strisunshinewentingwang/pytorch

Pull Requests Commits

log message at per-test level for`perfpipe_pytorch_test_times` (#43752)

Danqi Huang committed 4 years ago

71373276

Move torch/autograd/grad_mode.pyi stubs inline (#43415)

rgommers committed 4 years ago

4c19a1e3

Enable gcc coverage in OSS (#43883)

yujunzhao@devvm1621.atn0.facebook.com committed 4 years ago

e941a462

Move `fbcode` related coverage code to `fb/` folder and add `TARGETS` (#43800)

yujunzhao@devvm1621.atn0.facebook.com committed 4 years ago

da0e93a8

Implementing NumPy-like function torch.heaviside() (#42523)

Kiyosora committed 4 years ago

3682df77

Let linspace support bfloat16 and complex dtypes (#43578)

xuhdev committed 4 years ago

7680d87a

Skip target determination for codecov test (#43899)

malfet committed 4 years ago

3278beff

[pytorch][bot] update mobile op deps (#43871)

ljk53 committed 4 years ago

ffca81e3

Join-based API to support DDP uneven inputs (#42577)

rohan-varma committed 4 years ago

4e4626a2

Publish all_gather_object and gather_object docs (#43772)

mrshenli committed 4 years ago

2f527485

Revert D23385091: [quant][graphmode][fx] Add top level APIs

albanD committed 4 years ago

f7bae5b6

Revert D23385090: [quant][graphmode][fx] Add support for weight prepack folding

albanD committed 4 years ago

68304c52

[fix] torch.multinomial : fix for 0 size dim (#43775)

kshitij12345 committed 4 years ago

0394c5a2

Update aliasing in tensorexpr fuser (#43743)

Elias Ellison committed 4 years ago

3c8b1d73

use types in the IR instead of vmap (#43742)

Elias Ellison committed 4 years ago

5da8a7bf

Add passes to profiling executor pipeline (#43636)

Elias Ellison committed 4 years ago

259e5b7d

Use prim::TensorExprGroup interned symbol (#43635)

Elias Ellison committed 4 years ago

a7e7981c

Update requires grad property (#43634)

Elias Ellison committed 4 years ago

1c0faa75

add qr_backward functionality for wide case (#42216)

rlucas7 committed 4 years ago

2bede78a

[RPC profiling] Add test to ensure using record_function works for RPC (#43657)

rohan-varma committed 4 years ago

69dd0bab

Add __complex__ (#43844)

zasdfgbnm committed 4 years ago

4ef12be9

addmm/addmv should accept complex alpha and beta (#43827)

zasdfgbnm committed 4 years ago

c5d0f091

[fx] GraphModule.src -> GraphModule.code (#43655)

suo committed 4 years ago

89452a67

[NNC] Hook up registerizer to Cuda codegen [2/x] (#42878)

nickgg committed 4 years ago

1390cad2

Better msg (#43848)

Yinghai Lu committed 4 years ago

63dbef30

[quant][graphmode][fx] Add support for weight prepack folding (#43728)

jerryzh168 committed 4 years ago

ef08f920

[quant][graphmode][fx] Add top level APIs (#43581)

jerryzh168 committed 4 years ago

eb4199b0

Properly check that reduction strings are valid for l1_loss, smoothl1_loss, and mse_loss. (#43527)

gchanan committed 4 years ago

42c895de

[quant][graphmode][fx][fix] enable per channel quantization for functional ops (#43534)

jerryzh168 committed 4 years ago

b8d34547

Rewrite of ATen code generator (#42629)

ezyang committed 4 years ago

6ea89166

Newer Older