search

haotian-liu / LLaVA

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.
https://llava.hliu.cc
Apache License 2.0
20.42k stars 2.26k forks source link

feature_img_folder_None #1736

Open pegah-kh opened 1 month ago

pegah-kh commented 1 month ago

To solve the problem: when the image_folder is None in https://github.com/haotian-liu/LLaVA/blob/c121f0432da27facab705978f83c4ada465e46fd/llava/train/train.py#L701 an error is raised.