{"id":5694,"date":"2024-09-03T17:20:48","date_gmt":"2024-09-03T09:20:48","guid":{"rendered":"https:\/\/www.aisharenet.com\/?p=5694"},"modified":"2025-08-25T01:30:04","modified_gmt":"2025-08-24T17:30:04","slug":"marker","status":"publish","type":"post","link":"https:\/\/www.kdjingpai.com\/en\/marker\/","title":{"rendered":"Marker\uff1a\u5feb\u901f\u5c06PDF\u8f6c\u6362\u4e3aMarkdown\u7684\u5f00\u6e90\u5de5\u5177"},"content":{"rendered":"<p>Marker \u662f\u4e00\u4e2a\u57fa\u4e8e\u6df1\u5ea6\u5b66\u4e60\u7684\u6587\u6863\u5904\u7406\u5de5\u5177\uff0c\u65e8\u5728\u5c06 PDF \u6587\u4ef6\u5feb\u901f\u51c6\u786e\u5730\u8f6c\u6362\u4e3a Markdown \u683c\u5f0f\u3002\u5b83\u652f\u6301\u591a\u79cd\u6587\u6863\u7c7b\u578b\uff0c\u7279\u522b\u4f18\u5316\u4e86\u4e66\u7c4d\u548c\u79d1\u5b66\u8bba\u6587\u7684\u8f6c\u6362\u3002Marker \u80fd\u591f\u53bb\u9664\u9875\u7709\u9875\u811a\u7b49\u591a\u4f59\u5185\u5bb9\uff0c\u683c\u5f0f\u5316\u8868\u683c\u548c\u4ee3\u7801\u5757\uff0c\u5e76\u63d0\u53d6\u548c\u4fdd\u5b58\u56fe\u50cf\u3002\u5b83\u8fd8\u53ef\u4ee5\u5c06\u5927\u90e8\u5206\u516c\u5f0f\u8f6c\u6362\u4e3a LaTeX \u683c\u5f0f\uff0c\u652f\u6301\u5728 GPU\u3001CPU \u6216 MPS \u4e0a\u8fd0\u884c\u3002<\/p>\n<p>&nbsp;<\/p>\n<p><img decoding=\"async\" title=\"Marker\uff1a\u5feb\u901f\u5c06PDF\u8f6c\u6362\u4e3aMarkdown\u7684\u5f00\u6e90\u5de5\u5177-1\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2024\/09\/a06a04875c870ba.jpg\" alt=\"Marker\uff1a\u5feb\u901f\u5c06PDF\u8f6c\u6362\u4e3aMarkdown\u7684\u5f00\u6e90\u5de5\u5177-1\" \/><\/p>\n<p>&nbsp;<\/p>\n<h2>\u529f\u80fd\u5217\u8868<\/h2>\n<ul>\n<li>\u5c06 PDF \u6587\u4ef6\u8f6c\u6362\u4e3a Markdown \u683c\u5f0f<\/li>\n<li>\u652f\u6301\u591a\u79cd\u6587\u6863\u7c7b\u578b\uff0c\u5305\u62ec\u4e66\u7c4d\u548c\u79d1\u5b66\u8bba\u6587<\/li>\n<li>\u53bb\u9664\u9875\u7709\u9875\u811a\u7b49\u591a\u4f59\u5185\u5bb9<\/li>\n<li>\u683c\u5f0f\u5316\u8868\u683c\u548c\u4ee3\u7801\u5757<\/li>\n<li>\u63d0\u53d6\u5e76\u4fdd\u5b58\u56fe\u50cf<\/li>\n<li>\u5c06\u5927\u90e8\u5206\u65b9\u7a0b\u5f0f\u8f6c\u6362\u4e3a LaTeX \u683c\u5f0f<\/li>\n<li>\u652f\u6301 GPU\u3001CPU \u548c MPS \u8fd0\u884c<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<p>&nbsp;<\/p>\n<h2>\u4f7f\u7528\u5e2e\u52a9<\/h2>\n<h3>\u5b89\u88c5\u6d41\u7a0b<\/h3>\n<ol>\n<li><strong>\u5b89\u88c5\u4f9d\u8d56<\/strong>\uff1a\u786e\u4fdd\u5df2\u5b89\u88c5 Python 3.6 \u53ca\u4ee5\u4e0a\u7248\u672c\uff0c\u5e76\u5b89\u88c5\u4ee5\u4e0b\u4f9d\u8d56\uff1a\n<pre><code>pip install marker-pdf\r\n<\/code><\/pre>\n<\/li>\n<li><strong>\u8fd0\u884c\u793a\u4f8b<\/strong>\uff1a\n<pre><code>marker_single \/path\/to\/file.pdf \/path\/to\/output\/folder --batch_multiplier 2 --max_pages 10\r\n<\/code><\/pre>\n<\/li>\n<\/ol>\n<p>&nbsp;<\/p>\n<h3>\u4f7f\u7528\u6307\u5357<\/h3>\n<p>&nbsp;<\/p>\n<h4>\u8f6c\u6362\u5355\u4e2a\u6587\u4ef6<\/h4>\n<pre>marker_single \/path\/to\/file.pdf \/path\/to\/output\/folder --batch_multiplier 2 --max_pages 10<\/pre>\n<ul>\n<li><code>--batch_multiplier<\/code>\u00a0\u662f\u5728\u62e5\u6709\u989d\u5916 VRAM \u7684\u60c5\u51b5\u4e0b\uff0c\u9ed8\u8ba4\u6279\u5904\u7406\u5927\u5c0f\u7684\u500d\u6570\u3002\u66f4\u9ad8\u7684\u6570\u5b57\u5c06\u5360\u7528\u66f4\u591a\u7684 VRAM\uff0c\u4f46\u5904\u7406\u901f\u5ea6\u66f4\u5feb\u3002\u9ed8\u8ba4\u8bbe\u7f6e\u4e3a 2\u3002\u9ed8\u8ba4\u6279\u5904\u7406\u5927\u5c0f\u5927\u7ea6\u9700\u8981 3GB \u7684 VRAM\u3002<\/li>\n<li><code>--max_pages<\/code>\u00a0\u662f\u8981\u5904\u7406\u7684\u6700\u5927\u9875\u9762\u6570\u3002\u7701\u7565\u6b64\u9879\u5c06\u8f6c\u6362\u6574\u4e2a\u6587\u6863\u3002<\/li>\n<li><code>--langs<\/code>\u00a0\u662f\u53ef\u9009\u7684\u4ee5\u9017\u53f7\u5206\u9694\u7684\u6587\u6863\u8bed\u8a00\u5217\u8868\uff0c\u7528\u4e8e OCR\u3002\u9ed8\u8ba4\u662f\u53ef\u9009\u7684\uff0c\u5982\u679c\u4f7f\u7528 tesseract\uff0c\u5219\u9700\u8981\u63d0\u4f9b\u3002<\/li>\n<li><code>--ocr_all_pages<\/code>\u00a0\u662f\u4e00\u4e2a\u53ef\u9009\u53c2\u6570\uff0c\u7528\u4e8e\u5f3a\u5236\u5bf9 PDF \u7684\u6240\u6709\u9875\u9762\u8fdb\u884c OCR\u3002\u5982\u679c\u6b64\u53c2\u6570\u6216\u73af\u5883\u53d8\u91cf `OCR_ALL_PAGES` \u4e3a\u771f\uff0c\u5c06\u5f3a\u5236\u8fdb\u884c OCR\u3002<\/li>\n<\/ul>\n<p>\u652f\u6301\u7684 Surya OCR \u8bed\u8a00\u5217\u8868\u53ef\u4ee5\u5728 [<a href=\"https:\/\/github.com\/VikParuchuri\/surya\/blob\/master\/surya\/languages.py\">\u8fd9\u91cc<\/a>] \u627e\u5230\u3002\u5982\u679c\u60a8\u9700\u8981\u66f4\u591a\u8bed\u8a00\uff0c\u53ef\u4ee5\u4f7f\u7528\u4efb\u4f55\u652f\u6301\u7684\u8bed\u8a00\uff0c\u53ea\u9700\u5c06 <code>OCR_ENGINE<\/code>\u00a0\u8bbe\u7f6e\u4e3a <code>ocrmypdf<\/code>\u3002\u5982\u679c\u4e0d\u9700\u8981 OCR\uff0cmarker \u53ef\u4ee5\u652f\u6301\u4efb\u4f55\u8bed\u8a00\u3002<\/p>\n<p>&nbsp;<\/p>\n<h4>\u8f6c\u6362\u591a\u4e2a\u6587\u4ef6<\/h4>\n<pre>marker \/path\/to\/input\/folder \/path\/to\/output\/folder --workers 4 --max 10 --min_length 10000<\/pre>\n<ul>\n<li><code>--workers<\/code>\u00a0\u662f\u540c\u65f6\u8f6c\u6362\u7684 PDF \u6570\u91cf\u3002\u9ed8\u8ba4\u8bbe\u7f6e\u4e3a 1\uff0c\u4f46\u60a8\u53ef\u4ee5\u589e\u52a0\u6b64\u503c\u4ee5\u63d0\u9ad8\u541e\u5410\u91cf\uff0c\u4ee3\u4ef7\u662f\u589e\u52a0 