Age | Commit message (Collapse) | Author | |
---|---|---|---|
2017-02-23 | fix AVX2 detection bugsavx2detect | Soumith Chintala | |
2017-02-23 | Merge pull request #933 from twitter-forks/bitops | Nicholas Léonard | |
Bitwise operations | |||
2017-02-23 | Adding test for bitxor | Pavan Yalamanchili | |
2017-02-10 | Fixes to support short and char tensors for bitwise operations | Pavan Yalamanchili | |
2017-02-10 | Adding documentation for bitwise operations | Pavan Yalamanchili | |
2017-02-10 | Adding bitand, bitor and bitxor | Pavan Yalamanchili | |
2017-02-10 | Changes to shift operations | Pavan Yalamanchili | |
- renaming lsh -> lshift, rsh -> rshift - adding componentwise functions | |||
2017-02-10 | Avoiding duplicated unsigned as it causes error on gcc. | Conrado Miranda | |
2017-02-10 | Added shift operations. | Conrado Miranda | |
2017-02-09 | Merge pull request #926 from colesbury/view | Soumith Chintala | |
Add code for 'view' to TH | |||
2017-02-09 | Merge pull request #929 from colesbury/unsqueeze | Soumith Chintala | |
Add unsqueeze1d to TH | |||
2017-02-09 | Merge pull request #930 from colesbury/resize | Soumith Chintala | |
Expose rawSet and rawResize as resizeNd and setStorageNd | |||
2017-02-08 | Add unsqueeze1d to TH | Sam Gross | |
Unsqueeze inserts a singleton dimension. Unlike view, it doesn't require the tensor to be contiguous. | |||
2017-02-08 | Expose rawSet and rawResize as resizeNd and setStorageNd | Sam Gross | |
These methods are useful from C because they don't require constructing THLongStorages to wrap the sizes and strides, which can lead to leaked memory in case of an error. Instead the sizes and strides can be represented on the stack using standard C long arrays. | |||
2017-02-07 | Add code for 'view' to TH | Sam Gross | |
2017-02-04 | half<->float conversion cleanup (#901) | Boris Fomitchev | |
* half<->float conversion cleanup | |||
2017-02-02 | Merge pull request #922 from cliffwoolley/patch-1 | Soumith Chintala | |
Remove erroneous proprietary license header | |||
2017-02-01 | Update THHalf.c | Cliff Woolley | |
2017-02-01 | Remove erroneous proprietary license header | Cliff Woolley | |
This change was approved by NVIDIA Legal, and I am authorized to make the change on behalf of the company. | |||
2017-01-30 | Merge pull request #921 from apaszke/cat_fix | Soumith Chintala | |
Fix bug in cat (non-contiguous first input) | |||
2017-01-30 | Add Ubuntu 16.04 lib paths in CMake | Adam Paszke | |
2017-01-29 | Fix bug in cat (non-contiguous first input) | Adam Paszke | |
2017-01-24 | Merge pull request #915 from pavanky/convert | Soumith Chintala | |
Macros to convert between real and accreal | |||
2017-01-18 | Adding macros to convert between real and accreal | Pavan Yalamanchili | |
2017-01-18 | Remove redundant code in THGenerateAllTypes.h | Pavan Yalamanchili | |
2017-01-14 | Merge pull request #911 from gchanan/convWarning | Soumith Chintala | |
Avoid strict aliasing warning in float/half conversions. | |||
2017-01-14 | Avoid strict aliasing warning in float/half conversions. | Gregory Chanan | |
Verified that at least for GCC 4.47 this generates identical code. | |||
2017-01-11 | Merge pull request #908 from CDLuminate/really-fix-arm64-build | Soumith Chintala | |
simd.h: really fix the arm64 (i.e. Aarch64) build | |||
2017-01-11 | simd.h: really fix the arm64 (i.e. Aarch64) build | Zhou Mo | |
2017-01-11 | Merge pull request #906 from torch/installpaths | Ronan Collobert | |
provide install paths through require 'torch.paths' | |||
2017-01-11 | provide install paths through require 'torch.paths'installpaths | Ronan Collobert | |
2017-01-10 | fixing arm64 build | Soumith Chintala | |
2017-01-06 | Merge pull request #899 from gchanan/TensorCopyHalf | Soumith Chintala | |
Fix THHalf issues with MSVC. | |||
2017-01-05 | Fix THHalf issues with MSVC. | Gregory Chanan | |
2017-01-05 | Merge pull request #895 from colesbury/threadArgErrorHandler | Soumith Chintala | |
Fix condition for threadArgErrorHandler | |||
2017-01-05 | Fix condition for threadArgErrorHandler | Sam Gross | |
Some error handlers may not have any data associated with them | |||
2017-01-03 | Merge pull request #892 from temerick/patch-3 | Soumith Chintala | |
Update FindARM.cmake | |||
2017-01-03 | Update FindARM.cmake | Timothy Emerick | |
Fix typos | |||
2017-01-02 | small cmake fix | Soumith Chintala | |
2017-01-01 | Merge pull request #873 from nkoumchatzky/contiguous-cat-1d | Soumith Chintala | |
Different code path for contiguous tensors along dimension 1 in cat + fix cat bug + fix ambiguity | |||
2017-01-01 | Merge branch 'master' into contiguous-cat-1d | Soumith Chintala | |
2017-01-01 | Merge pull request #890 from torch/halfcleanup | Soumith Chintala | |
Removing Half conditional macros, and enabling half compilation by default | |||
2017-01-01 | Removing TH_GENERIC_USE_HALF, TH_NATIVE_HALF, TH_GENERIC_NO_MATH (replaced ↵ | soumith | |
where appropriate with TH_REAL_IS_HALF), removed half from THGenerateAllTypes, added an explicit THGenerateHalfType.h | |||
2017-01-01 | Merge pull request #889 from torch/revert-888-halfgeneric | Soumith Chintala | |
Revert "TH_GENERIC_USE_HALF=1 by default, half enabled by default" | |||
2017-01-01 | Revert "TH_GENERIC_USE_HALF=1 by default, half enabled by default"revert-888-halfgeneric | Soumith Chintala | |
2017-01-01 | Merge pull request #888 from torch/halfgeneric | Soumith Chintala | |
TH_GENERIC_USE_HALF=1 by default, half enabled by default | |||
2017-01-01 | TH_GENERIC_USE_HALF=1 by default, half enabled by defaulthalfgeneric | soumith | |
2016-12-31 | cmake C flags fix | Soumith Chintala | |
2016-12-31 | adding TH_SO_VERSION option | Soumith Chintala | |
2016-12-31 | Merge pull request #886 from torch/simdcpuidfix | Soumith Chintala | |
use __get_cpuid when available |