{"id":39446,"date":"2025-03-15T21:15:15","date_gmt":"2025-03-15T13:15:15","guid":{"rendered":"https:\/\/17aitech.com\/?p=39446"},"modified":"2025-04-07T14:52:02","modified_gmt":"2025-04-07T06:52:02","slug":"%e3%80%90%e6%a8%a1%e5%9e%8b%e6%b5%8b%e8%af%95%e3%80%91%e5%a4%a7%e6%a8%a1%e5%9e%8b%e8%af%84%e6%b5%8b%e5%b7%a5%e5%85%b7opencompass%e4%bd%bf%e7%94%a8%e6%96%b9%e6%b3%95%e6%80%bb%e7%bb%93","status":"publish","type":"post","link":"https:\/\/17aitech.com\/?p=39446","title":{"rendered":"\u3010\u6a21\u578b\u6d4b\u8bd5\u3011\u5927\u6a21\u578b\u8bc4\u6d4b\u5de5\u5177OpenCompass\u4f7f\u7528\u65b9\u6cd5\u603b\u7ed3"},"content":{"rendered":"<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_78 ez-toc-wrap-left-text counter-hierarchy ez-toc-counter ez-toc-light-blue ez-toc-container-direction\">\n<div class=\"ez-toc-title-container\">\n<p class=\"ez-toc-title\" style=\"cursor:inherit\">\u6587\u7ae0\u76ee\u5f55<\/p>\n<span class=\"ez-toc-title-toggle\"><a href=\"#\" class=\"ez-toc-pull-right ez-toc-btn ez-toc-btn-xs ez-toc-btn-default ez-toc-toggle\" aria-label=\"Toggle Table of Content\"><span class=\"ez-toc-js-icon-con\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #999;color:#999\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #999;color:#999\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/span><\/a><\/span><\/div>\n<nav><ul class='ez-toc-list ez-toc-list-level-1 ' ><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/17aitech.com\/?p=39446\/#%E5%89%8D%E8%A8%80\" >\u524d\u8a00<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/17aitech.com\/?p=39446\/#%E7%AE%80%E4%BB%8B\" >\u7b80\u4ecb<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/17aitech.com\/?p=39446\/#%E4%BD%BF%E7%94%A8%E6%96%B9%E6%B3%95\" >\u4f7f\u7528\u65b9\u6cd5<\/a><ul class='ez-toc-list-level-3' ><li class='ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/17aitech.com\/?p=39446\/#1_%E5%88%9B%E5%BB%BA%E8%99%9A%E6%8B%9F%E7%8E%AF%E5%A2%83\" >1. \u521b\u5efa\u865a\u62df\u73af\u5883<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"https:\/\/17aitech.com\/?p=39446\/#2_%E5%AE%89%E8%A3%85%E5%B7%A5%E5%85%B7\" >2. \u5b89\u88c5\u5de5\u5177<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"https:\/\/17aitech.com\/?p=39446\/#3_%E5%AE%89%E8%A3%85%E4%BE%9D%E8%B5%96\" >3. \u5b89\u88c5\u4f9d\u8d56<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-7\" href=\"https:\/\/17aitech.com\/?p=39446\/#4_%E8%8E%B7%E5%8F%96API_Key\" >4. \u83b7\u53d6API Key<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-8\" href=\"https:\/\/17aitech.com\/?p=39446\/#5_%E5%88%9B%E5%BB%BAAPI%E6%B5%8B%E8%AF%95%E8%84%9A%E6%9C%AC\" >5. \u521b\u5efaAPI\u6d4b\u8bd5\u811a\u672c<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-9\" href=\"https:\/\/17aitech.com\/?p=39446\/#6_%E9%85%8D%E7%BD%AE%E6%B5%8B%E8%AF%95%E6%95%B0%E6%8D%AE%E9%9B%86\" >6. \u914d\u7f6e\u6d4b\u8bd5\u6570\u636e\u96c6<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-10\" href=\"https:\/\/17aitech.com\/?p=39446\/#7_%E8%BF%90%E8%A1%8C%E6%B5%8B%E8%AF%95\" >7. \u8fd0\u884c\u6d4b\u8bd5<\/a><\/li><\/ul><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-11\" href=\"https:\/\/17aitech.com\/?p=39446\/#%E5%86%85%E5%AE%B9%E5%B0%8F%E7%BB%93\" >\u5185\u5bb9\u5c0f\u7ed3<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-12\" href=\"https:\/\/17aitech.com\/?p=39446\/#%E5%85%B6%E4%BB%96%E6%96%87%E7%AB%A0\" >\u5176\u4ed6\u6587\u7ae0<\/a><\/li><\/ul><\/nav><\/div>\n<h2><span