32 lines
938 B
JSON
32 lines
938 B
JSON
|
{
|
||
|
|
"document": "../../../test_documents/pdf/ocr_test_rotated_180.pdf",
|
||
|
|
"file_type": "pdf",
|
||
|
|
"file_size": 94703,
|
||
|
|
"expected_frameworks": [
|
||
|
|
"kreuzberg",
|
||
|
|
"docling",
|
||
|
|
"markitdown",
|
||
|
|
"mineru",
|
||
|
|
"pdfminer",
|
||
|
|
"pdfplumber",
|
||
|
|
"pdftotext",
|
||
|
|
"pymupdf4llm",
|
||
|
|
"pypdf",
|
||
|
|
"tika",
|
||
|
|
"unstructured"
|
||
|
|
],
|
||
|
|
"metadata": {
|
||
|
|
"description": "Rotated OCR test PDF - 92KB scanned document with rotation to test OCR engine handling of page orientation",
|
||
|
|
"category": "pdf-ocr",
|
||
|
|
"size_class": "small",
|
||
|
|
"requires_ocr": true,
|
||
|
|
"special_handling": "rotated_pages"
|
||
|
|
},
|
||
|
|
"ground_truth": {
|
||
|
|
"text_file": "../../../test_documents/ground_truth/pdf/pdf_ocr_rotated_180.md",
|
||
|
|
"source": "vision",
|
||
|
|
"markdown_file": "../../../test_documents/ground_truth/pdf/pdf_ocr_rotated_180.md"
|
||
|
|
},
|
||
|
|
"ground_truth_text": "Docling bundles PDF document conversion to JSON and Markdown in an easy self contained package"
|
||
|
|
}
|