search

Efficient-Large-Model / VILA

VILA - a multi-image visual language model with training, inference and evaluation recipe, deployable from cloud to edge (Jetson Orin and laptops)
Apache License 2.0
878 stars 55 forks source link

Would you consider releasing code that supports lora training 40b model? #58

Open Key-lei opened 1 month ago

Key-lei commented 1 month ago

Very excellent work! When using lora to train a 40b model in my task, I found during the loading inference process that lora did not save the weight of thevision tower, so the effect of my task was very poor. Would you consider supporting lora training and loading with official code?

Efficient-Large-Language-Model commented 1 month ago

Lora training is not well supported. I would recommend doing a regular finetuning.