class=\"ez-toc-section\" id=\"%E5%89%8D%E8%A8%80\"><\/span>\u524d\u8a00<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>\u7531\u4e8e\u4f7f\u7528 <code>lm_evaluation_harness<\/code> \u5de5\u5177\u8bc4\u6d4b\u65f6\uff0c\u9047\u5230\u8f83\u591a\u590d\u6742\u7684\u95ee\u9898\u4e0d\u597d\u5904\u7406\uff0c\u4f8b\u5982\uff1a<\/p>\n<ul>\n<li>\u8fde\u63a5 <code>huggingface<\/code> \u4e0b\u8f7d <code>tokenizer<\/code> \u88ab\u5899\uff1b<\/li>\n<li>\u8bc4\u6d4b <code>API<\/code> \u65f6\u9700\u8981\u670d\u52a1\u5668\u652f\u6301 <code>LogProbs<\/code> \u7b49\u95ee\u9898<\/li>\n<li>\u6e90\u4ee3\u7801\u8f83\u4e3a\u6666\u6da9\u96be\u61c2\uff0c<code>Readme<\/code> \u6587\u6863\u4e0d\u8be6\u7ec6&#8230;.<\/li>\n<\/ul>\n<p>\u5bfc\u81f4\u8bc4\u6d4b\u5de5\u5177\u7684\u4f7f\u7528\u6210\u672c\u4ee5\u53ca\u4f53\u9a8c\u4e0d\u4f73\uff0c\u56e0\u6b64\u6211\u4eec\u5bfb\u6c42\u4e00\u6b3e\u56fd\u4ea7\u7684\u3001\u6e90\u7801\u53ef\u8bfb\u6027\u9ad8\uff0c\u6587\u6863\u8be6\u7ec6\u7684\u8bc4\u6d4b\u5de5\u5177\uff1aOpenCompass\u3002<\/p>\n<h2><span class=\"ez-toc-section\" id=\"%E7%AE%80%E4%BB%8B\"><\/span>\u7b80\u4ecb<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>OpenCompass \u662f\u7531\u4e0a\u6d77\u4eba\u5de5\u667a\u80fd\u5b9e\u9a8c\u5ba4\u63a8\u51fa\u7684\u5f00\u6e90\u5927\u6a21\u578b\u8bc4\u6d4b\u4f53\u7cfb\uff0c\u4e3b\u8981\u7279\u70b9\u5305\u62ec\uff1a<\/p>\n<ol>\n<li><strong>\u5168\u80fd\u529b\u8bc4\u4f30<\/strong> &#8211; \u63d0\u4f9b50+\u8bc4\u6d4b\u6570\u636e\u96c6\uff0c\u8986\u76d6\u8bed\u8a00\/\u77e5\u8bc6\/\u63a8\u7406\/\u521b\u4f5c\u7b49\u4e03\u5927\u80fd\u529b\u7ef4\u5ea6<\/li>\n<li><strong>\u591a\u6a21\u578b\u652f\u6301<\/strong> &#8211; \u652f\u6301HuggingFace\/API\u7b4950+\u4e3b\u6d41\u6a21\u578b\u63a5\u5165\uff0c\u5305\u62ecLLaMA\/GLM\/ChatGPT\u7b49<\/li>\n<li><strong>\u672c\u571f\u5316\u4f18\u52bf<\/strong> &#8211; \u5185\u7f6e\u6587\u8a00\u6587\u7406\u89e3\u3001\u6cd5\u5f8b\u4f26\u7406\u7b49\u4e2d\u6587\u7279\u8272\u8bc4\u6d4b\u7ef4\u5ea6<\/li>\n<li><strong>\u8bf4\u660e\u6587\u6863\u5168<\/strong> &#8211; \u63d0\u4f9b\u4e86\u8be6\u7ec6\u7684\u8bf4\u660e\u6587\u6863\uff0c\u5305\u62ec\u5b89\u88c5\u3001\u4f7f\u7528\u3001\u914d\u7f6e\u7b49<\/li>\n<\/ol>\n<p>\u4ed3\u5e93\u5730\u5740\uff1a<a href=\"https:\/\/github.com\/open-compass\/opencompass\">https:\/\/github.com\/open-compass\/opencompass<\/a><br \/>\n\u8bf4\u660e\u6587\u6863\uff1a<a href=\"https:\/\/opencompass.readthedocs.io\/zh-cn\/latest\/get_started\/quick_start.html\">https:\/\/opencompass.readthedocs.io\/zh-cn\/latest\/get_started\/quick_start.html<\/a><br \/>\n<a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/03\/\u5de5\u5177\u5168\u666f\u56fe.