strisunshinewentingwang/pytorch

Pull Requests Commits

Dynamic quantization for bias. (#26057)

dskhudia committed 5 years ago

2b52c1d9

Clarified ambiguous docstring in NegativeBinomial

kyleengel committed 5 years ago

4a947b60

Add __s390x__ compiler define for s390 builds. (#26233)

abalib committed 5 years ago

327e94f5

Whiltelist and fusion support for quantized::linear - matmul(with bias) (#26204)

jerryzh168 committed 5 years ago

06c69ad8

Upgrade Caffe2 docker images to 306 to include roctracer and rocprofiler

bddppq committed 5 years ago

6f87a189

Kill defaults in nn.yaml. (#26282)

gchanan committed 5 years ago

ffbffb69

Disable broken unit tests (#26301)

smessmer committed 5 years ago

6df70db8

Add ProcessGroupGloo::createDefaultDevice (#26166)

pietern committed 5 years ago

f43a2c9c

Updating submodules

svcscm committed 5 years ago

7a7425cc

Whiltelist and fusion support for quantized::linear - matmul(without bias) (#26209)

jerryzh168 committed 5 years ago

fd3cc36f

Whiltelist and fusion support for quantized::linear - addmm (#26208)

jerryzh168 committed 5 years ago

f95d2b61

Move the CUDA implementation of round to ATen. (#25041)

xuhdev committed 5 years ago

c92ed8dd

Enabled conv methods for the bfloat16

izdeby committed 5 years ago

b6d1105e

Migrate away from using Variable( in test_nn.py (#26077)

rohan-varma committed 5 years ago

4e538ebc

fix hypothesis timeout (#26280)

smessmer committed 5 years ago

c0063560

Upgrade MKLDNN to v0.20.5 (#25757)

gujinghui committed 5 years ago

38b2bc14

Fix no auto batching bugs: cannot bulk load; not work with namedtuple (#26065)

ssnl committed 5 years ago

df9d8f90

Fix binary size of OpsAlreadyMovedToC10.cpp (#26237)

smessmer committed 5 years ago

24ae9b50

Switch to the new profiler infrastructure (#26174)

iotamudelta committed 5 years ago

976cefdb

Fix namedtensor ci (#26257)

zou3519 committed 5 years ago

91fc6f3b

Back out "[pytorch][PR] Refines test_torch.py generic device testing" (#26252)

Mike Ruberry committed 5 years ago

31139b5f

Fix CI (#26250)

smessmer committed 5 years ago

21ba320c

Fix Windows build (#26246)

smessmer committed 5 years ago

a2e5445f

Refines test_torch.py generic device testing (#26244)

Mike Ruberry committed 5 years ago

b6b2b4c1

Remove unboxedAutogradKernel from c10 (#26130)

smessmer committed 5 years ago

26d537d7

Call aten ops through c10 dispatcher (#23668)

smessmer committed 5 years ago

0e30e657

Use MIOpen for transpose convolutions (#26172)

lcskrishna committed 5 years ago

e86d99ae

Add a wrapper for inspect in JIT to produce better error message (#25415)

Dmytro Dzhulgakov committed 5 years ago

df338f80

Add type hint for cuda.set_rng_state (#26200)

yaroslavvb committed 5 years ago

7f3c4235

Ports most of test_torch.py to generic device type framework (#26232)

Mike Ruberry committed 5 years ago

b4b8f53a

Newer Older