strisunshinewentingwang/pytorch

Pull Requests Commits

Enable dist autograd tests (#28606)

mrshenli committed 4 years ago

261a13a8

Compute correct strides after type promotion (#28253)

nairbv committed 4 years ago

70e4548f

C++ parity, grid_sample functional (#28354)

lsrock1 committed 4 years ago

e885ce61

C++ nn::ConstantPad{1,2,3}d

Will Feng committed 4 years ago

92b39434

Speed up fill for half and bfloat16 on CPU. (#28397)

xuhdev committed 4 years ago

5cf64415

C++ nn::ZeroPad2d

Will Feng committed 4 years ago

7f9941c4

Enable Interpolate Tests for ONNX Opset 11 (#28560)

lara-hdr committed 4 years ago

d762ad09

Skip ProcessGroupNCCLTest if CUDA is not available (#28393)

rohan-varma committed 4 years ago

a7835637

C++ API parity: at::Tensor::requires_grad_

pbelevich committed 4 years ago

46f96d15

Minor followup on stringstream cleanups (#28300)

jjlilley committed 4 years ago

78039627

C++ nn::ReplicationPad{1,2,3}d

Will Feng committed 4 years ago

303527d7

C++ nn::ReflectionPad1d and nn::ReflectionPad2d

Will Feng committed 4 years ago

78375c02

jjsjann123 committed 4 years ago

e263dd38

Fix compute_non_overlapping_and_dense() (#28551)

ifedan committed 4 years ago

2020cc0c

Migrate remaining ops to the c10 dispatcher (#27978)

smessmer committed 4 years ago

8de8cab2

autograd/profiler: make python record_function use JIT methods

d4l3k committed 4 years ago

d8c66c15

CPU-Strided-Complex Support for reduce ops and linpack ops (#27653)

dylanbespalko committed 4 years ago

f8b758b1

torch.histc added a finite range check to resolve segfaults if tensor has inf. also added checks for nan values, min>max (#27712)

anjali411 committed 4 years ago

136bb07a

Kill TH(C)Tensor_squeeze which isn't used anymore.

gchanan committed 4 years ago

ae05e48f

Port is_set_to from TH/THC to ATen.

gchanan committed 4 years ago

4f0a3504

remove type information from docstrings of quantization functions (#28556)

ngoldbaum committed 4 years ago

139fec2d

C++ API parity: Linear

pbelevich committed 4 years ago

dd277e90

Make init_method url appending step re-usable by both init_process_group and init_model_parallel(init_rpc) (#28226)

xush6528 committed 4 years ago

59402f51

Make RRef::LocalValue return Future (#28025)

mrshenli committed 4 years ago

e31adeb4

Make RRef::toHere() return a jit::Future (#27943)

mrshenli committed 4 years ago

58873776

Updating submodules

svcscm committed 4 years ago

9c345473

static initialization order with mutex (#28243)

EscapeZero committed 4 years ago

61d40b80

workaround for raw string bug in VS2019 (#28349)

Mingbo Wan committed 4 years ago

80083223

Scripts for setting up benchmark projects (#28469)

xta0 committed 4 years ago

896b5d91

Ignore F401 in all __init__.py without putting noqa (#25823)

narumiruna committed 4 years ago

d83389d3

Newer Older