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/03\/\u5de5\u5177\u5168\u666f\u56fe.png\" alt=\"\" \/><\/a><\/p>\n<h2><span class=\"ez-toc-section\" id=\"%E4%BD%BF%E7%94%A8%E6%96%B9%E6%B3%95\"><\/span>\u4f7f\u7528\u65b9\u6cd5<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<h3><span class=\"ez-toc-section\" id=\"1_%E5%88%9B%E5%BB%BA%E8%99%9A%E6%8B%9F%E7%8E%AF%E5%A2%83\"><\/span>1. \u521b\u5efa\u865a\u62df\u73af\u5883<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<pre><code class=\"language-bash\">conda create --name opencompass python=3.10 -y\nconda activate opencompass<\/code><\/pre>\n<h3><span class=\"ez-toc-section\" id=\"2_%E5%AE%89%E8%A3%85%E5%B7%A5%E5%85%B7\"><\/span>2. \u5b89\u88c5\u5de5\u5177<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>\u8be5\u5de5\u5177\u63d0\u4f9b <code>pip install<\/code> \u548c <code>\u6e90\u7801<\/code> \u4e24\u79cd\u5b89\u88c5\u65b9\u5f0f\u3002\u7531\u4e8e\u6211\u4eec\u540e\u7eed\u9700\u8981\u81ea\u5b9a\u4e49\u8bc4\u6d4b\u6a21\u578b\u7684 <code>API<\/code> \u4ee5\u53ca\u6570\u636e\u96c6\uff0c\u6240\u4ee5\u6b64\u5904\u9009\u62e9\u6e90\u7801\u65b9\u5f0f\u5b89\u88c5\u3002<\/p>\n<pre><code class=\"language-bash\">git clone https:\/\/github.com\/open-compass\/opencompass opencompass\ncd opencompass\npip install -e .<\/code><\/pre>\n<h3><span class=\"ez-toc-section\" id=\"3_%E5%AE%89%E8%A3%85%E4%BE%9D%E8%B5%96\"><\/span>3. \u5b89\u88c5\u4f9d\u8d56<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<pre><code class=\"language-bash\"># \u5b89\u88c5sentencepiece\npip install torch sentencepiece protobuf<\/code><\/pre>\n<pre><code class=\"language-bash\"># \u5b89\u88c5torch\npip install torch torchvision torchaudio<\/code><\/pre>\n<h3><span class=\"ez-toc-section\" id=\"4_%E8%8E%B7%E5%8F%96API_Key\"><\/span>4. \u83b7\u53d6API Key<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>\u8bbf\u95eeDeepseek\u5b98\u7f51\u7684\u5f00\u53d1\u8005\u5e73\u53f0\uff0c\u83b7\u53d6 <code>API Key<\/code> \u4ee5\u53ca <code>API URL<\/code>\u3002<\/p>\n<p>\u5728\u672c\u5730\u65b0\u5efaJupyter Notebook\u6587\u4ef6\uff0c\u6d4b\u8bd5API\u662f\u5426\u53ef\u7528\u3002<\/p>\n<pre><code class=\"language-python\"># \u6d4b\u8bd5API\u53ef\u7528\nfrom openai import OpenAI\nopenai_api_key = &quot;sk-fe599*******&quot;\nopenai_api_base = &quot;https:\/\/api.deepseek.com\/&quot;\nmodel = &quot;deepseek-chat&quot;\n\nclient = OpenAI(\n    api_key=openai_api_key,\n    base_url=openai_api_base,\n)\n\nchat_response = client.chat.completions.create(\n    model=model,\n    messages=[\n        {&quot;role&quot;: &quot;system&quot;, &quot;content&quot;: &quot;\u4f60\u662f\u4e00\u4e2a\u5f88\u6709\u7528\u7684\u52a9\u624b\u3002&quot;},\n        {&quot;role&quot;: &quot;user&quot;, &quot;content&quot;: &quot;\u4e2d\u534e\u4eba\u6c11\u5171\u548c\u56fd\u7684\u9996\u90fd\u662f\u54ea\u91cc\uff1f&quot;},\n    ]\n)\nprint(chat_response.choices[0].message.content)<\/code><\/pre>\n<p>\u8fd0\u884c\u540e\uff0cAPI\u6b63\u5e38\u8fd4\u56de\uff1a<\/p>\n<pre><code