Groma icon indicating copy to clipboard operation
Groma copied to clipboard

unable to load local weight

Open liukc19 opened this issue 1 year ago • 5 comments

I have manually downloaded the model weights from Hugging Face and tried fine-tuning the model using the following command. bash scripts/vl_finetune.sh ./groma-7b-pretrain ./train_history/ But the program still tries to access the weights from the website. I set the following output for debug. image image And this is the output: image It seems that the program is unable to local model weight?

Looking forward to your reply.

liukc19 avatar Apr 30 '24 18:04 liukc19

Sorry, the issue seems to be occurring here. image image So what is the vis_encoder_path

liukc19 avatar May 01 '24 06:05 liukc19

Thanks for the feedback. The bug occurs as the program is looking for a local DINOv2-L checkpoint to initialize CustomDDETRModel. This is not an expected behavior. A quick fix is to delete vis_encoder_path: checkpoints/dinov2-large in groma-7b-pretrain/config.json. I will fix the initialization logic soon later.

machuofan avatar May 01 '24 06:05 machuofan

"Thank you for your prompt feedback. I have encountered a new issue. Do you know why this is happening?" image

liukc19 avatar May 01 '24 08:05 liukc19

I think this error comes from a failed installation of mmcv. Could you clarify the relationship between the mmcv folder in your repository and the mmcv package?

liukc19 avatar May 01 '24 12:05 liukc19

We inherited the mmcv folder from GPT4ROI. I think it is originated from mmcv==1.4.7.

machuofan avatar May 02 '24 07:05 machuofan