llava-hf

/

llava-v1.6-mistral-7b-hf

like 231

Follow

Llava Hugging Face 111

Yes it is! See https://huggingface.co/docs/transformers/main/en/model_doc/llava_next#multi-image-inference

@nielsr is this the correct promt for few shot training? I am providing the required images to processor

prompt = [
"[INST] \nWhat can you infer from the ecg? [/INST] The ECG shows 13 beats with a maximum amplitude of 0.6992553483687857. Beats Description: [Normal beat, Normal beat, Normal beat, Normal beat, Normal beat, Normal beat, Normal beat, Normal beat, Normal beat, Normal beat, Normal beat, Normal beat, Normal beat] [/INST]",
"[INST] \nWhat can you infer from the ecg? [/INST] The ECG shows 15 beats with a maximum amplitude of 0.20170827356791896. Beats Description: [Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat] [/INST]",
"[INST] \nWhat can you infer from the ecg? [/INST] The ECG shows 7 beats with a maximum amplitude of 0.9278580584124272. Beats Description: [Normal beat, Normal beat, Normal beat, Normal beat, Normal beat, Normal beat, Normal beat] [/INST]",
"[INST] \nWhat can you infer from the ecg? [/INST] The ECG shows 14 beats with a maximum amplitude of 0.4437582018494217. Beats Description: [Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat, Unclassifiable beat] [/INST]"
]