class=\"language-bash\">\u4e2d\u534e\u4eba\u6c11\u5171\u548c\u56fd\u7684\u9996\u90fd\u662f\u5317\u4eac\u3002\u5317\u4eac\u662f\u4e2d\u56fd\u7684\u653f\u6cbb\u3001\u6587\u5316\u3001\u56fd\u9645\u4ea4\u5f80\u548c\u79d1\u6280\u521b\u65b0\u4e2d\u5fc3\uff0c\u627f\u8f7d\u7740\u4e2d\u534e\u6c11\u65cf\u7684\u60a0\u4e45\u5386\u53f2\u548c\u707f\u70c2\u6587\u5316\uff0c\u540c\u65f6\u4e5f\u662f\u4e2d\u56fd\u5171\u4ea7\u515a\u4e2d\u592e\u59d4\u5458\u4f1a\u3001\u5168\u56fd\u4eba\u6c11\u4ee3\u8868\u5927\u4f1a\u548c\u56fd\u52a1\u9662\u7b49\u56fd\u5bb6\u6700\u9ad8\u6743\u529b\u673a\u5173\u7684\u6240\u5728\u5730\u3002\u5317\u4eac\u7684\u53d1\u5c55\u6210\u5c31\u5145\u5206\u5c55\u793a\u4e86\u4e2d\u56fd\u7279\u8272\u793e\u4f1a\u4e3b\u4e49\u5236\u5ea6\u7684\u4f18\u8d8a\u6027\u548c\u4e2d\u56fd\u5171\u4ea7\u515a\u9886\u5bfc\u7684\u4f1f\u5927\u529b\u91cf\u3002<\/code><\/pre>\n<h3><span class=\"ez-toc-section\" id=\"5_%E5%88%9B%E5%BB%BAAPI%E6%B5%8B%E8%AF%95%E8%84%9A%E6%9C%AC\"><\/span>5. \u521b\u5efaAPI\u6d4b\u8bd5\u811a\u672c<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>\u5728OpenCompass\u5de5\u7a0b\u76ee\u5f55\u4e0b\uff0c\u6309\u7167\u5982\u4e0b\u8def\u5f84\u521b\u5efa.py\u6587\u4ef6<\/p>\n<p>\u4ee3\u7801\u8def\u5f84\uff1a<code>opencompass\/configs\/models\/openai\/custom_api.py<\/code><\/p>\n<p>\u4ee3\u7801\u5185\u5bb9\uff1a<\/p>\n<pre><code class=\"language-python\">import os\nfrom opencompass.models import OpenAISDK\n\ninternlm_url = &#039;https:\/\/api.deepseek.com\/&#039;                                    # \u524d\u9762\u83b7\u5f97\u7684 api \u670d\u52a1\u5730\u5740\ninternlm_api_key = &quot;sk-fe5990***&quot;                                             # \u524d\u9762\u83b7\u5f97\u7684 API Key\n\nmodels = [\n    dict(\n        type=OpenAISDK,\n        path=&#039;deepseek-chat&#039;,               # \u8bf7\u6c42\u670d\u52a1\u65f6\u7684 model name\n        key=internlm_api_key, \n        openai_api_base=internlm_url, \n        rpm_verbose=True,                   # \u662f\u5426\u6253\u5370\u8bf7\u6c42\u901f\u7387\n        query_per_second=0.16,              # \u670d\u52a1\u8bf7\u6c42\u901f\u7387\n        max_out_len=1024,                   # \u6700\u5927\u8f93\u51fa\u957f\u5ea6\n        max_seq_len=4096,                   # \u6700\u5927\u8f93\u5165\u957f\u5ea6\n        temperature=0.01,                   # \u751f\u6210\u6e29\u5ea6\n        batch_size=1,                       # \u6279\u5904\u7406\u5927\u5c0f\n        retry=3,                            # \u91cd\u8bd5\u6b21\u6570\n    )\n]<\/code><\/pre>\n<h3><span class=\"ez-toc-section\" id=\"6_%E9%85%8D%E7%BD%AE%E6%B5%8B%E8%AF%95%E6%95%B0%E6%8D%AE%E9%9B%86\"><\/span>6. \u914d\u7f6e\u6d4b\u8bd5\u6570\u636e\u96c6<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>\u4ee3\u7801\u8def\u5f84\uff1a<code>opencompass\/configs\/datasets\/demo\/demo_cmmlu_chat_gen.py<\/code><\/p>\n<p>\u4ee3\u7801\u5185\u5bb9\uff1a<\/p>\n<pre><code class=\"language-python\">from mmengine import read_base\n\nwith read_base():\n    from ..cmmlu.cmmlu_gen_c13365 import cmmlu_datasets\n\n# \u6bcf\u4e2a\u6570\u636e\u96c6\u53ea\u53d6\u524d2\u4e2a\u6837\u672c\u8fdb\u884c\u8bc4\u6d4b\nfor d in cmmlu_datasets:\n    d[&#039;abbr&#039;] = &#039;demo_&#039; + d[&#039;abbr&#039;]\n    d[&#039;reader_cfg&#039;][&#039;test_range&#039;] = &#039;[0:1]&#039; # \u8fd9\u91cc\u6bcf\u4e2a\u6570\u636e\u96c6\u53ea\u53d61\u4e2a\u6837\u672c, \u65b9\u4fbf\u5feb\u901f\u8bc4\u6d4b.