Fine-tune multimodal models for vision and text use cases on Amazon SageMaker JumpStart
AWS Machine Learning Blog
NOVEMBER 15, 2024
In the metadata.jsonl file, each example is a dictionary that contains three keys named file_name , prompt , and completion. prompt defines the text input prompt and completion defines the text completion corresponding to the input prompt. jpg", "prompt": "what is the contact person name mentioned in letter?", "completion": "P.
Let's personalize your content