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

github.com/soumith/cudnn.torch.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
path: root/SpatialConvolution.lua
AgeCommit message (Expand)Author
2016-11-10Improved existing 16->32 fallback. Added performance-based fallback.Boris Fomitchev
2016-11-02clearState in FindExSergey Zagoruyko
2016-10-19Added new refactoring for convolution and filter descriptorsBoris Fomitchev
2016-10-19Code review changesBoris Fomitchev
2016-09-26Fixed sticky algo modesBoris Fomitchev
2016-09-23Revamped workspace handling in find.luaBoris Fomitchev
2016-08-30Addressed code review commentsBoris Fomitchev
2016-08-23Stream awareness restored. Better WS encapsulationBoris Fomitchev
2016-08-21FindEx implementetation + refactoring, take 3Boris Fomitchev
2016-08-06working double precisionsoumith
2016-08-06Revert "Refactoring CUDNN Find"revert-231-algoSoumith Chintala
2016-08-04Completing cudnnFind refactoring; addressing code review notesBoris Fomitchev
2016-08-03Refactoring cudnnFindBoris Fomitchev
2016-08-02Fixed algo locationBoris Fomitchev
2016-08-02Refactoring CUDNN FindBoris Fomitchev
2016-07-08handle half true/pseudoSergey Zagoruyko
2016-06-23half, double with testsSergey Zagoruyko
2016-06-23forward fp16 convolution worksSergey Zagoruyko
2016-04-20move gradInput resize to updateGradInputSergey Zagoruyko
2016-04-13R5 rebaseSergey Zagoruyko
2016-04-08use nn.utils.clearSergey Zagoruyko
2016-03-30move gradInput resize to updateGradInputSergey Zagoruyko
2016-03-22check if padW and padH existSergey Zagoruyko
2016-03-17adding _input and _gradOutput to clearStatesoumith
2016-02-26allow non-contiguous inputs for Spatial and Volumetric convolutionsoumith
2016-02-16removing streams optimizationsoumith
2016-02-12clearStateSergey Zagoruyko
2016-01-26adding :noBias() to SpatialConvolutionsoumith
2016-01-26cudnn<>nn conversionSergey Zagoruyko
2016-01-26syncing uptil master/R3 9bc3cbac4f054438f5e77824f868cd94e9e22f81soumith
2016-01-11Make bias optional in cudnn.SpatialConvolutionSam Gross
2015-11-06integrating changes from mastersoumith
2015-10-14Conmpleting the mergeBoris Fomitchev
2015-10-14Fixing conflictsBoris Fomitchev
2015-10-14Bugfix by AndreiBoris Fomitchev
2015-10-09Merge pull request #64 from borisfom/masterSoumith Chintala
2015-10-07cudnn.h fully synced for V4. addTensor_v2 uses changed to new AddTensor API.Boris Fomitchev
2015-10-05Merge pull request #62 from borisfom/masterSoumith Chintala
2015-10-03Initial cudnn_v4 compatibility patch. cudnn4 Batch Normaliztion Support added...Boris Fomitchev
2015-10-02fix typo SpatialConvolution with groupsAlban Desmaison
2015-10-01adding caching for autotuner resultssoumith
2015-09-15fastest mode disabled in Spatial convsoumith
2015-09-15adding proper documentation for new modessoumith
2015-09-15add back stuff that went missing in mergesoumith
2015-09-15whitespace cleanups, fixing logsoftmax testsoumith
2015-09-15fix as suggested in #44soumith
2015-08-25small fixessoumith
2015-08-23flag to enable or not to enable auto-tunersoumith
2015-08-21fix for cudnn r3 perf bugsoumith
2015-08-20shared workspacesoumith
generated by cgit v1.2.3 (git 2.25.1) at 2024-05-15 21:32:36 +0300