strisunshinewentingwang/pytorch

Pull Requests Commits

Revert D18499600: Add overload name to JIT prim operators.

zdevito committed 4 years ago

a5b4d78c

Revert D18499601: Add missing operators for PyText model.

zdevito committed 4 years ago

2a442f5d

add cuda sync when ops running on gpu (#29936)

mingzhe09088 committed 4 years ago

c5430345

fix missing lock in profiling graph compilation (#29886)

zdevito committed 4 years ago

f1860aea

Enable full error message for mobile builds (#29926)

linbinyu committed 4 years ago

5cad7d42

Turn off scalar_check for diag.

gchanan committed 4 years ago

c300f086

Turn off scalar_check for _th_max, _th_min.

gchanan committed 4 years ago

a6a31c6d

Turn off scalar_check for lstsq (gels), and test scalars for eig.

gchanan committed 4 years ago

6c7a0c68

Turn off scalar_check for sort. (#29874)

gchanan committed 4 years ago

79f06367

Fix memory leak in CUDA renorm, turn off scalar_check for renorm. (#29873)

gchanan committed 4 years ago

ee5201cd

Turn off scalar_checks for cumsum, cumprod.

gchanan committed 4 years ago

d87655f5

Turn off scalar_check for fmod. (#29871)

gchanan committed 4 years ago

fe575b44

Turn off scalar_check for remainder. (#29870)

gchanan committed 4 years ago

98362977

Turn off scalar_checks for multinomial_alias_setup_, which requires 1d tensors. (#29869)

gchanan committed 4 years ago

61df98a0

Stop generating maybe_zero_dim calls for "scalar_check: false" with multiple outputs. (#29868)

gchanan committed 4 years ago

92a512b5

Add missing operators for PyText model.

Martin Yuan committed 4 years ago

6c39e503

Add overload name to JIT prim operators.

Martin Yuan committed 4 years ago

ff4e782e

OPN ops TupleConstruct/Unpack and format. (#29635)

Martin Yuan committed 4 years ago

3003c5f9

Update fbjni and enable PyTorch JNI build

dreiss committed 4 years ago

d22f6143

fix device check in op bench (#29918)

mingzhe09088 committed 4 years ago

3f5dc95b

Updating submodules

svcscm committed 4 years ago

acb81008

Add TensorShapeAndType (#29848)

ChunliF committed 4 years ago

7807d449

Stop binding _th_resize_as_, which isn't used anymore.

gchanan committed 4 years ago

5ab6635d

Skip outputting scalar_checks if they are false. (#29866)

gchanan committed 4 years ago

8e61287d

Avoid keeping old histograms in the histogram observer to fix the OOM issue (#29768)

csummersea committed 4 years ago

4442fa59

Add `torch.version.hip` from cmake (#29815)

t-vi committed 4 years ago

7889e1e3

Expose is_signed for dtype (#29511)

vishwakftw committed 4 years ago

69e343f2

Revert "switch back to azure pipelines" (#29910)

suo committed 4 years ago

23fcc409

Set TORCH_CXX_FLAGS in minimal example (#29890)

take-cheeze committed 4 years ago

a9c719ba

Makes test_type_promotion generic (#29417)

mruberry committed 4 years ago

9ec1727e

Newer Older