{"id":18256,"date":"2025-01-14T10:29:06","date_gmt":"2025-01-14T02:29:06","guid":{"rendered":"https:\/\/www.aisharenet.com\/?p=18256"},"modified":"2025-08-25T00:39:52","modified_gmt":"2025-08-24T16:39:52","slug":"nv-ingest","status":"publish","type":"post","link":"https:\/\/www.kdjingpai.com\/pt\/nv-ingest\/","title":{"rendered":"NV Ingest\uff1a\u89e3\u6790\u590d\u6742\u683c\u5f0f\u6587\u6863\uff0c\u63d0\u53d6\u591a\u6a21\u6001\u6570\u636e\u4e3a\u5143\u6570\u636e\u548c\u6587\u672c"},"content":{"rendered":"<p>NV Ingest\uff08NVIDIA Ingest\uff09 \u662f\u4e00\u5957\u65e9\u671f\u8bbf\u95ee\u7684\u5fae\u670d\u52a1\uff0c\u4e13\u4e3a\u89e3\u6790\u6570\u5341\u4e07\u590d\u6742\u3001\u6df7\u4e71\u7684\u975e\u7ed3\u6784\u5316 PDF \u548c\u5176\u4ed6\u4f01\u4e1a\u6587\u6863\u800c\u8bbe\u8ba1\u3002\u5b83\u80fd\u591f\u5c06\u8fd9\u4e9b\u6587\u6863\u8f6c\u6362\u4e3a\u5143\u6570\u636e\u548c\u6587\u672c\uff0c\u4ee5\u4fbf\u5d4c\u5165\u5230\u68c0\u7d22\u7cfb\u7edf\u4e2d\u3002NVIDIA Ingest \u652f\u6301\u89e3\u6790 PDF\u3001Word \u548c PowerPoint \u6587\u6863\uff0c\u5229\u7528 NVIDIA NIM \u5fae\u670d\u52a1\u6765\u67e5\u627e\u3001\u4e0a\u4e0b\u6587\u5316\u548c\u63d0\u53d6\u6587\u672c\u3001\u8868\u683c\u3001\u56fe\u8868\u548c\u56fe\u50cf\uff0c\u4ee5\u4f9b\u4e0b\u6e38\u751f\u6210\u5e94\u7528\u4f7f\u7528\u3002\u8be5\u670d\u52a1\u80fd\u591f\u5e76\u884c\u5316\u5904\u7406\uff0c\u5c06\u6587\u6863\u62c6\u5206\u4e3a\u9875\u9762\uff0c\u5206\u7c7b\u5185\u5bb9\uff08\u5982\u8868\u683c\u3001\u56fe\u8868\u3001\u56fe\u50cf\u3001\u6587\u672c\uff09\uff0c\u5e76\u901a\u8fc7\u5149\u5b66\u5b57\u7b26\u8bc6\u522b\uff08OCR\uff09\u5c06\u5176\u63d0\u53d6\u4e3a\u5b9a\u4e49\u826f\u597d\u7684 JSON \u67b6\u6784\u3002NVIDIA Ingest \u8fd8\u53ef\u4ee5\u9009\u62e9\u6027\u5730\u7ba1\u7406\u5d4c\u5165\u5185\u5bb9\u7684\u8ba1\u7b97\uff0c\u5e76\u5c06\u5176\u5b58\u50a8\u5230\u5411\u91cf\u6570\u636e\u5e93 Milvus \u4e2d\u3002<\/p>\n<div id=\"attachment_18257\" style=\"width: 749px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" aria-describedby=\"caption-attachment-18257\" class=\"wp-image-18257\" title=\"NV Ingest\uff1a\u89e3\u6790\u590d\u6742\u683c\u5f0f\u6587\u6863\uff0c\u63d0\u53d6\u591a\u6a21\u6001\u6570\u636e\u4e3a\u5143\u6570\u636e\u548c\u6587\u672c-1\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/2ef0c6848d9a1fa.jpg\" alt=\"NV Ingest\uff1a\u89e3\u6790\u590d\u6742\u683c\u5f0f\u6587\u6863\uff0c\u63d0\u53d6\u591a\u6a21\u6001\u6570\u636e\u4e3a\u5143\u6570\u636e\u548c\u6587\u672c-1\" width=\"739\" height=\"416\" srcset=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/2ef0c6848d9a1fa.jpg 1000w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/2ef0c6848d9a1fa-300x169.jpg 300w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/2ef0c6848d9a1fa-768x432.jpg 