strisunshinewentingwang/pytorch

Pull Requests Commits

Dynamic shape support for functionalization of RNG ops

anijain2305 committed 1 year ago

4914d6db

[cuda rng] Making offset calculation independent of device properties

anijain2305 committed 1 year ago

987b6eaa

Python binding to set CUDA rng state offset

anijain2305 committed 1 year ago

3409d003

[pt2] add meta function for `linalg.cross` (#99279)

nkaretnikov committed 1 year ago

efc3887e

[pt2] add `SymInt` support for `column_stack` (#99276)

nkaretnikov committed 1 year ago

775dd869

Rename default branch to `main` (#99210)

malfet committed 1 year ago

Verified 2418b945

[PT2E][Quantization] Refactor Quantizer and QNNPACKQuantizer (#99063)

kimishpatel committed 1 year ago

31f311a8

fix fake tensor propagation for cross_entropy with smoothing (#99255)

ngimel committed 1 year ago

888c65b6

simplify codegen for integer min/max since they don't need to propaga… (#99249)

ngimel committed 1 year ago

fa502ab0

Remove obsolete HAS_PRIMS_REFS (#99252)

jansel committed 1 year ago

9ab5fdff

Revert "[quant][fix] Compare resnet with quantizer api with the prepare_fx and decomposed convert flow (#98905)"

pytorchmergebot committed 1 year ago

20a17881

make untemplated gemm calls data_ptr-correct (#99184)

mikey dagitses committed 1 year ago

be0b12ec

make ATen/native/cuda/Embedding.cu data_ptr-correct (#99183)

mikey dagitses committed 1 year ago

29ff5a0c

Add parameter for pin memory of storage to support other devices. (#98692)

wbigat committed 1 year ago

b08c3841

Revert "Expandable blocks in allocator (#96995)"

pytorchmergebot committed 1 year ago

851e89c8

[vmap] Register decomposition for huber_loss_backward (#99236)

qqaatw committed 1 year ago

6f181aae

Further improve symbolic shapes logging (#99159)

ezyang committed 1 year ago

e2923b52

Functionalization of torch.rand/rand_like ops (#97377)

anijain2305 committed 1 year ago

fdbc8625

[inductor] Refactor pre-grad passes into inductor.fx_passes (#99130)

jansel committed 1 year ago

6e1e27fc

[inductor][quant]Enable inductor vec codegen for quantization (#98489)

leslie-fang-intel committed 1 year ago

91279f94

Add invariant that all symbolic shapes must be bound in graph (#99089)

ezyang committed 1 year ago

039faf0d

[SPMD][BE] Generalize factory ops support in SPMD expansion (#99233)

mrshenli committed 1 year ago

c69d5488

[SPMD][BE] Remove deprecated aten.sym_numel branch (#99232)

mrshenli committed 1 year ago

6bb20822

[SPMD][BE] Consolidate DSymInt Branches (#99231)

mrshenli committed 1 year ago

39be9949

[SPMD] Refactor DSize to DSymInt to enable sym_numel (#99206)

mrshenli committed 1 year ago

544cd8e1

[SPMD] Enable aten.full.default with SymInt on sharded dims (#99190)

mrshenli committed 1 year ago

bafb9840

SymIntify randint and randperm (#98968)

ezyang committed 1 year ago

d350646f

Support large negative SymInt (#99157)

ezyang committed 1 year ago

756a86d5

[vmap] Add vmap support for nn.functional.huber_loss (#99235)

qqaatw committed 1 year ago

5c062e8b

fix breakage from #99027 (#99245)

mikey dagitses committed 1 year ago

c9403f12

Older