\n<\/code><\/pre>\n<p>\u89e3\u91ca\u8bf4\u660e\uff1a<br \/>\nCMMLU\uff08Chinese Massive Multitask Language Understanding\uff09\u662f\u4e00\u4e2a\u4e13\u95e8\u9488\u5bf9\u4e2d\u6587\u8bed\u8a00\u6a21\u578b\u8bbe\u8ba1\u7684\u7efc\u5408\u6027\u8bc4\u4f30\u57fa\u51c6\uff0c\u4e3b\u8981\u7279\u70b9\u5305\u62ec\uff1a<\/p>\n<ol>\n<li>\u9886\u57df\u8986\u76d6\uff1a\n<ul>\n<li>\u5305\u542b67\u4e2a\u5b66\u79d1\u4e3b\u9898<\/li>\n<li>\u6db5\u76d6\u81ea\u7136\u79d1\u5b66\uff08\u7269\u7406\/\u5316\u5b66\/\u751f\u7269\uff09<\/li>\n<li>\u793e\u4f1a\u79d1\u5b66\uff08\u5386\u53f2\/\u6cd5\u5f8b\/\u5fc3\u7406\u5b66\uff09<\/li>\n<li>\u5de5\u7a0b\u6280\u672f\uff08\u8ba1\u7b97\u673a\/\u7535\u5b50\u5de5\u7a0b\uff09<\/li>\n<li>\u4eba\u6587\u827a\u672f\uff08\u6587\u5b66\/\u54f2\u5b66\uff09\u7b49<\/li>\n<\/ul>\n<\/li>\n<li>\u9898\u76ee\u7c7b\u578b\uff1a\n<ul>\n<li>\u5355\u9879\u9009\u62e9\u9898<\/li>\n<li>\u591a\u9879\u9009\u62e9\u9898<\/li>\n<li>\u63a8\u7406\u5224\u65ad\u9898<\/li>\n<li>\u77e5\u8bc6\u5e94\u7528\u9898<\/li>\n<\/ul>\n<\/li>\n<li>\u8bc4\u4f30\u76ee\u6807\uff1a\n<ul>\n<li>\u6d4b\u8bd5\u6a21\u578b\u7684\u4e2d\u6587\u8bed\u8a00\u7406\u89e3\u80fd\u529b<\/li>\n<li>\u8bc4\u4f30\u8de8\u5b66\u79d1\u77e5\u8bc6\u638c\u63e1\u7a0b\u5ea6<\/li>\n<li>\u68c0\u9a8c\u590d\u6742\u63a8\u7406\u548c\u95ee\u9898\u89e3\u51b3\u80fd\u529b<\/li>\n<li>\u9a8c\u8bc1\u5b9e\u9645\u573a\u666f\u5e94\u7528\u80fd\u529b<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n<h3><span class=\"ez-toc-section\" id=\"7_%E8%BF%90%E8%A1%8C%E6%B5%8B%E8%AF%95\"><\/span>7. \u8fd0\u884c\u6d4b\u8bd5<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>\u5b8c\u6210\u4e0a\u8ff0\u7684\u4ee3\u7801\u4fee\u6539\u4ee5\u53ca\u914d\u7f6e\u540e\uff0c\u5728OpenCompass\u5de5\u7a0b\u76ee\u5f55\u4e0b\uff0c\u8fd0\u884c\u5982\u4e0b\u547d\u4ee4\uff1a<\/p>\n<pre><code class=\"language-bash\">python run.py --models custom_api.py --datasets demo_cmmlu_chat_gen.py --debug<\/code><\/pre>\n<p>\u8fd0\u884c\u7ed3\u679c\uff1a<br \/>\n<a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/03\/\u6267\u884c\u8bc4\u6d4b\u811a\u672c2.png\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/03\/\u6267\u884c\u8bc4\u6d4b\u811a\u672c2.png\" alt=\"\" \/><\/a><\/p>\n<h2><span class=\"ez-toc-section\" id=\"%E5%86%85%E5%AE%B9%E5%B0%8F%E7%BB%93\"><\/span>\u5185\u5bb9\u5c0f\u7ed3<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul>\n<li>\u76f8\u6bd4 <code>lm_evaluation_harness<\/code> \u5de5\u5177\uff0cOpenCompass \u7684\u6e90\u7801\u53ef\u8bfb\u6027\u66f4\u9ad8\uff0c\u6587\u6863\u66f4\u8be6\u7ec6\uff0c\u4f7f\u7528\u6210\u672c\u66f4\u4f4e\u3002<\/li>\n<li>OpenCompass \u652f\u6301\u591a\u79cd\u8bc4\u6d4b\u6570\u636e\u96c6\uff0c\u5305\u62ecCMMLU\u3001C-Eval\u3001C-MTEB\u7b49\uff0c\u65b9\u4fbf\u7528\u6237\u9009\u62e9\u5408\u9002\u7684\u8bc4\u6d4b\u6570\u636e\u96c6\u3002<\/li>\n<li>OpenCompass \u652f\u6301\u672c\u5730\u90e8\u7f72\u548cAPI\u8c03\u7528\u4e24\u79cd\u8bc4\u6d4b\u65b9\u5f0f\uff0c\u65b9\u4fbf\u7528\u6237\u9009\u62e9\u5408\u9002\u7684\u8bc4\u6d4b\u65b9\u5f0f\u3002<\/li>\n<li>\u9664\u6b64\u4e4b\u5916\uff0cOpenCompass \u662f\u679c\u7136\u8bc4\u6d4b\u8f6f\u4ef6\uff0c\u66f4\u52a0\u9002\u5408\u56fd\u60c5\u3002<\/li>\n<\/ul>\n<h2><span class=\"ez-toc-section\" id=\"%E5%85%B6%E4%BB%96%E6%96%87%E7%AB%A0\"><\/span>\u5176\u4ed6\u6587\u7ae0<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul>\n<li><a href=\"https:\/\/17aitech.com\/?p=38434\">\u3010\u6a21\u578b\u6d4b\u8bd5\u3011\u5927\u6a21\u578b\u6d4b\u8bc4\u4f53\u7cfb\u7684\u6784\u6210<\/a><\/li>\n<li><a href=\"https:\/\/17aitech.com\/?p=38565\">\u3010\u6a21\u578b\u6d4b\u8bd5\u3011\u5927\u6a21\u578b\u8bc4\u6d4b\u5de5\u5177lm-evaluation-harness\u7684\u4f7f\u7528\u65b9\u6cd5\u603b\u7ed3<\/a><\/li>\n<li><a href=\"https:\/\/17aitech.com\/?p=39446\">\u3010\u6a21\u578b\u6d4b\u8bd5\u3011\u5927\u6a21\u578b\u8bc4\u6d4b\u5de5\u5177OpenCompass\u4f7f\u7528\u65b9\u6cd5\u603b\u7ed3<\/a><\/li>\n<li><a href=\"https:\/\/17aitech.com\/?p=39926\">\u3010\u6a21\u578b\u6d4b\u8bd5\u3011ai-eval-system\u5728\u7ebf\u8bc4\u6d4b\u7cfb\u7edfv0.2\u9884\u89c8\u7248\u672c\u4ecb\u7ecd<\/a><\/li>\n<\/ul>\n<p align=\"center\">\u6b22\u8fce\u5173\u6ce8\u516c\u4f17\u53f7\u4ee5\u83b7\u5f97\u6700\u65b0\u7684\u6587\u7ae0\u548c\u65b0\u95fb<\/p>\n<p><a href=\"https:\/\/17aitech.com\/wp-content\/uploads\/2024\/09\/\u626b\u7801_\u641c\u7d22\u8054\u5408\u4f20\u64ad\u6837\u5f0f-\u767d\u8272\u7248.bmp\" data-fancybox=\"images\" data-fancybox=\"gallery\"><img decoding=\"async\" src=\"https:\/\/17aitech.com\/wp-content\/uploads\/2024\/09\/\u626b\u7801_\u641c\u7d22\u8054\u5408\u4f20\u64ad\u6837\u5f0f-\u767d\u8272\u7248.bmp\" alt=\"\" \/><\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u524d\u8a00 \u7531\u4e8e\u4f7f\u7528 lm_evaluatio [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":39450,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"aside","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","theme-transparent-header-meta":"default","adv-header-id-meta":"","stick-header-meta":"default","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"set","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[28],"tags":[81],"class_list":["post-39446","post","type-post","status-publish","format-aside","has-post-thumbnail","hentry","category-blog","tag-81","post_format-post-format-aside"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.4 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>\u3010\u6a21\u578b\u6d4b\u8bd5\u3011\u5927\u6a21\u578b\u8bc4\u6d4b\u5de5\u5177OpenCompass\u4f7f\u7528\u65b9\u6cd5\u603b\u7ed3 - \u4e00\u8d77AI\u6280\u672f<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/17aitech.com\/?p=39446\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/17aitech.com\/?p=39446\",\"url\":\"https:\/\/17aitech.com\/?p=39446\",\"name\":\"\u3010\u6a21\u578b\u6d4b\u8bd5\u3011\u5927\u6a21\u578b\u8bc4\u6d4b\u5de5\u5177OpenCompass\u4f7f\u7528\u65b9\u6cd5\u603b\u7ed3 - \u4e00\u8d77AI\u6280\u672f\",\"isPartOf\":{\"@id\":\"https:\/\/17aitech.