{"id":21065,"date":"2025-02-14T11:08:26","date_gmt":"2025-02-14T03:08:26","guid":{"rendered":"https:\/\/www.aisharenet.com\/?p=21065"},"modified":"2025-07-14T10:22:11","modified_gmt":"2025-07-14T02:22:11","slug":"simba","status":"publish","type":"post","link":"https:\/\/www.kdjingpai.com\/pt\/simba\/","title":{"rendered":"Simba\uff1a\u6536\u7eb3\u6587\u6863\u7684\u77e5\u8bc6\u7ba1\u7406\u7cfb\u7edf\uff0c\u65e0\u7f1d\u96c6\u6210\u5230\u4efb\u4f55RAG\u7cfb\u7edf"},"content":{"rendered":"<p>Simba\u662f\u4e00\u4e2a\u4fbf\u643a\u5f0f\u77e5\u8bc6\u7ba1\u7406\u7cfb\u7edf\uff08KMS\uff09\uff0c\u65e8\u5728\u4e0e\u4efb\u4f55\u68c0\u7d22\u589e\u5f3a\u751f\u6210\uff08RAG\uff09\u7cfb\u7edf\u65e0\u7f1d\u96c6\u6210\u3002\u8be5\u9879\u76ee\u7531GitHub\u7528\u6237GitHamza0206\u521b\u5efa\uff0c\u63d0\u4f9b\u4e86\u4e00\u79cd\u9ad8\u6548\u7684\u77e5\u8bc6\u7ba1\u7406\u89e3\u51b3\u65b9\u6848\uff0c\u9002\u7528\u4e8e\u5404\u79cd\u5e94\u7528\u573a\u666f\u3002Simba\u7684\u8bbe\u8ba1\u76ee\u6807\u662f\u7b80\u5316\u77e5\u8bc6\u7ba1\u7406\u6d41\u7a0b\uff0c\u63d0\u9ad8\u4fe1\u606f\u68c0\u7d22\u548c\u751f\u6210\u7684\u51c6\u786e\u6027\u548c\u6548\u7387\u3002\u901a\u8fc7\u96c6\u6210RAG\u7cfb\u7edf\uff0cSimba\u80fd\u591f\u5728\u5904\u7406\u590d\u6742\u6570\u636e\u548c\u751f\u6210\u5185\u5bb9\u65f6\u63d0\u4f9b\u5f3a\u5927\u7684\u652f\u6301\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter wp-image-21066\" title=\"Simba\uff1a\u6536\u7eb3\u6587\u6863\u7684\u77e5\u8bc6\u7ba1\u7406\u7cfb\u7edf\uff0c\u65e0\u7f1d\u96c6\u6210\u5230\u4efb\u4f55RAG\u7cfb\u7edf-1\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/02\/cd09ce039427672.jpg\" alt=\"Simba\uff1a\u6536\u7eb3\u6587\u6863\u7684\u77e5\u8bc6\u7ba1\u7406\u7cfb\u7edf\uff0c\u65e0\u7f1d\u96c6\u6210\u5230\u4efb\u4f55RAG\u7cfb\u7edf-1\" width=\"1023\" height=\"663\" srcset=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/02\/cd09ce039427672.jpg 1662w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/02\/cd09ce039427672-768x498.jpg 768w\" sizes=\"auto, (max-width: 1023px) 100vw, 1023px\" \/><\/p>\n<p>&nbsp;<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter wp-image-21067\" title=\"Simba\uff1a\u6536\u7eb3\u6587\u6863\u7684\u77e5\u8bc6\u7ba1\u7406\u7cfb\u7edf\uff0c\u65e0\u7f1d\u96c6\u6210\u5230\u4efb\u4f55RAG\u7cfb\u7edf-1\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/02\/fbd934518149c76.png\" alt=\"Simba\uff1a\u6536\u7eb3\u6587\u6863\u7684\u77e5\u8bc6\u7ba1\u7406\u7cfb\u7edf\uff0c\u65e0\u7f1d\u96c6\u6210\u5230\u4efb\u4f55RAG\u7cfb\u7edf-1\" width=\"1032\" height=\"545\" srcset=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/02\/fbd934518149c76.png 1422w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/02\/fbd934518149c76-768x406.png 768w\" sizes=\"auto, (max-width: 1032px) 100vw, 