CPU\/GPU \u7684\u4f7f\u7528\u7387\u3002\u6bcf\u4e2a\u5de5\u4f5c\u8fdb\u7a0b\u5728\u5cf0\u503c\u65f6\u5c06\u4f7f\u7528 5GB \u7684 VRAM\uff0c\u5e73\u5747\u4f7f\u7528 3.5GB\u3002<\/li>\n<li><code>--max<\/code>\u00a0\u662f\u8981\u8f6c\u6362\u7684 PDF \u6700\u5927\u6570\u91cf\u3002\u7701\u7565\u6b64\u9879\u5c06\u8f6c\u6362\u6587\u4ef6\u5939\u4e2d\u7684\u6240\u6709 PDF\u3002<\/li>\n<li><code>--min_length<\/code>\u00a0\u662f\u5728 PDF \u4e2d\u63d0\u53d6\u7684\u5b57\u7b26\u6570\u7684\u6700\u5c0f\u503c\uff0c\u53ea\u6709\u8d85\u8fc7\u8fd9\u4e2a\u503c\u7684 PDF \u624d\u4f1a\u88ab\u8003\u8651\u8fdb\u884c\u5904\u7406\u3002\u5982\u679c\u60a8\u8981\u5904\u7406\u5f88\u591a PDF\uff0c\u5efa\u8bae\u8bbe\u7f6e\u6b64\u503c\uff0c\u4ee5\u907f\u514d\u5bf9\u4e3b\u8981\u662f\u56fe\u50cf\u7684 PDF \u8fdb\u884c OCR\uff08\u8fd9\u4f1a\u51cf\u6162\u5904\u7406\u901f\u5ea6\uff09\u3002<\/li>\n<li><code>--metadata_file<\/code>\u00a0\u662f\u4e00\u4e2a\u53ef\u9009\u7684 JSON \u6587\u4ef6\u8def\u5f84\uff0c\u5305\u542b\u5173\u4e8e PDF \u7684\u5143\u6570\u636e\u3002\u5982\u679c\u63d0\u4f9b\u8be5\u6587\u4ef6\uff0c\u5c06\u7528\u4e8e\u4e3a\u6bcf\u4e2a PDF \u8bbe\u7f6e\u8bed\u8a00\u3002\u5bf9\u4e8e Surya\uff08\u9ed8\u8ba4\uff09\uff0c\u8bbe\u7f6e\u8bed\u8a00\u662f\u53ef\u9009\u7684\uff0c\u4f46\u5bf9\u4e8e Tesseract \u662f\u5fc5\u9700\u7684\u3002\u683c\u5f0f\u5982\u4e0b\uff1a<\/li>\n<\/ul>\n<pre>{\r\n\"pdf1.pdf\": {\"languages\": [\"English\"]},\r\n\"pdf2.pdf\": {\"languages\": [\"Spanish\", \"Russian\"]},\r\n...\r\n}<\/pre>\n<p>\u60a8\u53ef\u4ee5\u4f7f\u7528\u8bed\u8a00\u540d\u79f0\u6216\u4ee3\u7801\u3002\u5177\u4f53\u4ee3\u7801\u53d6\u51b3\u4e8e OCR \u5f15\u64ce\u3002\u6709\u5173 Surya \u4ee3\u7801\u7684\u5b8c\u6574\u5217\u8868\uff0c\u8bf7\u53c2\u89c1 [<a href=\"https:\/\/github.com\/VikParuchuri\/surya\/blob\/master\/surya\/languages.py\">\u8fd9\u91cc<\/a>]\uff0c\u6709\u5173 Tesseract \u7684\u8bf7\u89c1 [<a href=\"https:\/\/tesseract-ocr.github.io\/tessdoc\/Data-Files#data-files-for-version-400-november-29-2016\">\u8fd9\u91cc<\/a>]\u3002<\/p>\n<p>&nbsp;<\/p>\n<h2>\u5728FastGPT \u4e2d\u914d\u7f6eMarker\u73af\u5883\u53d8\u91cf<\/h2>\n<p>\u8981\u542f\u7528\u81ea\u5b9a\u4e49\u89e3\u6790\u670d\u52a1\uff0c\u9700\u8981\u5728 FastGPT \u4e2d\u914d\u7f6e\u4ee5\u4e0b\u73af\u5883\u53d8\u91cf\uff1a<\/p>\n<p>CUSTOM_READ_FILE_URL=http:\/\/xxxx.com\/v1\/parse\/file<br \/>\nCUSTOM_READ_FILE_EXTENSION=pdf<\/p>\n<ul>\n<li>CUSTOM_READ_FILE_URL &#8211; \u81ea\u5b9a\u4e49\u89e3\u6790\u670d\u52a1\u7684\u8bbf\u95ee\u5730\u5740\uff0c\u9700\u8981\u5c06 host \u4fee\u6539\u4e3a\u4f60\u90e8\u7f72\u7684\u89e3\u6790\u670d\u52a1\u5730\u5740\uff0cpath \u8def\u5f84\u4fdd\u6301\u4e0d\u53d8<\/li>\n<li>CUSTOM_READ_FILE_EXTENSION &#8211; \u6307\u5b9a\u652f\u6301\u89e3\u6790\u7684\u6587\u4ef6\u7c7b\u578b\u540e\u7f00\uff0c\u591a\u4e2a\u6587\u4ef6\u7c7b\u578b\u4e4b\u95f4\u7528\u9017\u53f7\u5206\u9694<\/li>\n<\/ul>\n<p><strong>\u9a8c\u8bc1\u89e3\u6790\u6548\u679c<\/strong><\/p>\n<p>\u5b8c\u6210\u914d\u7f6e\u540e\uff0c\u4f60\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u6b65\u9aa4\u9a8c\u8bc1\u89e3\u6790\u6548\u679c\uff1a<\/p>\n<ol>\n<li>\u5728\u77e5\u8bc6\u5e93\u4e2d\u4e0a\u4f20\u4e00\u4e2a PDF \u6587\u4ef6\u5e76\u786e\u8ba4\u4e0a\u4f20<\/li>\n<li>\u67e5\u770b\u7cfb\u7edf\u65e5\u5fd7(\u9700\u8981\u5c06 LOG_LEVEL \u8bbe\u7f6e\u4e3a info \u6216 debug \u7ea7\u522b)<\/li>\n<li>\u4f60\u4f1a\u53d1\u73b0\u901a\u8fc7 Marker \u89e3\u6790\u7684 PDF \u6587\u4ef6\u4e2d\u5305\u542b\u4e86\u5b8c\u6574\u7684\u56fe\u7247\u94fe\u63a5\uff0c\u8fd9\u8868\u660e\u89e3\u6790\u6210\u529f<\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"<p>Marker \u662f\u4e00\u4e2a\u57fa\u4e8e\u6df1\u5ea6\u5b66\u4e60\u7684\u6587\u6863\u5904\u7406\u5de5\u5177\uff0c\u65e8\u5728\u5c06 PDF \u6587\u4ef6\u5feb\u901f\u51c6\u786e\u5730\u8f6c\u6362\u4e3a Markdown \u683c\u5f0f\u3002\u5b83\u652f\u6301\u591a\u79cd\u6587\u6863\u7c7b\u578b\uff0c\u7279\u522b\u4f18\u5316\u4e86\u4e66\u7c4d\u548c\u79d1\u5b66\u8bba\u6587\u7684\u8f6c\u6362\u3002Marker \u80fd\u591f\u53bb\u9664\u9875\u7709\u9875\u811a\u7b49\u591a\u4f59\u5185\u5bb9\uff0c\u683c\u5f0f\u5316\u8868\u683c\u548c\u4ee3\u7801\u5757\uff0c\u5e76\u63d0\u53d6\u548c\u4fdd\u5b58\u56fe\u50cf&#8230;<\/p>\n","protected":false},"author":1,"featured_media":32782,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20,499],"tags":[230,252],"class_list":["post-5694","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-tool","category-document-extraction","tag-aikaiyuanxiangmu","tag-markdown"],"_links":{"self":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts\/5694","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/comments?post=5694"}],"version-history":[{"count":0,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts\/5694\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/media\/32782"}],"wp:attachment":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/media?parent=5694"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/categories?post=5694"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/tags?post=5694"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}