Skip to content

Latest commit

 

History

History
11 lines (6 loc) · 460 Bytes

File metadata and controls

11 lines (6 loc) · 460 Bytes

data process

DVQA_option.py --Construct a multiple-choice SFT fine-tuning dataset based on the OCR information from the DVQA dataset.

draw_radarmap.py --Plot a radar chart comparing the performance of large models.

gpt_api.py --General code framework for calling the GPT API.

sam_batchinfer.py --Batch inference of multiple images using the Segment Anything Model for dataset creation.

tokenize.py --Extract nouns from the sentence.