strisunshinewentingwang/pytorch

Pull Requests Commits

Respect order of Parameters in rnn.py (#18198)

David Riazati committed 5 years ago

f5435634

Refactor EmitLoopCommon to make it more amenable to future extensions (#19341)

Krovatkin committed 5 years ago

2d0d1532

Cleanup init_process_group (#19033)

Kutta Srinivasan committed 5 years ago

b7323a94

Sync FindCUDA/select_computer_arch.cmake from upstream (#19392)

peterjc123 committed 5 years ago

20a5aa96

Update module.py documentation. (#19347)

AlexMetsai committed 5 years ago

9e3bdb32

Add device-specific cuFFT plan caches (#19300)

ssnl committed 5 years ago

973d5107

Improve bmm() performance on CPU when input tensor is non-contiguous (#19338)

mingfeima committed 5 years ago

b8fb6eae

Optional inputs and outputs (#19289)

smessmer committed 5 years ago

12d6f79e

Add some tests (#19288)

smessmer committed 5 years ago

fa96de2b

Use string based schema for exposing caffe2 ops (#19287)

smessmer committed 5 years ago

601f36ba

Allow registering ops without specifying the full schema (#19286)

smessmer committed 5 years ago

5ca22cce

Add either type (#19285)

smessmer committed 5 years ago

a456e1e1

Allow ops without tensor args if only fallback kernel exists (#19284)

smessmer committed 5 years ago

12dcc77b

String-based schemas in op registration API (#19283)

smessmer committed 5 years ago

8036af39

Move function schema parser to ATen/core build target (#19282)

smessmer committed 5 years ago

41dc54e2

Automatic update of fbcode/onnx to ad7313470a9119d7e1afda7edf1d654497ee80ab (#19339)

houseroad committed 5 years ago

789c438d

Remove copy and copy_ special case on Type (#18972)

Roy Li committed 5 years ago

fbf505cb

Add constant folding to ONNX graph during export (Resubmission) (#18698)

Spandan Tiwari committed 5 years ago

a64cce32

Remove usages of TypeID (#19183)

Roy Li committed 5 years ago

01d7d3de

Fixing function schema parser for Android (#19281)

smessmer committed 5 years ago

c7b1fdb7

Split function schema parser from operator (#19280)

smessmer committed 5 years ago

094678c0

fix hub doc format

Ailing Zhang committed 5 years ago

a1174dbc

Clang-format torch/csrc/jit/passes/quantization.cpp. (#19385)

Mikhail Zolotukhin committed 5 years ago

b31bab78

Allow DDP to wrap multi-GPU modules (#19271)

mrshenli committed 5 years ago

6732358b

Add validator for optimizers when parameters are shared

chocjy committed 5 years ago

c48e1679

hub minor fixes (#19247)

Ailing Zhang committed 5 years ago

2787f1d8

fix wrong schema (#19370)

Elias Ellison committed 5 years ago

776fec0f

Fix printing format in examples in jit/README.md. (#19323)

Mikhail Zolotukhin committed 5 years ago

bf5f30f3

Allow for single-line deletions in clang_tidy.py (#19082)

efaust committed 5 years ago

48859e3a

Revert D14901379: [jit] Add options to Operator to enable registration of alias analysis passes

suo committed 5 years ago

242743ee

Newer Older