attempt to autodetect device #14
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ok running on mps, i think because the model is so small. maybe the autodetect logic should be improved. not 100% sure if i want to merge this in this module hmm...
|
This is indeed a known issue in pytorch specially on smaller kernels with MPS. specially since not all operations are made for MPS and can still revert to being run on the CPU That is why apple also has developed MLX which caters specifically to the unique apple M architecture with having unified memory the CPU and GPU can both work on the kernels, here is the docs if you would like to read more about the project |
|
Yes I have observed in my pytorch projects that cpu is faster than mps (for any scale that fits in my ram, and this is for M2 chip) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
but actually this is slower on my macbook running on mps, i think because the model is so small. maybe the autodetect logic should be improved. not 100% sure if i want to merge this in this module hmm...