]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
overview / pkg / ggml / sources / ggml / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: aec8550)
`ggml_cuda_cpy` support for 4d tensors and float16->float32 upcasting (#686)
authorJohn Balis <redacted>
Mon, 29 Jan 2024 12:37:33 +0000 (06:37 -0600)
committerGitHub <redacted>
Mon, 29 Jan 2024 12:37:33 +0000 (13:37 +0100)
commitb2a5c34073e68b841568b2787ed5d15ae6a8d475
tree89315299e22ed3700f55c118ccc1dc7fab4bc3d5tree
parentaec8550d966c82d3e6f10dd7de309afe5c424857commit | diff
`ggml_cuda_cpy` support for 4d tensors and float16->float32 upcasting (#686)

* added cuda float16->float32 upcasting to ggml_cuda_cpy

* added ability to copy 4d tensors with the cuda backend

* added tests for float16_>float32 upcast and 4d tensor cuda copys

* added 4d copy test for float32->float16 copy

* applied patch suggested by @iamlemec

* simplify cpy tests

---------

Co-authored-by: slaren <redacted>
src/ggml-cuda.cu diff | blob | history
tests/test-backend-ops.cpp diff | blob | history
Packaging of ggml-org/ggml