monkt/doctype

Document Classification It works per image input. Classes Class Processing Pipeline Training Datasets document_printed Docling pipeline HuggingFaceFW/finepdfs, RVL-CDIP document_handwritten Docling pipeline with handwriting OCR IAM Handwritten Forms, RVL-CDIP photo VLLM: "Describe this image" COCO 2017 diagram VLLM: "Describe this diagram" AI2D, UML Diagrams chart VLLM: "Analyze this chart/graph" PlotQA screenshot VLLM: "Analyze screenshot"… See the full description on the dataset page: https://huggingface.co/datasets/monkt/doctype.