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

github.com/torch/nn.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
path: root/ClassNLLCriterion.lua
AgeCommit message (Collapse)Author
2017-05-21Revert "Revert "ClassNLLCriterion supports missing targets""revert-1217-revert-1215-ClassNLLCriterion-missing-targetSoumith Chintala
2017-05-15Revert "ClassNLLCriterion supports missing targets"Soumith Chintala
2017-05-12ClassNLLCriterion supports missing targetsNicholas Leonard
2016-12-21fixing bug in ClassNLLCriterion for single targetssoumith
2016-10-26Convert ClassNLLCriterion targets for all cuda types.Gregory Chanan
2016-10-03nll: fixed bad cast target is CudaLongTensorMartin Simonovsky
2016-10-02fix 1.0-0 checkssoumith
2016-10-02The changes introduced in commit #979 have broken cutorch/cunn 1.0-0 because ↵Martin Simonovsky
there is no `cudaLong` there. This makes the conversion conditional.
2016-10-02fixing *NLLCriterion for non mini-batch casessoumith
2016-09-27making ClassNLLCriterion targets consistent between cpu and cudaclassnllfixsoumith
2016-04-26Fix ClassNLLCriterion bufferAdam Paszke
2015-12-31Add functional conversion of ClassNLLCriterionAndreas Köpf
Fixed indentation in THNN.lua to 3-spaces.
2015-11-06Incorrect type when target is a number.Clement Farabet
2015-09-13Set default value of total weight in ClassNLL to 1Jonghoon Jin
`total_weight_tensor` becomes to have a meaningful number after loss:forward(). Calling loss:backward() without loss:forward() will end up with zero output https://github.com/torch/nn/commit/ef2e3d290418c411add8eee8dd5ca7d486362ad8#diff-e3ab2ae1e4ca35f9a37e0702421d52ffR97. This commit consider the case.
2015-08-22Moved ClassNLLCriterion to C. Normalization done by total weight instead of sizeAdam Lerer
2015-07-24Aded weight support to ClassNLLCriterion cuda.Jonathan Tompson
2015-04-30#ClassNLLCriterion returns number of weightsdiz_vara
2015-04-21ClassNLLCriterion now can take target as a number or Tensor of dim1 with 1 ↵soumith
element in non-minibatch mode
2014-12-25correction in batch modeAysegul Dundar
2014-12-24fixing small cuda typing issue for ClassNLLCriterionsoumith
2014-11-21Support CUDA implementation of ClassNLLCriterionAndrew Tulloch
2014-10-27adding addtional weights parameter to ClassNLLCriterion, which is 1D tensor ↵Yossi Biton
assigning different weight to each class
2014-05-09fixed typo in ClassNLLCriterionnicholas-leonard
2012-01-25initial revamp of torch7 treeRonan Collobert
    generated by cgit v1.2.3 (git 2.25.1) at 2024-06-07 06:41:54 +0300