strisunshinewentingwang/pytorch

Pull Requests Commits

Dict mutability (#16884)

David Riazati committed 5 years ago

ac00a0cd

Fix static linkage cases and NO_DISTRIBUTED=1 + CUDA (#16705) (#17337)

soumith committed 5 years ago

3a47d569

Fix Insert Constant Lint Fail (#17316)

Elias Ellison committed 5 years ago

290b2a1d

Partial support for kwarg_only arguments in script (#17339)

zdevito committed 5 years ago

4c6da649

fix double backward for half softmax/logsoftmax (#17330)

Natalia Gimelshein committed 5 years ago

5fa78303

Revisit some native functions to increase number of jit matches (#17340)

cpuhrsch committed 5 years ago

9101dfc5

Add Value::isValidName method. (#17372)

Mikhail Zolotukhin committed 5 years ago

46f15b74

Fix #17218 by updating documentation (#17258)

Bharat123Rox committed 5 years ago

6feded88

fix lint (#17366)

soumith committed 5 years ago

45251fb5

switch to Operation in register_prim_ops.cpp (#17183)

Krovatkin committed 5 years ago

3145f46a

Use standard docker image for XLA build

kostmo committed 5 years ago

b312e9de

Modernize test_sparse. (#17324)

gchanan committed 5 years ago

25730f15

remove nn.Upsample deprecation warnings from tests (#17352)

soumith committed 5 years ago

c63af883

upgrade documentation in setup.py to NO_ -> USE_ (#17333)

soumith committed 5 years ago

3069c450

Enforce non-negativity of tensor construction (#17077)

Dmytro Dzhulgakov committed 5 years ago

5744d521

Fixing docstring in CTCLoss (#17307)

Igor Macedo Quintanilha committed 5 years ago

94a95a0c

Fix the slowness of mvn's log_prob (#17294)

fehiepsi committed 5 years ago

de81a274

Move argsort to C++

zasdfgbnm committed 5 years ago

722cbe30

Include vec256 headers in setup.py (#17220)

tridao committed 5 years ago

37890610

Enable MAX_JOBS for using Ninja on Windows

peterjc123 committed 5 years ago

51069186

Avoid unnecessary CPU-to-GPU copy of torch.load with CUDA (#17297)

lw committed 5 years ago

29f4f8f0

allow lists to contain any tensor type (#17321)

suo committed 5 years ago

2c302b6e

Skip convnets benchmark in rocm CI (#17331)

bddppq committed 5 years ago

d92ddcf7

Don't have malloc-free pairs that cross DLL boundaries. (#17302)

ezyang committed 5 years ago

b3b692a8

Add support to build for multiple amd gpu targets (#17329)

bddppq committed 5 years ago

c063a33e

batched cleanups (#17288)

suo committed 5 years ago

501d346d

(Permanently) fix CI breakage due to new docker version. (#17338)

ezyang committed 5 years ago

cb3d3d11

Implementation convolutionTranspose operator for mkl-dnn (#12866)

PenghuiCheng committed 5 years ago

376bb403

Support multi-device configuration for MKL-DNN (#12856)

PenghuiCheng committed 5 years ago

c02e2ff0

fix missing std (#17263)

Ailing Zhang committed 5 years ago

90950f79

Newer Older