strisunshinewentingwang/pytorch

Pull Requests Commits

[PyTorch] Avoid registering ops into dispatcher in lightweight dispatch (#74664)

larryliu0820 committed 2 years ago

4d82e5bf

Fix formatting of scalar tensors (don't call item)

ezyang committed 2 years ago

51e7a340

Implement _pad_circular in ATen

peterbell10 committed 2 years ago

f86bb2d6

Jinja2 for docs/cpp build set to version 3.0

b0noI committed 2 years ago

f7317d3c

[4/5]Testing jit module in flatbuffer in Python. (#74387)

qihqi committed 2 years ago

75d6cbe6

Add Python Version to Torch.Package metadata (#74610)

Jamie McCrindle committed 2 years ago

11894db9

Jinja2 version pinned to 3.0.* (#74690)

b0noI committed 2 years ago

7f996b85

Add support for backend to register reducer timer

jeejakp12 committed 2 years ago

13ebcf37

[Model Averaging] Make HierarchicalModelAverager a subclass of averagers.ModelAverager

wayi1 committed 2 years ago

5fbe8b19

Back out "Revert D34805092: Extend _save_for_mobile and _load_for_mobile to support flatbuffer format; Default format is pickle + Change buck targets to support `only pickle` and `pickle + flatbuffer` for migration" (#74594)

pavithranrao committed 2 years ago

fc2cf3d2

[quant] Remove assert for weight since it could be non-Tensor (#74365)

jerryzh168 committed 2 years ago

d64e7634

[pyper] to + lengths_to_offsets (#73879)

ajyu committed 2 years ago

f2ca4341

c10d: retry dns lookup failures (#74641)

d4l3k committed 2 years ago

5b915e84

Automated submodule update: FBGEMM (#74633)

facebook-github-bot committed 2 years ago

d0adb5ff

[Profiler] Pay for what you use (v2) (#74484)

Taylor Robie committed 2 years ago

2ecf7437

[reland] Process inputs and outputs in fx interpreter (#74637)

842974287 committed 2 years ago

3f164e03

[quant] Populate FakeQuantize quant_min/quant_max to observer (#74581)

jiaxuzhu92 committed 2 years ago

7c1f3cc8

Pull request to run CI for #72556 (#73404)

kstant0725 committed 2 years ago

ff58899b

Add meta support for [un]squeeze(), fix bug with set_()

bdhirsh committed 2 years ago

23383b1e

Adding gpuAtomicMin and gpuAtomicMax for non-integer types

mikaylagawarecki committed 2 years ago

4025ca8c

register meta allocator

bdhirsh committed 2 years ago

6dd6be63

Decorate test_pdist_large for requiring large memory (#74574)

janeyx99 committed 2 years ago

3f9115dc

Add cuda_atomic_ops_test to run_tests.sh

mikaylagawarecki committed 2 years ago

3024bcff

[Dynamic RPC] Allow newly joined ranks to communicate with existing ranks (#73373)

H-Huang committed 2 years ago

9270bcca

[Dynamic RPC] Allow for optional world_size argument in init_rpc (#73372)

H-Huang committed 2 years ago

f76d1c02

[quant] Add default symmetric qat qconfig for qnnpack (#74507)

digantdesai committed 2 years ago

09f32eba

Skip specifying rcond for gelsy driver in tests

IvanYashchuk committed 2 years ago

ca4ba2ee

Add forward AD support for clamp when bounds are tensors

soulitzer committed 2 years ago

a4c81b13

Remove high priority as an owner for tests (#74555)

janeyx99 committed 2 years ago

a1e284d9

[shard] Add ReplicatedTensor (#73529)

wanchaol committed 2 years ago

0524b282

Newer Older