Skip to content

clin155/CSE595Final

Repository files navigation

CSE595Final

How to run: set HUGGINGFACE_HUB_TOKEN and WANDB_API_KEY environment variables for fine tuning: python finetune_llava.py args: force_reload_dataset -> dataset is cached in DATASET_DIR, add this arg to refresh it zip_directory -> location of the generated_images.zip (the zip of the chat-gpt generated images) include_gen_images -> whether to include the generated chatgpt images test_only -> only run inference no fine tuning python finetune_llama.py args: all the same args except for test_only

python qwen-vl-7b-sft.py args: all the same args except for test_only

make sure torch==2.4.1

for zero-shot: python zero_shot_llava.py

python test_qwen.py

python fintune_llama --zero_shot=True

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages