Age | Commit message (Collapse) | Author | |
---|---|---|---|
2015-03-03 | Add batch mode to VolumetricMaxPooling | Max Losch | |
2015-02-23 | CosineEmbeddingCriterion: zero grads below margin | CeĢdric Deltheil | |
2015-02-15 | Merge pull request #165 from szagoruyko/prelu | Soumith Chintala | |
Parametric ReLU | |||
2015-02-15 | added PReLU | Sergey Zagoruyko | |
2015-02-06 | Module:listModules() | Nicholas Leonard | |
2015-02-05 | multimargin supports p=2 | Sergey Zagoruyko | |
2015-01-31 | fixed L1Penalty constructor arguments | Soumith Chintala | |
2015-01-27 | Matrix matrix multiplication layer. | Dominik Grewe | |
Multiplies two matrices (or two batches of matrices). This is a pure Lua implementation taking advantage of the new torch.bmm function for minibatch inputs. | |||
2015-01-20 | Merge https://github.com/torch/nn into convMM_dw | soumith | |
2015-01-20 | SpatialConvolutionMM supports dw/dh != 1 | Aysegul Dundar | |
2015-01-19 | Merge https://github.com/fmassa/nn | soumith | |
2015-01-18 | removed global variable in ParallelTable test | Sergey Zagoruyko | |
2015-01-16 | Added SpatialAdaptiveMaxPooling | fsuzanomassa | |
2015-01-16 | Revert "Added SpatialAdaptiveMaxPooling" | Soumith Chintala | |
2015-01-16 | Added SpatialAdaptiveMaxPooling | fsuzanomassa | |
2015-01-10 | Merge pull request #128 from nicholas-leonard/euclidean2 | Soumith Chintala | |
Euclidean batch support | |||
2015-01-10 | Merge pull request #133 from nicholas-leonard/weightedeuclidean2 | Soumith Chintala | |
WeightedEuclidean optimizations | |||
2015-01-10 | WeightedEuclidean optimizations | nicholas-leonard | |
2015-01-10 | Euclidean | nicholas-leonard | |
2015-01-08 | Added more informative pretty-printing. | Jonathan J Hunt | |
2015-01-08 | CMul optimizations, doc and unit tests | nicholas-leonard | |
2015-01-07 | Merge pull request #136 from nicholas-leonard/copy | Soumith Chintala | |
new Copy constructor arguments | |||
2015-01-07 | Merge pull request #135 from nicholas-leonard/parallel | Soumith Chintala | |
Parallel, Container & cie | |||
2015-01-07 | Merge branch 'master' of github.com:torch/nn | Soumith Chintala | |
2015-01-07 | new Copy constructor arguments | nicholas-leonard | |
2015-01-06 | Parallel optimization. ParallelTable inherits Container. unit tests | nicholas-leonard | |
2015-01-03 | refactoring all the common container code into nn.Container | soumith | |
2014-12-23 | Merge pull request #119 from nicholas-leonard/weightedeuclidean | Soumith Chintala | |
WeightedEuclidean batch mode | |||
2014-12-23 | WeightedEuclidean batch mode | Nicholas Leonard | |
2014-12-21 | VolumetricConvolution batch mode + test | Soumith Chintala | |
2014-12-16 | adding batch processing to CMul | germank | |
Changed CMul reset of weights to random initialization | |||
2014-12-04 | removing the requirement for providing size in nn.Mul | Soumith Chintala | |
2014-12-03 | Merge branch 'master' of github.com:torch/nn | Soumith Chintala | |
2014-11-25 | SpatialAveragePooling | Nicholas Leonard | |
2014-11-24 | [Torch] Move test.lua to the top level | Andrew Tulloch | |
Summary: This removes some CMakeLists.txt manipulation. Test Plan: ``` luarocks make rocks/nn-scm-1.rockspec && th -lnn -e "nn.test()" ... Completed 710 asserts in 81 tests with 0 errors ``` |