strisunshinewentingwang/pytorch

Pull Requests Commits

Fix torch.arange traced as constant (#25363)

lara-hdr committed 5 years ago

8ca93ec3

add the tensor_observer to record the runtime tensor for quantization … (#25830)

lly-zero-one committed 5 years ago

62767077

Improve error message when input is not in the right format (#25928)

dskhudia committed 5 years ago

ec48280a

enable unit tests (#25963)

iotamudelta committed 5 years ago

00d967c3

remove pthreadpool.a from install directory (#25977)

ljk53 committed 5 years ago

075adb4d

Updating submodules

svcscm committed 5 years ago

54f3cb8f

Add torch.backends.mkldnn.enabled flag (#25459)

jiayisunx committed 5 years ago

b9bf91fe

Simply code generation - phase 1 (#25961)

kostmo committed 5 years ago

c79a13b7

indentation for hypothesis profile and proper inheritance for QuantizationTestCase (#25934)

jerryzh168 committed 5 years ago

76487e16

Change order of activation and weight in QConfig (#25950)

jerryzh168 committed 5 years ago

c475ef72

Fix typo in OpenBLAS cmake detection

albanD committed 5 years ago

63df9ffd

Add VariableTensorId, store it in TensorTypeSet (#25597)

ezyang committed 5 years ago

2080a158

C++ MaxPool Module

ShahriarSS committed 5 years ago

ba9fda14

L1Loss module (#25902)

ShahriarSS committed 5 years ago

e0483600

The float version of calc_digamma should return float type. (#25488)

xuhdev committed 5 years ago

1a58a9e4

Remove global group name tracking for ProcessGroupNCCL (#25905)

pietern committed 5 years ago

ebdb32c7

Remove superfluous check for POLLIN in TCPStore (#25911)

pietern committed 5 years ago

929764ac

Make running Gloo tests conditional on availability

pietern committed 5 years ago

e4cd807c

Increase failure threshold for timing based assert (#25867)

pietern committed 5 years ago

ebeb2a35

Updates autograd engine to respect streams set in forward (#8354)

mruberry committed 5 years ago

87a2c926

Updating submodules

svcscm committed 5 years ago

6b3f9689

Fix LBFGS on GPU (#25909)

vishwakftw committed 5 years ago

9815739d

Add assert to ensure the divisor is not 0 (#25960)

jianyuh committed 5 years ago

3185b455

Add torch.nn.LSTM into the default dynamic quantize mappings (#25954)

jianyuh committed 5 years ago

9b4f3fd7

fix use-after-free bug

zdevito committed 5 years ago

21e9d114

Delete tools/autograd/env.py (#25920)

zou3519 committed 5 years ago

dc015a1a

compute common dtype based on inputs only (#25593)

ifedan committed 5 years ago

e69a6bab

add support for ModuleDict (#25715)

Elias Ellison committed 5 years ago

8f7020bb

Simplify header inclusion in test/cpp/api/modules.cpp (#25921)

Will Feng committed 5 years ago

a88f3101

remove protobuf from Dependencies.cmake for libtorch mobile build (#25958)

ljk53 committed 5 years ago

74b48f21

Newer Older