768w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/01\/2ef0c6848d9a1fa-18x10.jpg 18w\" sizes=\"auto, (max-width: 739px) 100vw, 739px\" \/><p id=\"caption-attachment-18257\" class=\"wp-caption-text\">\u5e2e\u52a9\u6587\u6863\uff1ahttps:\/\/docs.nvidia.com\/nv-ingest\/<\/p><\/div>\n<p>&nbsp;<\/p>\n<h2>\u529f\u80fd\u5217\u8868<\/h2>\n<ul>\n<li>\u652f\u6301\u89e3\u6790 PDF\u3001Word \u548c PowerPoint \u6587\u6863<\/li>\n<li>\u5229\u7528 NVIDIA NIM \u5fae\u670d\u52a1\u67e5\u627e\u3001\u4e0a\u4e0b\u6587\u5316\u548c\u63d0\u53d6\u6587\u672c\u3001\u8868\u683c\u3001\u56fe\u8868\u548c\u56fe\u50cf<\/li>\n<li>\u5e76\u884c\u5316\u5904\u7406\u6587\u6863\uff0c\u5c06\u5176\u62c6\u5206\u4e3a\u9875\u9762\u5e76\u5206\u7c7b\u5185\u5bb9<\/li>\n<li>\u901a\u8fc7 OCR \u63d0\u53d6\u5185\u5bb9\u5e76\u8f6c\u6362\u4e3a JSON \u67b6\u6784<\/li>\n<li>\u652f\u6301\u591a\u79cd\u6587\u6863\u7c7b\u578b\u7684\u63d0\u53d6\u65b9\u6cd5\uff0c\u4ee5\u5e73\u8861\u541e\u5410\u91cf\u548c\u51c6\u786e\u6027<\/li>\n<li>\u652f\u6301\u5404\u79cd\u9884\u5904\u7406\u548c\u540e\u5904\u7406\u64cd\u4f5c\uff0c\u5305\u62ec\u6587\u672c\u62c6\u5206\u548c\u5206\u5757\u3001\u8f6c\u6362\u548c\u8fc7\u6ee4\u3001\u5d4c\u5165\u751f\u6210\u548c\u56fe\u50cf\u5378\u8f7d\u5230\u5b58\u50a8<\/li>\n<li>\u53ef\u9009\u62e9\u6027\u5730\u7ba1\u7406\u5d4c\u5165\u5185\u5bb9\u7684\u8ba1\u7b97\u548c\u5b58\u50a8\u5230\u5411\u91cf\u6570\u636e\u5e93 Milvus \u4e2d<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2>\u4f7f\u7528\u5e2e\u52a9<\/h2>\n<h3>\u5b89\u88c5\u6d41\u7a0b<\/h3>\n<ol>\n<li>\u514b\u9686 NVIDIA Ingest \u4ed3\u5e93\uff1a<\/li>\n<\/ol>\n<pre><code>   git clone https:\/\/github.com\/NVIDIA\/nv-ingest.git\r\n<\/code><\/pre>\n<ol start=\"2\">\n<li>\u8fdb\u5165\u9879\u76ee\u76ee\u5f55\uff1a<\/li>\n<\/ol>\n<pre><code>   cd nv-ingest\r\n<\/code><\/pre>\n<ol start=\"3\">\n<li>\u5b89\u88c5\u4f9d\u8d56\u9879\uff1a<\/li>\n<\/ol>\n<pre><code>   pip install -r requirements.txt\r\n<\/code><\/pre>\n<ol start=\"4\">\n<li>\u914d\u7f6e\u73af\u5883\u53d8\u91cf\uff1a<\/li>\n<\/ol>\n<pre><code>   source setup_env.sh\r\n<\/code><\/pre>\n<ol start=\"5\">\n<li>\u542f\u52a8\u670d\u52a1\uff1a<\/li>\n<\/ol>\n<pre><code>   docker-compose up\r\n<\/code><\/pre>\n<h3>\u4f7f\u7528\u6d41\u7a0b<\/h3>\n<ol>\n<li><strong>\u63d0\u4ea4\u6587\u6863\u89e3\u6790\u4efb\u52a1<\/strong>\uff1a\n<ul>\n<li>\u901a\u8fc7 API \u63d0\u4ea4\u5305\u542b\u6587\u6863\u8d1f\u8f7d\u548c\u89e3\u6790\u4efb\u52a1\u7684 JSON \u4f5c\u4e1a\u63cf\u8ff0\u3002<\/li>\n<li>\u793a\u4f8b JSON \u4f5c\u4e1a\u63cf\u8ff0\uff1a<\/li>\n<\/ul>\n<pre><code> {\r\n\"document_payload\": \"base64_encoded_document\",\r\n\"ingestion_tasks\": [\"parse_text\", \"extract_metadata\"]\r\n}\r\n<\/code><\/pre>\n<\/li>\n<li><strong>\u68c0\u7d22\u89e3\u6790\u7ed3\u679c<\/strong>\uff1a\n<ul>\n<li>\u901a\u8fc7 API \u68c0\u7d22\u4f5c\u4e1a\u7ed3\u679c\uff0c\u7ed3\u679c\u4e3a\u5305\u542b\u63d0\u53d6\u5bf9\u8c61\u5143\u6570\u636e\u3001\u5904\u7406\u6ce8\u91ca\u548c\u65f6\u95f4\/\u8ddf\u8e2a\u6570\u636e\u7684 JSON \u5b57\u5178\u3002<\/li>\n<li>\u793a\u4f8b API \u8c03\u7528\uff1a<\/li>\n<\/ul>\n<pre><code> curl -X GET \"http:\/\/localhost:5000\/api\/results\/{job_id}\"\r\n<\/code><\/pre>\n<\/li>\n<li><strong>\u652f\u6301\u7684\u6587\u6863\u7c7b\u578b\u548c\u63d0\u53d6\u65b9\u6cd5<\/strong>\uff1a\n<ul>\n<li>PDF \u6587\u6863\uff1a\u652f\u6301\u901a\u8fc7 pdfium\u3001Unstructured.io \u548c Adobe Content Extraction Services \u8fdb\u884c\u63d0\u53d6\u3002<\/li>\n<li>Word \u6587\u6863\uff1a\u652f\u6301\u901a\u8fc7 Microsoft Office API \u8fdb\u884c\u63d0\u53d6\u3002<\/li>\n<li>PowerPoint \u6587\u6863\uff1a\u652f\u6301\u901a\u8fc7 Microsoft Office API \u8fdb\u884c\u63d0\u53d6\u3002<\/li>\n<li>\u56fe\u50cf\uff1a\u652f\u6301\u901a\u8fc7 OCR \u8fdb\u884c\u63d0\u53d6\u3002<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u9884\u5904\u7406\u548c\u540e\u5904\u7406\u64cd\u4f5c<\/strong>\uff1a\n<ul>\n<li>\u6587\u672c\u62c6\u5206\u548c\u5206\u5757\uff1a\u5c06\u957f\u6587\u672c\u62c6\u5206\u4e3a\u8f83\u5c0f\u7684\u5757\uff0c\u4ee5\u4fbf\u66f4\u597d\u5730\u5904\u7406\u548c\u5206\u6790\u3002<\/li>\n<li>\u8f6c\u6362\u548c\u8fc7\u6ee4\uff1a\u5bf9\u63d0\u53d6\u7684\u6587\u672c\u8fdb\u884c\u8f6c\u6362\u548c\u8fc7\u6ee4\uff0c\u4ee5\u63d0\u9ad8\u6570\u636e\u8d28\u91cf\u3002<\/li>\n<li>\u5d4c\u5165\u751f\u6210\uff1a\u8ba1\u7b97\u63d0\u53d6\u5185\u5bb9\u7684\u5d4c\u5165\uff0c\u4ee5\u4fbf\u5728\u5411\u91cf\u6570\u636e\u5e93\u4e2d\u5b58\u50a8\u548c\u68c0\u7d22\u3002<\/li>\n<li>\u56fe\u50cf\u5378\u8f7d\u5230\u5b58\u50a8\uff1a\u5c06\u63d0\u53d6\u7684\u56fe\u50cf\u5378\u8f7d\u5230\u5916\u90e8\u5b58\u50a8\uff0c\u4ee5\u4fbf\u8fdb\u4e00\u6b65\u5904\u7406\u548c\u5206\u6790\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n<h3>\u8be6\u7ec6\u64cd\u4f5c\u6d41\u7a0b<\/h3>\n<ol>\n<li><strong>\u63d0\u4ea4\u6587\u6863\u89e3\u6790\u4efb\u52a1<\/strong>\uff1a\n<ul>\n<li>\u901a\u8fc7 API \u63d0\u4ea4\u5305\u542b\u6587\u6863\u8d1f\u8f7d\u548c\u89e3\u6790\u4efb\u52a1\u7684 JSON \u4f5c\u4e1a\u63cf\u8ff0\u3002<\/li>\n<li>\u793a\u4f8b JSON \u4f5c\u4e1a\u63cf\u8ff0\uff1a<\/li>\n<\/ul>\n<pre><code> {\r\n\"document_payload\": \"base64_encoded_document\",\r\n\"ingestion_tasks\": [\"parse_text\", \"extract_metadata\"]\r\n}\r\n<\/code><\/pre>\n<\/li>\n<li><strong>\u68c0\u7d22\u89e3\u6790\u7ed3\u679c<\/strong>\uff1a\n<ul>\n<li>\u901a\u8fc7 API \u68c0\u7d22\u4f5c\u4e1a\u7ed3\u679c\uff0c\u7ed3\u679c\u4e3a\u5305\u542b\u63d0\u53d6\u5bf9\u8c61\u5143\u6570\u636e\u3001\u5904\u7406\u6ce8\u91ca\u548c\u65f6\u95f4\/\u8ddf\u8e2a\u6570\u636e\u7684 JSON \u5b57\u5178\u3002<\/li>\n<li>\u793a\u4f8b API \u8c03\u7528\uff1a<\/li>\n<\/ul>\n<pre><code> curl -X