1032px\" \/><\/p>\n<p>&nbsp;<\/p>\n<h2>\u529f\u80fd\u5217\u8868<\/h2>\n<ul>\n<li><strong>\u77e5\u8bc6\u7ba1\u7406<\/strong>\uff1a\u63d0\u4f9b\u5168\u9762\u7684\u77e5\u8bc6\u7ba1\u7406\u529f\u80fd\uff0c\u652f\u6301\u77e5\u8bc6\u7684\u5b58\u50a8\u3001\u5206\u7c7b\u548c\u68c0\u7d22\u3002<\/li>\n<li><strong>RAG\u7cfb\u7edf\u96c6\u6210<\/strong>\uff1a\u4e0e\u68c0\u7d22\u589e\u5f3a\u751f\u6210\u7cfb\u7edf\u65e0\u7f1d\u96c6\u6210\uff0c\u63d0\u5347\u4fe1\u606f\u751f\u6210\u7684\u51c6\u786e\u6027\u3002<\/li>\n<li><strong>\u4fbf\u643a\u6027<\/strong>\uff1a\u8bbe\u8ba1\u4e3a\u4fbf\u643a\u5f0f\u7cfb\u7edf\uff0c\u6613\u4e8e\u90e8\u7f72\u548c\u4f7f\u7528\u3002<\/li>\n<li><strong>\u5f00\u6e90\u9879\u76ee<\/strong>\uff1a\u4f5c\u4e3a\u5f00\u6e90\u9879\u76ee\uff0c\u7528\u6237\u53ef\u4ee5\u81ea\u7531\u8bbf\u95ee\u6e90\u4ee3\u7801\u5e76\u8fdb\u884c\u5b9a\u5236\u3002<\/li>\n<li><strong>\u9ad8\u6548\u68c0\u7d22<\/strong>\uff1a\u4f18\u5316\u7684\u4fe1\u606f\u68c0\u7d22\u529f\u80fd\uff0c\u5feb\u901f\u627e\u5230\u6240\u9700\u77e5\u8bc6\u3002<\/li>\n<li><strong>\u7528\u6237\u53cb\u597d\u754c\u9762<\/strong>\uff1a\u63d0\u4f9b\u76f4\u89c2\u7684\u7528\u6237\u754c\u9762\uff0c\u7b80\u5316\u64cd\u4f5c\u6d41\u7a0b\u3002<\/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><strong>\u514b\u9686\u4ed3\u5e93<\/strong>\uff1a\u9996\u5148\uff0c\u4f7f\u7528Git\u547d\u4ee4\u514b\u9686Simba\u9879\u76ee\u7684GitHub\u4ed3\u5e93\u3002<\/li>\n<\/ol>\n<pre><code>   git clone https:\/\/github.com\/GitHamza0206\/simba.git\r\n<\/code><\/pre>\n<ol start=\"2\">\n<li><strong>\u5b89\u88c5\u4f9d\u8d56<\/strong>\uff1a\u8fdb\u5165\u9879\u76ee\u76ee\u5f55\u5e76\u5b89\u88c5\u6240\u9700\u7684\u4f9d\u8d56\u5305\u3002<\/li>\n<\/ol>\n<pre><code>   cd simba\r\n<\/code><\/pre>\n<h3>\u672c\u5730\u5f00\u53d1<\/h3>\n<ol>\n<li><strong>\u540e\u7aef\u8bbe\u7f6e<\/strong>\uff1a\n<ul>\n<li>\u8fdb\u5165\u540e\u7aef\u76ee\u5f55\uff1a<\/li>\n<\/ul>\n<pre><code> cd backend\r\n<\/code><\/pre>\n<ul>\n<li>\u786e\u4fdd\u5728\u64cd\u4f5c\u7cfb\u7edf\u4e2d\u5b89\u88c5\u4e86Redis\uff1a<\/li>\n<\/ul>\n<pre><code> redis-server\r\n<\/code><\/pre>\n<ul>\n<li>\u8bbe\u7f6e\u73af\u5883\u53d8\u91cf\uff1a<\/li>\n<\/ul>\n<pre><code> cp .env.example .env\r\n<\/code><\/pre>\n<p>\u7136\u540e\u7f16\u8f91.env\u6587\u4ef6\uff0c\u586b\u5165\u60a8\u7684\u503c\uff1a<\/p>\n<pre><code> OPENAI_API_KEY=\"\"\r\nLANGCHAIN_TRACING_V2= #(optional - for langsmith tracing)\r\nLANGCHAIN_API_KEY=\"\" #(optional - for langsmith tracing)\r\nREDIS_HOST=redis\r\nCELERY_BROKER_URL=redis:\/\/redis:6379\/0\r\nCELERY_RESULT_BACKEND=redis:\/\/redis:6379\/1\r\n<\/code><\/pre>\n<ul>\n<li>\u5b89\u88c5\u4f9d\u8d56\uff1a<\/li>\n<\/ul>\n<pre><code> poetry install\r\npoetry shell\r\n<\/code><\/pre>\n<p>\u6216\u8005\u5728Mac\/Linux\u4e0a\uff1a<\/p>\n<pre><code> source .venv\/bin\/activate\r\n<\/code><\/pre>\n<p>\u5728Windows\u4e0a\uff1a<\/p>\n<pre><code> .venv\\Scripts\\activate\r\n<\/code><\/pre>\n<ul>\n<li>\u8fd0\u884c\u540e\u7aef\u670d\u52a1\uff1a<\/li>\n<\/ul>\n<pre><code> python main.py\r\n<\/code><\/pre>\n<p>\u6216\u8005\u4f7f\u7528\u81ea\u52a8\u91cd\u8f7d\uff1a<\/p>\n<pre><code> uvicorn main:app --reload\r\n<\/code><\/pre>\n<p>\u7136\u540e\u5bfc\u822a\u5230<code>http:\/\/0.0.0.0:8000\/docs<\/code>\u8bbf\u95eeSwagger UI\uff08\u53ef\u9009\uff09\u3002<\/p>\n<ul>\n<li>\u4f7f\u7528Celery\u8fd0\u884c\u89e3\u6790\u5668\uff1a<\/li>\n<\/ul>\n<pre><code> celery -A tasks.parsing_tasks worker --loglevel=info\r\n<\/code><\/pre>\n<ul>\n<li>\u6839\u636e\u9700\u8981\u4fee\u6539<code>config.yaml<\/code>\u6587\u4ef6\uff1a<\/li>\n<\/ul>\n<pre><code> project:\r\nname: \"Simba\"\r\nversion: \"1.0.0\"\r\napi_version: \"\/api\/v1\"\r\npaths:\r\nbase_dir: null  # Will be set programmatically\r\nmarkdown_dir: \"markdown\"\r\nfaiss_index_dir: \"vector_stores\/faiss_index\"\r\nvector_store_dir: \"vector_stores\"\r\nllm:\r\nprovider: \"openai\" #or <a href=\"https:\/\/www.kdjingpai.com\/de\/ollama\/\">ollama<\/a> (<a href=\"https:\/\/www.kdjingpai.com\/de\/vllm\/\">vllm<\/a> coming soon)\r\nmodel_name: \"gpt-4o\" #or ollama model name\r\ntemperature: 0.0\r\nmax_tokens: null\r\nstreaming: true\r\nadditional_params: {}\r\nembedding:\r\nprovider: \"huggingface\" #or openai\r\nmodel_name: \"BAAI\/bge-base-en-v1.5\" #or any HF model name\r\ndevice: \"cpu\"  # mps,cuda,cpu\r\nadditional_params: {}\r\nvector_store:\r\nprovider: \"faiss\"\r\ncollection_name: \"migi_collection\"\r\nadditional_params: {}\r\nchunking:\r\nchunk_size: 512\r\nchunk_overlap: 200\r\n<a href=\"https:\/\/www.kdjingpai.com\/de\/retrieval\/\">retrieval<\/a>:\r\nk: 5 #number of chunks to retrieve\r\nfeatures:\r\nenable_parsers: true  # Set to false to disable parsing\r\ncelery:\r\nbroker_url: ${CELERY_BROKER_URL:-redis:\/\/redis:6379\/0}\r\nresult_backend: ${CELERY_RESULT_BACKEND:-redis:\/\/redis:6379\/1}\r\n<\/code><\/pre>\n<\/li>\n<li><strong>\u524d\u7aef\u8bbe\u7f6e<\/strong>\uff1a\n<ul>\n<li>\u786e\u4fdd\u5728Simba\u6839\u76ee\u5f55\u4e0b\uff1a <code>bash<br \/>\ncd frontend<br \/>\n<\/code><\/li>\n<li>\u5b89\u88c5\u4f9d\u8d56\uff1a <code>bash<br \/>\nnpm install<br \/>\n<\/code><\/li>\n<li>\u8fd0\u884c\u524d\u7aef\u670d\u52a1\uff1a <code>bash<br \/>\nnpm run dev<br \/>\n<\/code> \u7136\u540e\u5bfc\u822a\u5230<code>http:\/\/localhost:5173<\/code>\u8bbf\u95ee\u524d\u7aef\u754c\u9762\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n<h3>\u4f7f\u7528Docker\u542f\u52a8\uff08\u63a8\u8350\uff09<\/h3>\n<ol>\n<li>\u5bfc\u822a\u5230Simba\u6839\u76ee\u5f55\uff1a<\/li>\n<\/ol>\n<pre><code>   export OPENAI_API_KEY=\"\" #(optional)\r\ndocker-compose up --build\r\n<\/code><\/pre>\n<h3>\u9879\u76ee\u7ed3\u6784<\/h3>\n<pre><code>simba\/\r\n\u251c\u2500\u2500 backend\/              # \u6838\u5fc3\u5904\u7406\u5f15\u64ce\r\n\u2502   \u251c\u2500\u2500 api\/              # FastAPI\u7aef\u70b9\r\n\u2502   \u251c\u2500\u2500 services\/         # \u6587\u6863\u5904\u7406\u903b\u8f91\r\n\u2502   \u251c\u2500\u2500 tasks\/            # Celery\u4efb\u52a1\u5b9a\u4e49\r\n\u2502   \u2514\u2500\u2500 models\/           # Pydantic\u6570\u636e\u6a21\u578b\r\n\u251c\u2500\u2500 frontend\/             # \u57fa\u4e8eReact\u7684UI\r\n\u2502   \u251c\u2500\u2500 public\/           # \u9759\u6001\u8d44\u6e90\r\n\u2502   \u2514\u2500\u2500 src\/              # React\u7ec4\u4ef6\r\n\u251c\u2500\u2500 docker-compose.yml    # \u5f00\u53d1\u73af\u5883\r\n\u2514\u2500\u2500 docker-compose.prod.yml # \u751f\u4ea7\u73af\u5883\u8bbe\u7f6e\r\n<\/code><\/pre>\n<h3>\u914d\u7f6e<\/h3>\n<p><code>config.yaml<\/code>\u6587\u4ef6\u7528\u4e8e\u914d\u7f6e\u540e\u7aef\u5e94\u7528\u7a0b\u5e8f\u3002\u60a8\u53ef\u4ee5\u66f4\u6539\u4ee5\u4e0b\u5185\u5bb9\uff1a<\/p>\n<ul>\n<li>\u5d4c\u5165\u6a21\u578b<\/li>\n<li>\u5411\u91cf\u5b58\u50a8<\/li>\n<li>\u5206\u5757<\/li>\n<li>\u68c0\u7d22<\/li>\n<li>\u529f\u80fd<\/li>\n<li>\u89e3\u6790\u5668<\/li>\n<\/ul>\n<p>\u6709\u5173\u66f4\u591a\u4fe1\u606f\uff0c\u8bf7\u5bfc\u822a\u5230<code>backend\/README.md<\/code>\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Simba\u662f\u4e00\u4e2a\u4fbf\u643a\u5f0f\u77e5\u8bc6\u7ba1\u7406\u7cfb\u7edf\uff08KMS\uff09\uff0c\u65e8\u5728\u4e0e\u4efb\u4f55\u68c0\u7d22\u589e\u5f3a\u751f\u6210\uff08RAG\uff09\u7cfb\u7edf\u65e0\u7f1d\u96c6\u6210\u3002\u8be5\u9879\u76ee\u7531GitHub\u7528\u6237GitHamza0206\u521b\u5efa\uff0c\u63d0\u4f9b\u4e86\u4e00\u79cd\u9ad8\u6548\u7684\u77e5\u8bc6\u7ba1\u7406\u89e3\u51b3\u65b9\u6848\uff0c\u9002\u7528\u4e8e\u5404\u79cd\u5e94\u7528\u573a\u666f\u3002Simba\u7684\u8bbe\u8ba1\u76ee\u6807\u662f\u7b80\u5316\u77e5\u8bc6\u7ba1\u7406\u6d41\u7a0b\uff0c&#8230;<\/p>\n","protected":false},"author":1,"featured_media":61831,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[425,20,455],"tags":[230,243],"class_list":["post-21065","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-professional","category-tool","category-knowledge-base","tag-aikaiyuanxiangmu","tag-aizhishikuyukefu"],"_links":{"self":[{"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/posts\/21065","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=21065"}],"version-history":[{"count":0,"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/posts\/21065\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/media\/61831"}],"wp:attachment":[{"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/media?parent=21065"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/categories?post=21065"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/pt\/wp-json\/wp\/v2\/tags?post=21065"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}