mllife

Results 38 comments of mllife

Leverage word bbox from pdf-parser-v2 in the layout- and table-model

@PeterStaar-IBM , any update to https://github.com/DS4SD/docling-ibm-models for table-parsing to utilise this?

Leverage word bbox from pdf-parser-v2 in the layout- and table-model

@PeterStaar-IBM , I am using your table model and integrated to my code based on the tests which uses iocr.parse format https://github.com/DS4SD/docling-ibm-models/blob/main/tests/test_data/samples/ADS.2007.page_123.png_iocr.parse_format.json ; what are changes required for me? It...

docling identified my entire page as a picture

This is a scanned document. You should use OCR argument to parse it.

how to approach model distillation, for creating a smaller + faster model

Hello, @dimitri009 ; You can train your own custom object detection model that is faster rt-detr or the newer yolo v11 https://docs.ultralytics.com/models/rtdetr/ ; Personally, I have now moved to TableFormer...

Using Docling with costume layout and table recognition models

@PeterStaar-IBM , have you benchmarked docling_v2 on the dp-bench? How good it is on tables?

Using Docling with costume layout and table recognition models

@PeterStaar-IBM , I found this - https://huggingface.co/ds4sd/docling-models#tableformer is this the same? Model (TEDS) | Simple table | Complex table | All tables -- | -- | -- | -- Tabula...

Using Docling with costume layout and table recognition models

Hello, @PeterStaar-IBM , I see the the OTSL paper have some numbers (95.5) but they are on PubLayNet tables, which tables do dp-bench uses? Will you share the new results...

Using Docling with costume layout and table recognition models

@PeterStaar-IBM , any update to this?