30 lines
709 B
JSON
30 lines
709 B
JSON
|
{
|
||
|
|
"document": "../../../test_documents/pdf/scanned.pdf",
|
||
|
|
"file_type": "pdf",
|
||
|
|
"file_size": 69966,
|
||
|
|
"expected_frameworks": [
|
||
|
|
"kreuzberg",
|
||
|
|
"docling",
|
||
|
|
"markitdown",
|
||
|
|
"mineru",
|
||
|
|
"pdfminer",
|
||
|
|
"pdfplumber",
|
||
|
|
"pdftotext",
|
||
|
|
"pymupdf4llm",
|
||
|
|
"pypdf",
|
||
|
|
"tika",
|
||
|
|
"unstructured"
|
||
|
|
],
|
||
|
|
"metadata": {
|
||
|
|
"description": "Scanned PDF requiring OCR - 68KB non-searchable document",
|
||
|
|
"category": "pdf-scanned",
|
||
|
|
"size_class": "small",
|
||
|
|
"requires_ocr": true
|
||
|
|
},
|
||
|
|
"ground_truth": {
|
||
|
|
"text_file": "../../../test_documents/ground_truth/pdf/pdf_scanned_ocr.md",
|
||
|
|
"source": "vision",
|
||
|
|
"markdown_file": "../../../test_documents/ground_truth/pdf/pdf_scanned_ocr.md"
|
||
|
|
}
|
||
|
|
}
|