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/17aitech.com\/?p=39446#primaryimage\"},\"image\":{\"@id\":\"https:\/\/17aitech.com\/?p=39446#primaryimage\"},\"thumbnailUrl\":\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/03\/\u5c01\u9762-scaled.jpeg\",\"datePublished\":\"2025-03-15T13:15:15+00:00\",\"dateModified\":\"2025-04-07T06:52:02+00:00\",\"author\":{\"@id\":\"https:\/\/17aitech.com\/#\/schema\/person\/3d23bb6f7f115fcefc9ae7803a691739\"},\"breadcrumb\":{\"@id\":\"https:\/\/17aitech.com\/?p=39446#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/17aitech.com\/?p=39446\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/17aitech.com\/?p=39446#primaryimage\",\"url\":\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/03\/\u5c01\u9762-scaled.jpeg\",\"contentUrl\":\"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/03\/\u5c01\u9762-scaled.jpeg\",\"width\":2560,\"height\":1558},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/17aitech.com\/?p=39446#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/17aitech.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"\u3010\u6a21\u578b\u6d4b\u8bd5\u3011\u5927\u6a21\u578b\u8bc4\u6d4b\u5de5\u5177OpenCompass\u4f7f\u7528\u65b9\u6cd5\u603b\u7ed3\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/17aitech.com\/#website\",\"url\":\"https:\/\/17aitech.com\/\",\"name\":\"\u4e00\u8d77AI\u6280\u672f\",\"description\":\"\u8ba9AI\u77e5\u8bc6\u89e6\u624b\u53ef\u53ca\",\"alternateName\":\"\u4e00\u8d77AI\u6280\u672f\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/17aitech.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/17aitech.com\/#\/schema\/person\/3d23bb6f7f115fcefc9ae7803a691739\",\"name\":\"Dongming\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/17aitech.com\/#\/schema\/person\/image\/\",\"url\":\"\/\/17aitech.com\/wp-content\/uploads\/member\/avatars\/238a0b923820dcc5.1732798681.jpg\",\"contentUrl\":\"\/\/17aitech.com\/wp-content\/uploads\/member\/avatars\/238a0b923820dcc5.1732798681.jpg\",\"caption\":\"Dongming\"},\"description\":\"\u89c1\u5929\u5730\uff0c\u89c1\u4f17\u751f\uff0c\u89c1\u81ea\u5df1\u3002\",\"sameAs\":[\"http:\/\/17aitech.com\"],\"url\":\"https:\/\/17aitech.com\/?page_id=33738&user=1\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"\u3010\u6a21\u578b\u6d4b\u8bd5\u3011\u5927\u6a21\u578b\u8bc4\u6d4b\u5de5\u5177OpenCompass\u4f7f\u7528\u65b9\u6cd5\u603b\u7ed3 - \u4e00\u8d77AI\u6280\u672f","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/17aitech.com\/?p=39446","schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/17aitech.com\/?p=39446","url":"https:\/\/17aitech.com\/?p=39446","name":"\u3010\u6a21\u578b\u6d4b\u8bd5\u3011\u5927\u6a21\u578b\u8bc4\u6d4b\u5de5\u5177OpenCompass\u4f7f\u7528\u65b9\u6cd5\u603b\u7ed3 - \u4e00\u8d77AI\u6280\u672f","isPartOf":{"@id":"https:\/\/17aitech.