strisunshinewentingwang/pytorch

Pull Requests Commits

Fixing a few links in distributed CONTRIBUTING.md (#44753)

pritamdamania committed 4 years ago

dbf17a1d

CUDA BFloat16 pow (#44760)

zasdfgbnm committed 4 years ago

06036f76

Add a test to ensure DDP join works with RPC (#44439)

rohan-varma committed 4 years ago

63469da3

[quant][qat] Ensure observers and fq modules are scriptable (#44749)

supriyar committed 4 years ago

3f512b0d

[CI] Add profiling-te benchmarks. (#44756)

Mikhail Zolotukhin committed 4 years ago

b85568a5

[TensorExpr] Fuser: try merging adjacent fusion groups. (#43671)

Mikhail Zolotukhin committed 4 years ago

d66520ba

lr_schedule.py redundant code (#44613)

gauenk committed 4 years ago

2efc618f

[fx] remove delegate, replace with tracer (#44566)

zdevito committed 4 years ago

2c1b215b

Convert num_kernels to int64 before calling into CUDA GET_BLOCKS (#44688)

soulitzer committed 4 years ago

993b4651

Revert D23583017: move rebuild buckets from end of first iteration to beginning of second iteration

Ailing Zhang committed 4 years ago

fb085d90

[WIP][JIT] Add benchmarking support of NV Fuser with FP16 dtype support (#44101)

kevinstephano committed 4 years ago

26a91a9f

[jit] Speed up saving in case of many classes (#44589)

Dmytro Dzhulgakov committed 4 years ago

2f4c31ce

Enable fp16 for UniformFill (#44540)

Yan Xie committed 4 years ago

285ba0d0

[NNC] make inlining immediate (take 3) (#44231)

nickgg committed 4 years ago

69839ea3

Fix fallback graph in specialize autogradzero (#44654)

Elias Ellison committed 4 years ago

8df0400a

Enable fp16 for CUDA SparseLengthsSum/Mean (#44089)

Yan Xie committed 4 years ago

4ce6af35

Run vmap tests in CI (#44656)

zou3519 committed 4 years ago

07cba8b1

ci: Add anaconda pruning to CI pipeline (#44651)

seemethere committed 4 years ago

d62994a9

[docs] torch.min/max: remove incorrect warning from docs (#44615)

kshitij12345 committed 4 years ago

1d733d66

Use amax/maximum instead of max in optimizers (#43797)

zasdfgbnm committed 4 years ago

6bc77f4d

Fix doc builds for bool kwargs (#44686)

muthuArivoli committed 4 years ago

9c364da9

move rebuild buckets from end of first iteration to beginning of second iteration (#44326)

zhaojuanmao committed 4 years ago

f5d231d5

qat conv_fused.py: one more patch for forward compatibility (#44671)

vkuzo committed 4 years ago

5f692a67

Upgrade oneDNN (mkl-dnn) to v1.6 (#44706)

pinzhenx committed 4 years ago

72b5665c

Revert D23323486: DPP Async Tracing

Mike Ruberry committed 4 years ago

7036e91a

Fix FP16 fastAtomicAdd for one case where tensor start address is not 32 bit aligned (#44642)

mcarilli committed 4 years ago

2435d941

Small clarification to amp gradient penalty example (#44667)

mcarilli committed 4 years ago

2fd142a2

Deleted docker images for rocm 3.3 and rocm 3.5 (#44672)

KyleCZH committed 4 years ago

aedce773

add dilation to transposeconv's _output_padding method (#43793)

vchiley committed 4 years ago

c71ce10c

Split CUDA_NVCC_FLAGS by space (#44603)

peterjc123 committed 4 years ago

ed862d36

Newer Older