Abdul Fatir 5242d986f4 Remove float32 casting for cumsum (#53) ... *Description of changes:* This PR removes casting to `fp32` for the `cumsum` operation and upgrades `mlx` to `~=0.10.0` which adds `bf16` support for `cumsum`. Related: https://github.com/ml-explore/mlx/issues/959 By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice. Co-authored-by: Abdul Fatir Ansari <ansarnd@amazon.com>		2024-04-12 20:41:12 +02:00
..
dummy-chronos-model	Upload code	2024-03-13 09:58:39 +01:00
test_chronos_mlx.py	Remove float32 casting for cumsum (#53)	2024-04-12 20:41:12 +02:00