Age | Commit message (Collapse) | Author | |
---|---|---|---|
2016-12-29 | Add THHalfTensor support to cutorch (#655) | gchanan | |
* Add THHalfTensor support to cutorch. | |||
2015-12-26 | Add generic CudaTensor types to cutorch | Adam Lerer | |
2015-12-26 | Move files to generic folders (preparation) | Adam Lerer | |
2015-09-08 | Add copyAsync for asynchronous copies between host and device. | Dominik Grewe | |
2015-05-13 | Revert "Auto device: API changes, bug fixes, README.md" | Adam Lerer | |
This reverts commit d88ac24c712e3a40d4aaf3ac2d043bd79ba4280e. Revert "Auto device mode, plus allocation helper functions." This reverts commit 47a2f6de252c2254234edfc1c6115229b5383bac. | |||
2015-04-29 | Auto device mode, plus allocation helper functions. | Adam Lerer | |
This diff introduces an alternative way of writing multi-GPU cutorch code. In this mode, the location of each tensor is specified, and the appropriate GPU for each kernel is determined automatically based on the location of its argument tensors. It's backwards-compatible and interoperable with the old-style multi-GPU API. | |||
2015-04-07 | adding cutorch streams | soumith | |
2015-04-01 | revamps TensorMath to remove sync points at many places, adds maskedSelect ↵ | soumith | |
and maskedFill operations (and tests). Also adds generic Reduce and Apply kernels that can be reused. | |||
2015-01-14 | Pass a state to every THC function. | Dominik Grewe | |
Every THC function gets a THCState pointer as the first argument. Some generic files that were previously included have been instantiated because TH functions currently don't get a state parameter. | |||
2014-11-25 | Cast allocation to avoid compiler warnings. | Dominik Grewe | |
2014-11-12 | adding getDevice for tensor, manualSeedAll and seedAll | soumith | |
2014-06-19 | added indexCopy | Hycis | |
2014-06-15 | Revert "Merge pull request #20 from hycis/master" | Soumith Chintala | |
This reverts commit b6fa093ecf020147f697310d4ccbc485280dcbd9, reversing changes made to 12eeb82762e9db0d950047f56cda4b950a52f55d. | |||
2014-06-15 | optimized indexCopy | Hycis | |
2014-06-13 | added indexFill and simplify index | Hycis | |
2014-06-12 | optimized indexSelect | Hycis | |
2014-04-08 | Add indexCopy and indexFill implementations | Ioannis Antonoglou | |
2014-04-04 | add some comments | Ioannis Antonoglou | |
2014-04-04 | Add index in cutorch | Ioannis Antonoglou | |
2013-08-11 | introduce THC_API and work around log1p issue. | Leon Bottou | |
2013-04-16 | indexing operators work with CUDA, well they give out errors. | koray kavukcuoglu | |
2013-04-16 | implement cuda versions for index operators | koray kavukcuoglu | |
2012-08-13 | cuda package now complies with the new luaT API | Ronan Collobert | |
2012-02-06 | little fix (maskedSelect does not exist yet on Cuda) | Ronan Collobert | |
2012-02-03 | Merge branch 'logical' into newpack | Ronan Collobert | |
Conflicts: pkg/torch/TensorMathWrap.lua | |||
2012-01-25 | initial revamp of torch7 tree | Ronan Collobert | |