summaryrefslogtreecommitdiff
path: root/candle-core/src/dummy_cuda_backend.rs
Commit message (Expand)AuthorAgeFilesLines
* 20241118 docs (#2629)zachcp2024-11-191-0/+2
* Add support for cuda streams. (#2532)Laurent Mazare2024-10-021-0/+6
* Make it possible to use TF32 accumulation in F32 matmuls. (#2178)Laurent Mazare2024-05-111-0/+10
* Add a toggle for F16/BF16 accumulation in gemm. (#2141)Laurent Mazare2024-04-291-0/+20
* Add StorageRef. (#2113)Laurent Mazare2024-04-231-0/+4
* Add a synchronize method to devices. (#2055)Laurent Mazare2024-04-141-0/+4
* Add the alloc_uninit function. (#1901)Laurent Mazare2024-03-221-0/+4
* Async tensor copying. (#1900)Laurent Mazare2024-03-211-0/+4
* Optimize the cat operation on contiguous tensors (#1855)Laurent Mazare2024-03-171-0/+13
* Add the conv-transpose1d op. (#1251)Laurent Mazare2023-11-031-0/+10
* Make the cuda rng seedable. (#1056)Laurent Mazare2023-10-081-0/+4
* Add 1d upsampling. (#839)Laurent Mazare2023-09-131-0/+4
* Add the powf op. (#664)Laurent Mazare2023-08-291-0/+4
* Add conv-transpose. (#635)Laurent Mazare2023-08-281-0/+10
* add max_pool2d (#371)LeeeSe2023-08-091-0/+4
* Add more conv2d support. (#340)Laurent Mazare2023-08-081-0/+10
* CPU implementation for upsample-nearest2d. (#339)Laurent Mazare2023-08-071-0/+4
* Some CLIP fixes for stable diffusion. (#338)Laurent Mazare2023-08-071-0/+4
* Remove the embedding ops in favor of index-select. (#299)Laurent Mazare2023-08-021-3/+0
* Softmax numerical stability. (#267)Laurent Mazare2023-07-281-4/+0
* Add the gather op. (#219)Laurent Mazare2023-07-221-0/+16
* Start adding index-add.laurent2023-07-211-0/+11
* Add the index-select op. (#209)Laurent Mazare2023-07-201-0/+3
* Op refactor (#208)Laurent Mazare2023-07-201-8/+3
* Add the comparison operations. (#207)Laurent Mazare2023-07-201-1/+6
* Add some more developed training examples. (#199)Laurent Mazare2023-07-191-1/+1
* Expliciting how to enable cuda.Nicolas Patry2023-07-141-1/+1
* Modular backends (#138)Laurent Mazare2023-07-111-63/+55
* Allow for uniform initialization in a single step. (#136)Laurent Mazare2023-07-111-1/+1
* Random initializers. (#128)Laurent Mazare2023-07-101-0/+8
* Add the elu op. (#113)Laurent Mazare2023-07-091-0/+4
* Add more of the conv1d op.laurent2023-07-041-2/+1
* Sketch the conv1d op.laurent2023-07-041-0/+11
* Simplify the pattern matching logic in the cuda backend.laurent2023-06-291-1/+1
* Adapt the cuda bits.laurent2023-06-281-1/+1
* Get the cpu backend to compile.laurent2023-06-281-31/+15
* Refactor the hierarchy.Nicolas Patry2023-06-271-0/+136
generated by cgit v1.2.3 (git 2.39.1) at 2025-04-05 08:22:22 +0000