strisunshinewentingwang/pytorch

Pull Requests Commits

Add mark_static (#98427)

ezyang committed 1 year ago

f98c1809

Turn off divisible_by_16 for dynamic shapes; support ablation (#98471)

ezyang committed 1 year ago

bdb79a8f

[quant][pt2e] Support quantizer API in prepare_pt2e_quantizer (#97994)

jerryzh168 committed 1 year ago

3142ce20

[Inductor] Fix convolution lowering if stride or padding or dilation is 1 element list (#98448)

yanboliang committed 1 year ago

ccc27bc3

Print collective (#97544)

rohan-varma committed 1 year ago

b8cf0101

[PG Wrapper] Add sequence number (#97462)

rohan-varma committed 1 year ago

dab1a7e6

[FSDP] records for composable (#98428)

rohan-varma committed 1 year ago

428c531d

[ROCm] Enable FSDP BF16 comm hooks unit tests (#97517)

lcskrishna committed 1 year ago

eadd84d0

Make caling type on user defined class UserError (#98366)

tugsbayasgalan committed 1 year ago

37dc47a1

[SPMD] Dedup collectives from DTensor expansion (#98216)

mrshenli committed 1 year ago

1cd1d9c2

Enable LogSoftmax for SPMD tracing (#98380)

mrshenli committed 1 year ago

11b0a84f

backport std::bit_cast from c++20 to c10 (#98417)

mikey dagitses committed 1 year ago

e2c81e44

Support neg calls to dyn shapes (#94068)

voznesenskym committed 1 year ago

ab95b7a0

fix grain size setting for baddbmm_cpu_kernel (#98297)

mingfeima committed 1 year ago

11890156

Add PrivateUse1 for dispatching PyTorch Distributed Collectives. (#98137)

shaoyf42 committed 1 year ago

cc5f6495

Clean up CUDA 11.6 Docker images (#98395)

huydhn committed 1 year ago

d3adbbf4

[BE] Fix `collect_env` for python-path-with-space (#98415)

malfet committed 1 year ago

bd785320

[EASY] Fix some more places where we incorrectly assume only Tensor (#98310)

ezyang committed 1 year ago

680bf14a

Disable persistent reductions with dynamic shapes (#98405)

ezyang committed 1 year ago

478df47f

[CI] Update update_expected.py to make it generate a combined csv file (#98407)

desertfire committed 1 year ago

007587aa

[quant][pt2e][fix] Fix the internal test failures caused by refactor (#98378)

jerryzh168 committed 1 year ago

a7611483

Add check for 0 to 1 inclusive for elements of target tensor in BCE loss (#97814)

kiersten-stokes committed 1 year ago

2a48f43f

Export function symbols to enable Windows build of Intel Extension for PyTorch (#98054)

z71258847 committed 1 year ago

3112d2a2

[inductor] Move `tl.broadcast` call out codegen.common (#98304)

peterbell10 committed 1 year ago

013c7f5b

[inductor] Enable CSE on masked loads (#98303)

peterbell10 committed 1 year ago

bb4174d2

rewrite at::vec::*::convert_to_int_of_same_size (#98429)

mikey dagitses committed 1 year ago

aa7850c2

Forward fix for DataLoader to accept custom Sharding DataPipe (#97287)

ejguan committed 1 year ago

29d2e4b7

Add detect_fake_mode (#98321)

ezyang committed 1 year ago

d01ee10b

Extract ExtraMeta symbolic shape fields into a dedicated SymbolicShap… (#98399)

ppiskorski committed 1 year ago

5854923c

Add slow workflow to upload test stats workflow (#98447)

clee2000 committed 1 year ago

5af47dbb

Newer Older