summaryrefslogtreecommitdiff
path: root/candle-core/tests
Commit message (Expand)AuthorAgeFilesLines
* UG metal integration. (#2580)Laurent Mazare2024-10-271-5/+11
* Support for UG kernels. (#2579)Laurent Mazare2024-10-271-0/+30
* Testcases (#2567)Anubhab Bandyopadhyay2024-10-171-0/+274
* Efficient implementation of `Tensor::ones()` for `metal` (#2512)Anubhab Bandyopadhyay2024-10-011-0/+30
* Bugfix for the metal elu kernel. (#2490)Laurent Mazare2024-09-211-0/+13
* Add a minimal test for the metal bf16 matmul. (#2381)Laurent Mazare2024-08-011-0/+20
* Fix log_sum_exp to handle large positive/negative inputs (#2367)Yun-Jhong Wu2024-08-011-3/+21
* Fix for backprop in ConvTranspose2D with stride of 2 (#2337)Ivor Wanders2024-07-171-0/+97
* Add a slice_set op. (#2193)Laurent Mazare2024-05-181-0/+25
* Add SliceSafetensors. (#2179)Laurent Mazare2024-05-111-0/+47
* Add the cuda dequantize f16 kernels. (#2137)Laurent Mazare2024-04-281-1/+120
* Add a sort function. (#2134)Laurent Mazare2024-04-281-0/+18
* Add argsort. (#2132)Laurent Mazare2024-04-271-0/+17
* Add more QMMV cuda kernels. (#2077)Laurent Mazare2024-04-181-7/+15
* Add the mmv kernels for small batch sizes. (#2075)Laurent Mazare2024-04-161-1/+35
* Fix for the batch dim in the quantized matmul example. (#2073)Laurent Mazare2024-04-151-36/+36
* Handle zero dims in some simple operations. (#2064)Laurent Mazare2024-04-151-0/+22
* Add support for "sign" on tensors (#2012)Thomas Santerre2024-04-041-0/+8
* Fix the matmul layout for accelerate & mkl. (#2011)Laurent Mazare2024-04-042-22/+4
* Optimize the gelu f16 opt. (#2008)Laurent Mazare2024-04-041-0/+8
* Improve the handling of matmul with squeezed layouts. (#1998)Laurent Mazare2024-04-022-100/+119
* Support i64 in index-select on metal. (#1951)Laurent Mazare2024-03-271-37/+42
* More flexible matmul contiguity checks. (#1949)Laurent Mazare2024-03-271-0/+25
* Preliminary support for inplace ops. (#1921)Laurent Mazare2024-03-231-0/+31
* Backwards for ConvTranspose2D (#1910)Kirpal Grewal2024-03-231-7/+154
* Add support for conv_transpose2d on Metal backend (#1903)Thomas Santerre2024-03-211-69/+55
* Avoid copying the data on squeeze and unsqueeze. (#1884)Laurent Mazare2024-03-201-1/+15
* Add support for conv_transpose1d for metal backend (#1874)Thomas Santerre2024-03-191-5/+0
* Add avg_pool2d metal implementation for the metal backend (#1869)Thomas Santerre2024-03-181-3/+0
* Add support for max_pool2d for Metal backend (#1863)Thomas Santerre2024-03-181-3/+0
* Optimize the cat operation on contiguous tensors (#1855)Laurent Mazare2024-03-174-59/+121
* Fast CPU kernel for transposed 1d convolutions. (#1822)Laurent Mazare2024-03-081-20/+24
* Metal random-generation bug fixes (#1811)Niklas Hallqvist2024-03-081-0/+25
* Add a cuda kernel for dequantizing q8_0. (#1804)Laurent Mazare2024-03-051-4/+0
* Handle Q5_0 and Q5_1 quants in cuda.laurent2024-02-291-8/+0
* Fix the block size for some cuda kernels. (#1767)Laurent Mazare2024-02-271-32/+0
* Support for attention bias in gemma + refactor things a bit. (#1744)Laurent Mazare2024-02-221-46/+28
* Add grads for interpolate1d (#1742)Kirpal Grewal2024-02-221-0/+33
* Bugfix for conv-transpose1d (#1734)Laurent Mazare2024-02-191-0/+14
* Support for groups in conv-transpose1d. (#1731)Laurent Mazare2024-02-181-1/+1
* feat: add silu activation function (#1706)OlivierDehaene2024-02-142-0/+20
* ConvTranspose1d cuda support. (#1697)Laurent Mazare2024-02-121-11/+9
* Add support for loading Fortran contiguous tensors (#1672)Dilshod Tadjibaev2024-02-073-0/+44
* Enhance pickle to retrieve state_dict with a given key (#1671)Dilshod Tadjibaev2024-02-063-1/+11
* Fix rustfmt. (#1669)Laurent Mazare2024-02-061-1/+1
* Fix: pth files don't load on Windows (#1661)Roma Klapaukh2024-02-063-0/+14
* Quantized GGUF style (#1523)Nicolas Patry2024-01-171-143/+430
* Add the pow operator. (#1583)Laurent Mazare2024-01-131-2/+14
* Bugfix for dequantizing q5k layers. (#1569)Laurent Mazare2024-01-111-1/+1
* Simpler repro for the neon optimization issue + bugfix (#1544)Laurent Mazare2024-01-071-16/+41
generated by cgit v1.2.3 (git 2.39.1) at 2025-04-05 00:34:17 +0000