Hi, thank you for sharing your work. I am using single_turn_eval.py to do chart summarization tasks. But my model keeps got killed at initializing the model ' model = MetaModel(args.llama_type, args.llama_config, args.tokenizer_path, with_visual=True)'. My GPU has 64GB Ram. Would it be possible to change to a smaller model? Can you give some hints?
Hi, thank you for sharing your work. I am using single_turn_eval.py to do chart summarization tasks. But my model keeps got killed at initializing the model ' model = MetaModel(args.llama_type, args.llama_config, args.tokenizer_path, with_visual=True)'. My GPU has 64GB Ram. Would it be possible to change to a smaller model? Can you give some hints?
Thank you