search

kroggen / mamba.c

Inference of Mamba models in pure C
175 stars 10 forks source link

Backpropagation / Training loop #3

Open Xezusa opened 6 months ago

Xezusa commented 6 months ago

Just wondering if you had any plans to implement cross_entropy_loss or backpropagation for training.

kroggen commented 1 month ago

This guy made it:

https://github.com/Named666/mamba.c

I have not tested, so not sure if it works already