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/17aitech.com\/?p=39446#primaryimage"},"image":{"@id":"https:\/\/17aitech.com\/?p=39446#primaryimage"},"thumbnailUrl":"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/03\/\u5c01\u9762-scaled.jpeg","datePublished":"2025-03-15T13:15:15+00:00","dateModified":"2025-04-07T06:52:02+00:00","author":{"@id":"https:\/\/17aitech.com\/#\/schema\/person\/3d23bb6f7f115fcefc9ae7803a691739"},"breadcrumb":{"@id":"https:\/\/17aitech.com\/?p=39446#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/17aitech.com\/?p=39446"]}]},{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/17aitech.com\/?p=39446#primaryimage","url":"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/03\/\u5c01\u9762-scaled.jpeg","contentUrl":"https:\/\/17aitech.com\/wp-content\/uploads\/2025\/03\/\u5c01\u9762-scaled.jpeg","width":2560,"height":1558},{"@type":"BreadcrumbList","@id":"https:\/\/17aitech.com\/?p=39446#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/17aitech.com\/"},{"@type":"ListItem","position":2,"name":"\u3010\u6a21\u578b\u6d4b\u8bd5\u3011\u5927\u6a21\u578b\u8bc4\u6d4b\u5de5\u5177OpenCompass\u4f7f\u7528\u65b9\u6cd5\u603b\u7ed3"}]},{"@type":"WebSite","@id":"https:\/\/17aitech.com\/#website","url":"https:\/\/17aitech.com\/","name":"\u4e00\u8d77AI\u6280\u672f","description":"\u8ba9AI\u77e5\u8bc6\u89e6\u624b\u53ef\u53ca","alternateName":"\u4e00\u8d77AI\u6280\u672f","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/17aitech.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/17aitech.com\/#\/schema\/person\/3d23bb6f7f115fcefc9ae7803a691739","name":"Dongming","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/17aitech.com\/#\/schema\/person\/image\/","url":"\/\/17aitech.com\/wp-content\/uploads\/member\/avatars\/238a0b923820dcc5.1732798681.jpg","contentUrl":"\/\/17aitech.com\/wp-content\/uploads\/member\/avatars\/238a0b923820dcc5.1732798681.jpg","caption":"Dongming"},"description":"\u89c1\u5929\u5730\uff0c\u89c1\u4f17\u751f\uff0c\u89c1\u81ea\u5df1\u3002","sameAs":["http:\/\/17aitech.com"],"url":"https:\/\/17aitech.com\/?page_id=33738&user=1"}]}},"_links":{"self":[{"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/posts\/39446","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/17aitech.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=39446"}],"version-history":[{"count":2,"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/posts\/39446\/revisions"}],"predecessor-version":[{"id":39960,"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/posts\/39446\/revisions\/39960"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/17aitech.com\/index.php?rest_route=\/wp\/v2\/media\/39450"}],"wp:attachment":[{"href":"https:\/\/17aitech.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=39446"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/17aitech.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=39446"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/17aitech.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=39446"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}