search

TinyLLaVA / TinyLLaVA_Factory

A Framework of Small-scale Large Multimodal Models
https://arxiv.org/abs/2402.14289
Apache License 2.0
564 stars 53 forks source link

Data download #43

Open wangfengjuan opened 5 months ago

wangfengjuan commented 5 months ago

image Follow the link of the ocr data set to download the relevant image, but when I execute the fine-tuning instruction CUDA_VISIBLE_DEVICES=0,1 bash scripts/tiny_llava/finetune/finetune_lora.sh, I always encounter the above problems. I wonder if you have encountered similar problems. , I checked in the image folder, it does exist this image. looking forward to your reply!

baichuanzhou commented 5 months ago

Can you try to open the image with PIL.Image.open('your_path')?

wangfengjuan commented 5 months ago

Can you try to open the image with PIL.Image.open('your_path')? Thank you for your reply. There is indeed a problem with the image and it cannot be opened. I found that many of the downloaded images have problems. I wonder if it would be convenient for you to share the OCR-VQA data set you downloaded? You can use the network disk or leave your contact information. I've been struggling for a week, but the download is still incomplete. Thank you very much for your help.

baichuanzhou commented 5 months ago

For legal reasons, we cannot directly distribute these datasets.