strisunshinewentingwang/pytorch

Pull Requests Commits

Add missing algorithm header to Array utility (#22157)

Uladzislau Paulovich committed 5 years ago

a7cb07eb

Remove THD (#22065)

pietern committed 5 years ago

6ff0c6ca

Port symeig to ATen and enable batching of inputs (#21858)

vishwakftw committed 5 years ago

bcb5fd8f

Show deprecation warning when stateful lambdas are used as kernels (#21885)

smessmer committed 5 years ago

4ec6fbef

serialize torch.Size object (#20952)

ailzhang committed 5 years ago

c6811938

porting convtranspose3d to ATen (#22019)

xmnlab committed 5 years ago

7daa96a3

Not expose mkldnn reshape and transpose (#22193)

XiaobingSuper committed 5 years ago

9af8ea1c

Switch autograd to use a pool of workers for each device (#21911)

mal committed 5 years ago

c8b5f1d2

Optimization of the Embedding and Embedding-Bag CUDA Kernel (#22016)

madsbk committed 5 years ago

94e83da5

Further remove redundant CMake option passing code for those CMake variables that are directly controlled by environment variables but with a different name. (#22154)

xuhdev committed 5 years ago

b0bd8758

Remove more build options not needed to be explicitly set in Python build scripts. (#22153)

xuhdev committed 5 years ago

ce1a9653

Fixes bugs in torch.multinomial without replacement (#22183)

syed-ahmed committed 5 years ago

839b496f

Optimize InstanceNormOp forward (#22130)

xiaomengy committed 5 years ago

b61693c0

support both regularizable and sofmax re-weighting on sparse features in dot product (#22176)

Le Fang committed 5 years ago

ac4913ee

Use latest stable flake8-bugbear in CI and fix B011 flake8 error. (#21944)

xuhdev committed 5 years ago

299ea84a

Don't end on inplace operators in einsum (#22111)

t-vi committed 5 years ago

f5df0c91

Enabled mul for bool tensors on CUDA (#21771)

izdeby committed 5 years ago

ede08492

Add missing whitespace in error message (#21904)

cdancette committed 5 years ago

3b700a43

AveragePool: expand incomplete kernel_size for the C++ API

skrah committed 5 years ago

88cdc168

DilatedMaxPool: expand incomplete kernel_size for the C++ API (#22073)

skrah committed 5 years ago

2372e7ed

Make Dropout.__repr__ consistent with other modules (#22110)

apaszke committed 5 years ago

b2a39314

Cast return value of vector.at() to void to avoid nodiscard warning in MSVC. (#22061)

xuhdev committed 5 years ago

273b6c5b

Quick fix for #18215, the CPU case (#21910)

zh217 committed 5 years ago

0ac28c89

Improve repr for IncompatibleKeys (#22119)

apaszke committed 5 years ago

41d0525d

Fix minor issues with #21736 (#22074)

apaszke committed 5 years ago

f1775796

Document the Boolean tensor type.

xuhdev committed 5 years ago

a4589893

Fix in ivalue::Future (#22114)

Ilia Cherniavskii committed 5 years ago

7c420649

Fix sequential MKL case (#22062)

Ilia Cherniavskii committed 5 years ago

6350dbdd

Better trace comments

Krovatkin committed 5 years ago

21da33f0

De-deprecate some warnings that hurt usability (#21999)

bwasti committed 5 years ago

f1c7fa05

Newer Older