GET \"http:\/\/localhost:5000\/api\/results\/{job_id}\"\r\n<\/code><\/pre>\n<\/li>\n<li><strong>\u652f\u6301\u7684\u6587\u6863\u7c7b\u578b\u548c\u63d0\u53d6\u65b9\u6cd5<\/strong>\uff1a\n<ul>\n<li>PDF \u6587\u6863\uff1a\u652f\u6301\u901a\u8fc7 pdfium\u3001Unstructured.io \u548c Adobe Content Extraction Services \u8fdb\u884c\u63d0\u53d6\u3002<\/li>\n<li>Word \u6587\u6863\uff1a\u652f\u6301\u901a\u8fc7 Microsoft Office API \u8fdb\u884c\u63d0\u53d6\u3002<\/li>\n<li>PowerPoint \u6587\u6863\uff1a\u652f\u6301\u901a\u8fc7 Microsoft Office API \u8fdb\u884c\u63d0\u53d6\u3002<\/li>\n<li>\u56fe\u50cf\uff1a\u652f\u6301\u901a\u8fc7 OCR \u8fdb\u884c\u63d0\u53d6\u3002<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u9884\u5904\u7406\u548c\u540e\u5904\u7406\u64cd\u4f5c<\/strong>\uff1a\n<ul>\n<li>\u6587\u672c\u62c6\u5206\u548c\u5206\u5757\uff1a\u5c06\u957f\u6587\u672c\u62c6\u5206\u4e3a\u8f83\u5c0f\u7684\u5757\uff0c\u4ee5\u4fbf\u66f4\u597d\u5730\u5904\u7406\u548c\u5206\u6790\u3002<\/li>\n<li>\u8f6c\u6362\u548c\u8fc7\u6ee4\uff1a\u5bf9\u63d0\u53d6\u7684\u6587\u672c\u8fdb\u884c\u8f6c\u6362\u548c\u8fc7\u6ee4\uff0c\u4ee5\u63d0\u9ad8\u6570\u636e\u8d28\u91cf\u3002<\/li>\n<li>\u5d4c\u5165\u751f\u6210\uff1a\u8ba1\u7b97\u63d0\u53d6\u5185\u5bb9\u7684\u5d4c\u5165\uff0c\u4ee5\u4fbf\u5728\u5411\u91cf\u6570\u636e\u5e93\u4e2d\u5b58\u50a8\u548c\u68c0\u7d22\u3002<\/li>\n<li>\u56fe\u50cf\u5378\u8f7d\u5230\u5b58\u50a8\uff1a\u5c06\u63d0\u53d6\u7684\u56fe\u50cf\u5378\u8f7d\u5230\u5916\u90e8\u5b58\u50a8\uff0c\u4ee5\u4fbf\u8fdb\u4e00\u6b65\u5904\u7406\u548c\u5206\u6790\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"<p>NV Ingest\uff08NVIDIA Ingest\uff09 \u662f\u4e00\u5957\u65e9\u671f\u8bbf\u95ee\u7684\u5fae\u670d\u52a1\uff0c\u4e13\u4e3a\u89e3\u6790\u6570\u5341\u4e07\u590d\u6742\u3001\u6df7\u4e71\u7684\u975e\u7ed3\u6784\u5316 PDF \u548c\u5176\u4ed6\u4f01\u4e1a\u6587\u6863\u800c\u8bbe\u8ba1\u3002\u5b83\u80fd\u591f\u5c06\u8fd9\u4e9b\u6587\u6863\u8f6c\u6362\u4e3a\u5143\u6570\u636e\u548c\u6587\u672c\uff0c\u4ee5\u4fbf\u5d4c\u5165\u5230\u68c0\u7d22\u7cfb\u7edf\u4e2d\u3002NVIDIA Ingest \u652f\u6301\u89e3\u6790 PD&#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-18256","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\/pt\/wp-json\/wp\/v2\/posts\/18256","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/comments?post=18256"}],"version-history":[{"count":0,"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/posts\/18256\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/media\/32782"}],"wp:attachment":[{"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/media?parent=18256"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/categories?post=18256"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/tags?post=18256"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}