Curious, does MLX allow to just port my pytorch code? I don't want to rewrite it. It takes a lot of time to write my pytorch code.

[brando90]

Curious, does MLX allow to just port my pytorch code? I don't want to rewrite it. It takes a lot of time to write my pytorch code.

[awni]

If you are using high-level PyToch e.g. nn then mostly you can replace those modules with mlx.nn. But right now the replacement is not perfectly 1:1 so it does require some work.

[brando90]

I have hugging face code I do NOT want to rewrite. That's the best use case. ----- Brando Miranda Ph.D. Student Computer Science, Stanford University EDGE Scholar, Stanford University ***@***.*** website: https://brando90.github.io/brandomiranda/home.html mentorship opportunities: https://brando90.github.io/brandomiranda/prospective-collaborations.html On Jan 29, 2024, at 10:57 AM, Awni Hannun ***@***.***> wrote: If you are using high-level PyToch e.g. nn then mostly you can replace those modules with mlx.nn. But right now the replacement is not perfectly 1:1 so it does require some work. — Reply to this email directly, view it on GitHub<#582 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AAOE6LSSB3IICHFVCUCT7JLYQ7WIZAVCNFSM6AAAAABCP6LYIKVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4DEOBYHA2TE>. You are receiving this because you authored the thread.Message ID: ***@***.***>