strisunshinewentingwang/pytorch

Pull Requests Commits

add tests for add_custom_scalars and others (#20987)

lanpa committed 5 years ago

1fc240e5

Expose torch.empty(sizes, *, names, ...) to Python (#21648)

zou3519 committed 5 years ago

0d6eb209

Fix flaky nuclear_norm() test (#21638)

skrah committed 5 years ago

71082187

Adding the quantized namespace to torch.nn and importing it from torch (#21600)

zaf committed 5 years ago

ff8c3fd5

Deprecate unordered_map and vector in IValues (#21712)

smessmer committed 5 years ago

9a1dc43f

Define __setstate__ on _ConvNd to handle pre-padding_mode pickles. (#21687)

ezyang committed 5 years ago

029a9682

Fix handling of kwargs from common method invocations (#21499)

nairbv committed 5 years ago

7284f448

Add ONNX py3 CI cases (#21715)

houseroad committed 5 years ago

c1744a6c

Add aten mkldnn zero_ operator (#20573)

XiaobingSuper committed 5 years ago

c06ccbe6

rebuild_storage_fd retry on EINTR (#21723)

ssnl committed 5 years ago

bc628102

Throwing errors for min and max reductions in empty CUDA tensors (#19612)

LeviViana committed 5 years ago

deb2140c

When building extensions, honor options set in CMake. (#21653)

xuhdev committed 5 years ago

b811b6d5

When converting to NumPy, throw TypeError when type is not supported (#21608)

xuhdev committed 5 years ago

4001e715

Fix with emit_nvtx, also allow shape information to appear in nvtx ranges. (#21691)

definitelynotmcarilli committed 5 years ago

2d5ce519

Fix the issue of sizes vs size for tensor creation ops (#21686)

shagunsodhani committed 5 years ago

b9675efb

Query caffe2 operator stats for detailed execution info (#20924)

Benny Chen committed 5 years ago

1e7bd758

backend.py: _getattr__ must raise AttributeError (#21763)

Aapo Kyrola committed 5 years ago

d9eec4ef

Handling numel() == 0 in convTranspose (#21652)

anshuljain1 committed 5 years ago

044809f1

Implement at::empty(IntArrayRef, DimnameList?, TensorOptions) in aten (#21647)

zou3519 committed 5 years ago

5c0e0589

Make it possible to trigger all tests by pushing to ci-all/ branch (#21750)

ezyang committed 5 years ago

3e790363

better example for local weights (#21685)

Ailing Zhang committed 5 years ago

16b4a12e

Add batch id to tracer event (#21446)

Sherman Wong committed 5 years ago

adc99efb

schema_matching.cpp: improve error messages.

Mikhail Zolotukhin committed 5 years ago

fbecb462

Tune elementwise ops for ROCm (#21754)

colesbury committed 5 years ago

cfd8c58b

Fix spelling errors (#21665)

chosungmann committed 5 years ago

f5958121

fix multihead attention for half (#21658)

Natalia Gimelshein committed 5 years ago

efd20de2

Use expect to fill in pytorchbot token (#20459)

Will Feng committed 5 years ago

4716409f

Document that no_grad is thread local. (#21755)

ezyang committed 5 years ago

b858f42e

Bug fix: ONNX export full operator (#21669)

BowenBao committed 5 years ago

3e8dc565

Added dim check for index_copy_ (#21617)

izdeby committed 5 years ago

4b45f08f

Newer Older