open_flamingo icon indicating copy to clipboard operation
open_flamingo copied to clipboard

Published 20 hours ago • verified publisher icon mlfoundations

How to try my own fine-tuning experiments

Open minghehe-nobug opened this issue 11 months ago • 1 comments

Hello! I'm new to multimodal training. Inspired by this exciting project, I hope to try my own fine-tuning experiments on interleaved data.

Currently, I have downloaded the pro-trained model (3B) and completed the inference process. But can anyone help me how to write the parameters for the "torchrun" script?

The challenge for me is how to change the two parameters "--laion_shards" and "--mmc4_shards" to my own. And how to modify the original code without using "LAION-2B" (the data set is too large)?

Thanks!

minghehe-nobug avatar Mar 05 '24 03:03 minghehe-nobug

I want to fine-tune the model, too. Can I ask how much GPU resource is needed. I could have 4x A5000 GPUs, are they enough to complete it?

Roberyan avatar May 02 '24 03:05 Roberyan