33765)

dskhudia committed 4 years ago

a8e7ed48

Migrate `fmod` and `fmod_` from TH to ATen (CPU) (#33592)

peterbell10 committed 4 years ago

2eb95d8f

Remove the use of macros in defining binary ops for base Vec256 (#33733)

xuhdev committed 4 years ago

f87b0b25

Fix deprecated python "add" calls (#33428)

xwang233 committed 4 years ago

c1dd7068

Feature/vonmises upstream (#33418)

ahmadsalim committed 4 years ago

24659d28

[Lite interpreter] Pass shared_ptr properly (#33667)

Martin Yuan committed 4 years ago

758ad516

[WIP] Reanimate gradient scaling API with original scale update heuristic (#33366)

definitelynotmcarilli committed 4 years ago

fc6a1536

Skip manual backward for `cdist` with case `p=2` (#31167)

Emilio Castillo committed 4 years ago

a836c4ca

pad_packed_sequence: doc improvement (#33768)

stas00 committed 4 years ago

9a5ea713

removed padding and dilation from LPPool2d Doc (#33714)

joerg-de committed 4 years ago

5bac7feb

Disable printing of the histogram when dump (#33749)

Haixin Liu committed 4 years ago

038ee013

[quant][graphmode][refactor] Factor out insertDequantCall (#33172)

jerryzh168 committed 4 years ago

86673791

Newer Older