Welcome to mirror list, hosted at ThFree Co, Russian Federation.

github.com/torch/cunn.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2017-04-05Update to ignore zero targetsJin-Hwa Kim
If the target is zero, loss and gradient of input are set to zero. It is useful for variable-length natural language generation models.
2017-03-31Merge pull request #456 from twitter-forks/addmm-fixesSoumith Chintala
Using temporary variables when performing transpose + addmm
2017-03-30Using temporary variables when performing transpose + addmmPavan Yalamanchili
2017-03-25Improving the performance of IndexLinear:updateOutputPavan Yalamanchili
- Removes separate kernel for updateOutputTrain
2017-03-24Merge pull request #454 from ngimel/patch-1Soumith Chintala
Fix inconsistent in-place and out-of-place for HardTanh
2017-03-24Fix inconsistent in-place and out-of-place for HardTanhngimel
in-place and out-of-place updateGradOutput results are different where input=min_val or input=max_val
2017-03-24Adding support for flattened inputs for IndexLinearPavan Yalamanchili
- Adding relevant tests
2017-03-24IndexLinear support for cunnPavan Yalamanchili
2017-03-22Merge pull request #453 from apaszke/lookup_renormSoumith Chintala
Cast accumulator in LookupTable renorm to accreal
2017-03-22Added support for multidimensional tensors in PReLU; Channel number now in ↵Hardik
second dimension
2017-03-22Cast accumulator in LookupTable renorm to accrealAdam Paszke
2017-03-15Merge pull request #449 from gchanan/precision_testingSoumith Chintala
Improve precision testing
2017-03-15Merge pull request #451 from wickedfoo/faster-lookup-tableSoumith Chintala
Improve cunn LookupTable performance for large batch sizes
2017-03-13change lookup table sortJeff Johnson
2017-03-06Improve precision testingGregory Chanan
1) In cases where tests were failing with some regularity (SmoothL1, l1cost, mse, SoftShrink-backward) scale error bounds by (absolute) value being tested. 2) Fix some spacing issues in error messages.
2017-02-22Merge pull request #447 from gchanan/accrealSoumith Chintala
Remove accreal typedef because it is defined multiple times
2017-02-22Remove accreal typedef because it is defined multiple timesGregory Chanan
with nn.
2017-02-21Merge pull request #446 from lvdmaaten/masterSoumith Chintala
[DPT] Add option for synchronous copy of model to GPUs
2017-02-21[DPT] Add option for synchronous copy of model to GPUsLaurens van der Maaten
2017-02-21Merge pull request #418 from ruotianluo/adaptiveAverageSoumith Chintala
Add SpatialAdaptiveAveragePooling.
2017-02-21Merge pull request #434 from bottler/masterSoumith Chintala
VolumetricFractionalMaxPooling like spatial
2017-02-21Merge pull request #442 from twitter-forks/half-fixesSoumith Chintala
Convert real to accreal in libTHCUNN
2017-02-20Merge pull request #445 from apaszke/nll_criterionSoumith Chintala
Do SpatialClassNLLCriterion sizeAverage in a separate kernel
2017-02-20Do SpatialClassNLLCriterion sizeAverage in a separate kernelAdam Paszke
2017-02-18Merge pull request #444 from apaszke/fixesSoumith Chintala
Improvements for spatial functions
2017-02-18Use a more stable formula for spatial LogSoftMaxAdam Paszke
2017-02-18Add shape checks for SpatialClassNLLCriterionAdam Paszke
2017-02-17Convert real to accreal in libTHCUNNPavan Yalamanchili
- This reverts commit 0d85922d116879448485ef88ae21e83a9255a0b0. - Includes fixes for TemporalRowConvolution
2017-02-15add localsoumith
2017-02-15implementation of bias=False for VolConv.cuEli Stevens
2017-02-08Merge pull request #440 from gchanan/glugenericSoumith Chintala
Use THCTensor rather than THCudaTensor in GatedLinearUnit
2017-02-08Use THCTensor rather than THCudaTensor in THCUNN.h definition ofGregory Chanan
GatedLinearUnit.
2017-02-03Merge pull request #438 from huihuifan/betterCudaGLUSoumith Chintala
cuda implementation of Gated Linear Unit, fixed issues with genericiz…
2017-02-03cuda implementation of Gated Linear Unit, fixed issues with genericizationAngela Fan
2017-02-02Merge pull request #437 from torch/revert-430-newCudaGLUSoumith Chintala
Revert "cuda implementation of Gated Linear Unit, fixed issues with genericization"
2017-02-02Revert "cuda implementation of Gated Linear Unit, fixed issues with ↵revert-430-newCudaGLUSoumith Chintala
genericization"
2017-02-02Merge pull request #430 from huihuifan/newCudaGLUSoumith Chintala
cuda implementation of Gated Linear Unit, fixed issues with genericization
2017-02-01VolumetricFractionalMaxPooling like spatialjeremy
2017-01-31Merge pull request #433 from apaszke/unpooling_leakSoumith Chintala
Fix memory leak in SpatialMaxUnpooling
2017-01-31Fix memory leak in SpatialMaxUnpoolingAdam Paszke
2017-01-29Merge pull request #431 from apaszke/assertsSoumith Chintala
Compile with asserts by default
2017-01-29Compile with asserts by defaultAdam Paszke
2017-01-27Added cunn support for TemporalRowConvolutionMM (#415)Will Frey
* Added cunn TemporalRowConvolutionMM support
2017-01-27cuda implementation of Gated Linear Unit, fixed issues with genericizationAngela Fan
2017-01-26Merge pull request #428 from colesbury/CMakeSoumith Chintala
Install THCUNN.h and generic/THCUNN.h
2017-01-26Install THCUNN.h and generic/THCUNN.hSam Gross
The THCApply.cuh is moved to the .cu files so that THCUNN.h can be compiled by a standard C compiler.
2017-01-26Merge pull request #427 from torch/revert-416-half-fixesSoumith Chintala
Revert "Convert real to accreal in libTHCUNN"
2017-01-26Revert "Convert real to accreal in libTHCUNN"revert-416-half-fixesSoumith Chintala
2017-01-25Merge pull request #426 from torch/revert-421-cudaGLUSoumith Chintala
Revert "cuda implementation of Gated Linear Unit"
2017-01-25Revert "cuda implementation of Gated Linear Unit"revert-421-cudaGLUSoumith Chintala
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-15 07:47:06 +0300