COMMITS
/ tests/python/gpu/test_operator_gpu.py April 9, 2018
C
CUDNN not training on backward pass similar to pytorch (#10470)
Chris Olivier committed
C
[MXNET-80] Fix average pooling kernel size assignment error (#10000)
CoinCheung committed
April 3, 2018
A
[MXNET-81] Fix crash with mx.nd.ones (#10014)
Anirudh Subramanian committed
March 31, 2018
C
fix crash when profiler not enabled (#10306)
Chris Olivier committed
March 22, 2018
March 20, 2018
March 10, 2018
J
Cpu lstm inference (#9977)
Jerry Zhang committed
February 18, 2018
D
Add seed management to python tests (#9791)
Dick Carter committed
February 15, 2018
D
Refactor operators and add MKLDNN (#9677)
Da Zheng committed
February 13, 2018
A
Better Exception Handling for Operators (#9681)
Anirudh Subramanian committed
February 9, 2018
T
Check padding size for global pooling (#9730)
Tao Lv committed
February 1, 2018
C
Revert "Refactor operators & MKLDNN (#8302)" (#9661)
Chris Olivier committed
January 31, 2018
D
Refactor operators & MKLDNN (#8302)
Da Zheng committed
January 30, 2018
D
Expand gpu-kernel-launch synchronous error checking. (#9560)
Dick Carter committed
January 3, 2018
Z
sparse output for binary scalar op with zero (#9227)
Ziyue Huang committed
January 2, 2018
D
1 d conv with cudnn (#9184)
Dick Carter committed
December 13, 2017
M
Disable test_operator_gpu.test_svmoutput_with_type (#9059)
Marco de Abreu committed
December 5, 2017
Z
rsp[:] = const_value, used in constant initializer for rsp weight (#8891)
Ziyue Huang committed
December 1, 2017
Z
fix race when temp space is used in copy & fix instance overwrite in g2c (#8867)
Ziyue Huang committed
November 26, 2017
S
Update test_operator_gpu.py (#8819)
Sheng Zha committed
November 25, 2017
S
multi-device support (#8812)
Sheng Zha committed
November 15, 2017
Z
check_format of sparse ndrray (#8259)
Ziyue Huang committed
October 25, 2017
H
dtype default to source_array.dtype for sparse ndarrays (#8403)
Haibin Lin committed
October 15, 2017
I
Temporarily disable 'test_row_sparse_pull' on GPU. (#8265)
Indhu Bharathi committed
October 13, 2017
I
Temporarily disable some unit tests to fix the build (#8253)
Indhu Bharathi committed
October 6, 2017
A
Add infer storage type function for sparse slice operator (#8154)
Anirudh Subramanian committed
October 2, 2017
E
Autograd with multiple devices (#8124)
Eric Junyuan Xie committed
September 26, 2017
E
Add cuda rtc module (#8017)
Eric Junyuan Xie committed
September 11, 2017
S
change contrib CTC to zero-indexed, in log space. sequence mask for grad (#7727)
Sheng Zha committed
August 31, 2017
H
Add more gpu tests and fix nested try catch in Imperative Invoke (#7676)
Haibin Lin committed
August 30, 2017
C
1x1 convolution acceleration (#7613)
chinakook committed
S
fix ctc on softmax grad and req option (#7660)
Sheng Zha committed
August 29, 2017
R
Refactor random linalg contrib namespaces (#7604)
reminisce committed
R
skip failing test temporarily (#7648)
Rahul Huilgol committed
August 27, 2017
S
fix tests (#7633)
Sheng Zha committed
August 24, 2017
S
contrib ctc interface changes, cudnn7 CTC, and gluon CTC (#7442)
Sheng Zha committed
August 22, 2017
H
Sparse Tensor: request for reviews (#7082)
Haibin Lin committed
August 18, 2017
D
Changed FullyConnected to use new linalg gemm, plus TensorCore if fp16 I/O. (#7505)
Dick Carter committed
August 15, 2017
E
fix autograd memory cost (#7478)
Eric Junyuan Xie committed
August 12, 2017
W
fix consistency of cpu/gpu in stn (#7374)
Wei Wu committed
August 8, 2017
M
Add license header (#7379)
Mu Li committed
August 7, 2017
S
fix build (#7358)
Sheng Zha committed
July 28, 2017
L
July 17, 2017
M
Optimized sequence reverse operator (#6946)
Marek Kolodziej committed
July 10, 2017
E
rename Foo to Gluon (#6980)
Eric Junyuan Xie committed
E
rename nn.Layer to foo.Layer (#6959)
Eric Junyuan Xie committed
June 27, 2017
E
refactor rnn layers (#6823)
Eric Junyuan Xie committed
June 21, 2017
E
[RFC]HybridLayer (#6772)
Eric Junyuan Xie committed
June 5, 2017
E
reorg code to temporary namespace foo (#6568)
Eric Junyuan Xie committed
May 31, 2017
E
nn interface (#6501)
Eric Junyuan Xie committed