{"id":14908,"date":"2024-12-05T10:53:43","date_gmt":"2024-12-05T02:53:43","guid":{"rendered":"https:\/\/www.aisharenet.com\/?p=14908"},"modified":"2024-12-08T10:58:14","modified_gmt":"2024-12-08T02:58:14","slug":"aigongchengxueyuan23","status":"publish","type":"post","link":"https:\/\/www.kdjingpai.com\/de\/aigongchengxueyuan23\/","title":{"rendered":"AI\u5de5\u7a0b\u5b66\u9662\uff1a2.3BM25 RAG \uff08\u68c0\u7d22\u589e\u5f3a\u751f\u6210\uff09"},"content":{"rendered":"<h2>\u7b80\u4ecb<\/h2>\n<p><a href=\"https:\/\/www.kdjingpai.com\/de\/bm25\/\">BM25<\/a> \u68c0\u7d22\u589e\u5f3a\u751f\u6210\uff08BM25 RAG\uff09\u662f\u4e00\u79cd\u9ad8\u7ea7\u6280\u672f\uff0c\u5c06\u7528\u4e8e\u4fe1\u606f\u68c0\u7d22\u7684 BM25\uff08Best Matching 25\uff09\u7b97\u6cd5\u4e0e\u5927\u8bed\u8a00\u6a21\u578b\u7ed3\u5408\uff0c\u7528\u4e8e\u6587\u672c\u751f\u6210\u3002\u901a\u8fc7\u4f7f\u7528\u7ecf\u8fc7\u9a8c\u8bc1\u7684\u6982\u7387\u68c0\u7d22\u6a21\u578b\uff0c\u6b64\u65b9\u6cd5\u53ef\u63d0\u9ad8\u751f\u6210\u54cd\u5e94\u7684\u51c6\u786e\u6027\u548c\u76f8\u5173\u6027\u3002<\/p>\n<p>&nbsp;<\/p>\n<h2 dir=\"auto\">BM25 RAG \u5de5\u4f5c\u6d41<\/h2>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone  wp-image-14909\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2024\/12\/3d290c71f5ec82d.jpg\" alt=\"\" width=\"663\" height=\"1243\" srcset=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2024\/12\/3d290c71f5ec82d.jpg 1120w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2024\/12\/3d290c71f5ec82d-160x300.jpg 160w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2024\/12\/3d290c71f5ec82d-546x1024.jpg 546w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2024\/12\/3d290c71f5ec82d-768x1440.jpg 768w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2024\/12\/3d290c71f5ec82d-819x1536.jpg 819w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2024\/12\/3d290c71f5ec82d-1092x2048.jpg 1092w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2024\/12\/3d290c71f5ec82d-6x12.jpg 6w\" sizes=\"auto, (max-width: 663px) 100vw, 663px\" \/><\/p>\n<p>&nbsp;<\/p>\n<h2>\u5feb\u901f\u5f00\u59cb<\/h2>\n<h3>Notebook<\/h3>\n<p>\u60a8\u53ef\u4ee5\u8fd0\u884c\u6b64\u4ee3\u7801\u5e93\u4e2d\u63d0\u4f9b\u7684 Jupyter notebook\uff0c\u8be6\u7ec6\u63a2\u7d22 BM25 RAG\u3002https:\/\/github.com\/adithya-s-k\/AI-Engineering.academy\/tree\/main\/RAG\/01_BM25_RAG<\/p>\n<h3>\u804a\u5929\u5e94\u7528\u7a0b\u5e8f<\/h3>\n<ol>\n<li>\u5b89\u88c5\u4f9d\u8d56\u9879\uff1a\n<pre><code>pip install -r requirements.txt\r\n<\/code><\/pre>\n<\/li>\n<li>\u8fd0\u884c\u5e94\u7528\u7a0b\u5e8f\uff1a\n<pre><code>python app.py\r\n<\/code><\/pre>\n<\/li>\n<li>\u52a8\u6001\u6444\u53d6\u6570\u636e\uff1a\n<pre><code>python app.py --ingest --data_dir \/path\/to\/documents\r\n<\/code><\/pre>\n<\/li>\n<\/ol>\n<h3>\u670d\u52a1\u5668<\/h3>\n<p>\u8fd0\u884c\u670d\u52a1\u5668\uff1a<\/p>\n<pre><code>python server.py\r\n<\/code><\/pre>\n<p>\u670d\u52a1\u5668\u5305\u542b\u4e24\u4e2a\u7aef\u70b9\uff1a<\/p>\n<ul>\n<li><code>\/api\/ingest<\/code>\uff1a\u7528\u4e8e\u6444\u53d6\u65b0\u6587\u6863<\/li>\n<li><code>\/api\/query<\/code>\uff1a\u7528\u4e8e\u67e5\u8be2 BM25 <a href=\"https:\/\/www.kdjingpai.com\/de\/rag\/\">RAG<\/a> \u7cfb\u7edf<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2>BM25 RAG \u7684\u5173\u952e\u7279\u6027<\/h2>\n<ol>\n<li><strong>\u6982\u7387\u68c0\u7d22<\/strong>\uff1aBM25 \u4f7f\u7528\u6982\u7387\u6a21\u578b\u5bf9\u6587\u6863\u8fdb\u884c\u6392\u5e8f\uff0c\u4e3a\u68c0\u7d22\u63d0\u4f9b\u4e86\u7406\u8bba\u4e0a\u7684\u53ef\u9760\u57fa\u7840\u3002<\/li>\n<li><strong>\u8bcd\u9891\u9971\u548c<\/strong>\uff1aBM25 \u8003\u8651\u4e86\u91cd\u590d\u8bcd\u7684\u8fb9\u9645\u6536\u76ca\u9012\u51cf\uff0c\u63d0\u9ad8\u4e86\u68c0\u7d22\u8d28\u91cf\u3002<\/li>\n<li><strong>\u6587\u6863\u957f\u5ea6\u5f52\u4e00\u5316<\/strong>\uff1a\u7b97\u6cd5\u8003\u8651\u6587\u6863\u957f\u5ea6\uff0c\u51cf\u5c11\u5bf9\u8f83\u957f\u6587\u6863\u7684\u504f\u5411\u3002<\/li>\n<li><strong>\u4e0a\u4e0b\u6587\u76f8\u5173\u6027<\/strong>\uff1a\u901a\u8fc7\u57fa\u4e8e\u68c0\u7d22\u4fe1\u606f\u751f\u6210\u54cd\u5e94\uff0cBM25 RAG \u63d0\u4f9b\u66f4\u51c6\u786e\u548c\u76f8\u5173\u7684\u56de\u7b54\u3002<\/li>\n<li><strong>\u53ef\u6269\u5c55\u6027<\/strong>\uff1aBM25 \u68c0\u7d22\u6b65\u9aa4\u53ef\u9ad8\u6548\u5904\u7406\u5927\u578b\u6587\u6863\u96c6\u3002<\/li>\n<\/ol>\n<p>&nbsp;<\/p>\n<h2>BM25 RAG \u7684\u4f18\u52bf<\/h2>\n<ol>\n<li><strong>\u63d0\u9ad8\u51c6\u786e\u6027<\/strong>\uff1a\u7ed3\u5408\u6982\u7387\u68c0\u7d22\u4e0e\u795e\u7ecf\u6587\u672c\u751f\u6210\u7684\u4f18\u70b9\u3002<\/li>\n<li><strong>\u53ef\u89e3\u91ca\u6027<\/strong>\uff1aBM25 \u7684\u8bc4\u5206\u673a\u5236\u76f8\u6bd4\u5bc6\u96c6\u5411\u91cf\u68c0\u7d22\u65b9\u6cd5\u66f4\u5177\u53ef\u89e3\u91ca\u6027\u3002<\/li>\n<li><strong>\u5904\u7406\u957f\u5c3e\u67e5\u8be2<\/strong>\uff1a\u5728\u9700\u8981\u7279\u5b9a\u6216\u7a00\u6709\u4fe1\u606f\u7684\u67e5\u8be2\u4e2d\u8868\u73b0\u4f18\u5f02\u3002<\/li>\n<li><strong>\u65e0\u9700\u5d4c\u5165<\/strong>\uff1a\u4e0e\u57fa\u4e8e\u5411\u91cf\u7684 RAG \u4e0d\u540c\uff0cBM25 \u4e0d\u9700\u8981\u6587\u6863\u5d4c\u5165\uff0c\u964d\u4f4e\u4e86\u8ba1\u7b97\u5f00\u9500\u3002<\/li>\n<\/ol>\n<p>&nbsp;<\/p>\n<h2>\u524d\u7f6e\u6761\u4ef6<\/h2>\n<ul>\n<li>Python 3.7+<\/li>\n<li>Jupyter Notebook \u6216 JupyterLab\uff08\u7528\u4e8e\u8fd0\u884c notebook\uff09<\/li>\n<li>\u6240\u9700 Python \u5305\uff08\u89c1\u00a0<code>requirements.txt<\/code>\uff09<\/li>\n<li>\u6240\u9009\u8bed\u8a00\u6a21\u578b\u7684 API \u5bc6\u94a5\uff08\u5982 OpenAI API \u5bc6\u94a5\uff09<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>\u7b80\u4ecb BM25 \u68c0\u7d22\u589e\u5f3a\u751f\u6210\uff08BM25 RAG\uff09\u662f\u4e00\u79cd\u9ad8\u7ea7\u6280\u672f\uff0c\u5c06\u7528\u4e8e\u4fe1\u606f\u68c0\u7d22\u7684 BM25\uff08Best Matching 25\uff09\u7b97\u6cd5\u4e0e\u5927\u8bed\u8a00\u6a21\u578b\u7ed3\u5408\uff0c\u7528\u4e8e\u6587\u672c\u751f\u6210\u3002\u901a\u8fc7\u4f7f\u7528\u7ecf\u8fc7\u9a8c\u8bc1\u7684\u6982\u7387\u68c0\u7d22\u6a21\u578b\uff0c\u6b64\u65b9\u6cd5\u53ef\u63d0\u9ad8\u751f\u6210\u54cd\u5e94\u7684\u51c6\u786e\u6027\u548c\u76f8\u5173\u6027\u3002 &#038;nb&#8230;<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[34],"tags":[],"class_list":["post-14908","post","type-post","status-publish","format-standard","hentry","category-knowledge"],"_links":{"self":[{"href":"https:\/\/www.kdjingpai.com\/de\/wp-json\/wp\/v2\/posts\/14908","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.kdjingpai.com\/de\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.kdjingpai.com\/de\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/de\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/de\/wp-json\/wp\/v2\/comments?post=14908"}],"version-history":[{"count":0,"href":"https:\/\/www.kdjingpai.com\/de\/wp-json\/wp\/v2\/posts\/14908\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.kdjingpai.com\/de\/wp-json\/wp\/v2\/media?parent=14908"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/de\/wp-json\/wp\/v2\/categories?post=14908"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/de\/wp-json\/wp\/v2\/tags?post=14908"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}