LucasZhao

[Question] train lightning llava-7b-v1 got relatively higher loss curve.

1

### Question Here is my training script: ```shell torchrun --nnodes=1 --nproc_per_node=8 --master_port=25001 \ llava/train/train_mem.py \ --model_name_or_path llama-vicuna-7b-v1.1 \ --version v1 \ --data_path CC3M/chat.json \ --image_folder CC3M/images \ --vision_tower openai/clip-vit-large-patch14 \...

ChartQA augmented & CMMMU

1

It would be nice if VLMEvalKit can support evaluation on ChartQA augmented set and CMMMU, since it has supported ChartQA human set and MMMU

How to test the model in DUE-benchmark?

1

The DUE-benchmark provides the ocr results of pdf-type documnets and other models use the ocr results as input to eval their model. What are the input of your model when...