{"id":29401,"date":"2025-03-27T19:06:28","date_gmt":"2025-03-27T11:06:28","guid":{"rendered":"https:\/\/www.aisharenet.com\/?p=29401"},"modified":"2025-07-14T11:53:32","modified_gmt":"2025-07-14T03:53:32","slug":"rankify","status":"publish","type":"post","link":"https:\/\/www.kdjingpai.com\/ja\/rankify\/","title":{"rendered":"Rankify\uff1a\u652f\u6301\u4fe1\u606f\u68c0\u7d22\u4e0e\u91cd\u6392\u5e8f\u7684Python\u5de5\u5177\u5305"},"content":{"rendered":"<p>Rankify \u662f\u7531\u5965\u5730\u5229\u56e0\u65af\u5e03\u9c81\u514b\u5927\u5b66\u6570\u636e\u79d1\u5b66\u5c0f\u7ec4\u5f00\u53d1\u7684\u5f00\u6e90 Python \u5de5\u5177\u5305\u3002\u5b83\u4e13\u6ce8\u4e8e\u4fe1\u606f\u68c0\u7d22\u3001\u91cd\u6392\u5e8f\u548c\u68c0\u7d22\u589e\u5f3a\u751f\u6210\uff08RAG\uff09\uff0c\u63d0\u4f9b\u4e86\u4e00\u4e2a\u7edf\u4e00\u7684\u6846\u67b6\u3002\u5de5\u5177\u5305\u5185\u7f6e 40 \u4e2a\u9884\u68c0\u7d22\u57fa\u51c6\u6570\u636e\u96c6\uff0c\u652f\u6301 7 \u79cd\u68c0\u7d22\u6280\u672f\u548c 24 \u79cd\u91cd\u6392\u5e8f\u6a21\u578b\uff0c\u8fd8\u5305\u62ec\u591a\u79cd <a href=\"https:\/\/www.kdjingpai.com\/pt\/rag\/\">RAG<\/a> \u65b9\u6cd5\u3002Rankify \u91c7\u7528\u6a21\u5757\u5316\u8bbe\u8ba1\uff0c\u6613\u4e8e\u6269\u5c55\uff0c\u9002\u5408\u7814\u7a76\u4eba\u5458\u548c\u5f00\u53d1\u8005\u7528\u4e8e\u5b9e\u9a8c\u548c\u57fa\u51c6\u6d4b\u8bd5\u3002\u4ee3\u7801\u5f00\u653e\uff0c\u6587\u6863\u8be6\u5c3d\uff0c\u652f\u6301 Python 3.10 \u53ca\u4ee5\u4e0a\u7248\u672c\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter wp-image-29402\" title=\"Rankify\uff1a\u652f\u6301\u4fe1\u606f\u68c0\u7d22\u4e0e\u91cd\u6392\u5e8f\u7684Python\u5de5\u5177\u5305-1\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9f7ed9b4f575854.jpg\" alt=\"Rankify\uff1a\u652f\u6301\u4fe1\u606f\u68c0\u7d22\u4e0e\u91cd\u6392\u5e8f\u7684Python\u5de5\u5177\u5305-1\" width=\"759\" height=\"1028\" srcset=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9f7ed9b4f575854.jpg 1521w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9f7ed9b4f575854-768x1041.jpg 768w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9f7ed9b4f575854-1134x1536.jpg 1134w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9f7ed9b4f575854-1511x2048.jpg 1511w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9f7ed9b4f575854-9x12.jpg 9w\" sizes=\"auto, (max-width: 759px) 100vw, 759px\" \/><\/p>\n<p>&nbsp;<\/p>\n<h2>\u529f\u80fd\u5217\u8868<\/h2>\n<ul>\n<li>\u63d0\u4f9b 7 \u79cd\u68c0\u7d22\u6280\u672f\uff0c\u5305\u62ec BM25\u3001DPR\u3001ColBERT\u3001ANCE\u3001BGE\u3001Contriever \u548c HYDE\u3002<\/li>\n<li>\u652f\u6301 24 \u79cd\u91cd\u6392\u5e8f\u6a21\u578b\uff0c\u5982 MonoT5\u3001RankGPT\u3001Sentence <a href=\"https:\/\/www.kdjingpai.com\/pt\/transformer\/\">Transformer<\/a> \u7b49\uff0c\u63d0\u5347\u68c0\u7d22\u7ed3\u679c\u51c6\u786e\u6027\u3002<\/li>\n<li>\u96c6\u6210\u68c0\u7d22\u589e\u5f3a\u751f\u6210\uff08RAG\uff09\uff0c\u652f\u6301 GPT\u3001LLaMA\u3001T5 \u7b49\u6a21\u578b\u751f\u6210\u56de\u7b54\u3002<\/li>\n<li>\u5185\u7f6e 40 \u4e2a\u9884\u68c0\u7d22\u6570\u636e\u96c6\uff0c\u6db5\u76d6\u95ee\u7b54\u3001\u5bf9\u8bdd\u3001\u5b9e\u4f53\u94fe\u63a5\u7b49\u573a\u666f\u3002<\/li>\n<li>\u63d0\u4f9b\u8bc4\u4f30\u5de5\u5177\uff0c\u8ba1\u7b97\u68c0\u7d22\u3001\u91cd\u6392\u5e8f\u548c\u751f\u6210\u7ed3\u679c\u7684\u6307\u6807\uff0c\u5982 Top-K\u3001EM\u3001Recall\u3002<\/li>\n<li>\u652f\u6301\u9884\u6784\u5efa\u7d22\u5f15\uff08\u5982 Wikipedia \u548c MS MARCO\uff09\uff0c\u65e0\u9700\u81ea\u5df1\u5efa\u7d22\u5f15\u3002<\/li>\n<li>\u6a21\u5757\u5316\u7ed3\u6784\uff0c\u5141\u8bb8\u7528\u6237\u81ea\u5b9a\u4e49\u6570\u636e\u96c6\u3001\u68c0\u7d22\u5668\u548c\u6a21\u578b\u3002<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2>\u4f7f\u7528\u5e2e\u52a9<\/h2>\n<p>Rankify \u7684\u5b89\u88c5\u548c\u4f7f\u7528\u7b80\u5355\u660e\u4e86\u3002\u4ee5\u4e0b\u662f\u8be6\u7ec6\u6b65\u9aa4\u548c\u64cd\u4f5c\u6307\u5357\uff0c\u5e2e\u52a9\u4f60\u5feb\u901f\u4e0a\u624b\u3002<\/p>\n<h3>\u5b89\u88c5\u6d41\u7a0b<\/h3>\n<p>Rankify \u9700\u8981 Python 3.10 \u6216\u4ee5\u4e0a\u7248\u672c\u3002\u5efa\u8bae\u4f7f\u7528\u865a\u62df\u73af\u5883\u5b89\u88c5\uff0c\u907f\u514d\u4f9d\u8d56\u51b2\u7a81\u3002<\/p>\n<ol>\n<li><strong>\u521b\u5efa\u865a\u62df\u73af\u5883\uff08\u63a8\u8350\uff09<\/strong><br \/>\n\u4f7f\u7528 Conda \u521b\u5efa\u73af\u5883\uff1a<\/li>\n<\/ol>\n<pre><code>conda create -n rankify python=3.10\r\nconda activate rankify\r\n<\/code><\/pre>\n<p>\u6216\u4f7f\u7528 Python \u81ea\u5e26\u5de5\u5177\uff1a<\/p>\n<pre><code>python -m venv rankify_env\r\nsource rankify_env\/bin\/activate  # Linux\/Mac\r\nrankify_env\\Scripts\\activate    # Windows\r\n<\/code><\/pre>\n<ol start=\"2\">\n<li><strong>\u5b89\u88c5 PyTorch\uff08\u63a8\u8350 2.5.1 \u7248\uff09<\/strong><br \/>\n\u5982\u679c\u6709 GPU\uff0c\u5b89\u88c5\u5e26 CUDA 12.4 \u7684\u7248\u672c\uff1a<\/li>\n<\/ol>\n<pre><code>pip install torch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1 --index-url https:\/\/download.pytorch.org\/whl\/cu124\r\n<\/code><\/pre>\n<p>\u65e0 GPU \u5219\u5b89\u88c5 CPU \u7248\uff1a<\/p>\n<pre><code>pip install torch==2.5.1\r\n<\/code><\/pre>\n<ol start=\"3\">\n<li><strong>\u57fa\u7840\u5b89\u88c5<\/strong><br \/>\n\u5b89\u88c5 Rankify \u6838\u5fc3\u529f\u80fd\uff1a<\/li>\n<\/ol>\n<pre><code>pip install rankify\r\n<\/code><\/pre>\n<ol start=\"4\">\n<li><strong>\u5b8c\u6574\u5b89\u88c5\uff08\u63a8\u8350\uff09<\/strong><br \/>\n\u5b89\u88c5\u6240\u6709\u529f\u80fd\uff1a<\/li>\n<\/ol>\n<pre><code>pip install \"rankify[all]\"\r\n<\/code><\/pre>\n<ol start=\"5\">\n<li><strong>\u6309\u9700\u5b89\u88c5\uff08\u53ef\u9009\uff09<\/strong><br \/>\n\u53ea\u5b89\u88c5\u68c0\u7d22\u529f\u80fd\uff1a<\/li>\n<\/ol>\n<pre><code>pip install \"rankify[retriever]\"\r\n<\/code><\/pre>\n<p>\u53ea\u5b89\u88c5\u91cd\u6392\u5e8f\u529f\u80fd\uff1a<\/p>\n<pre><code>pip install \"rankify[reranking]\"\r\n<\/code><\/pre>\n<ol start=\"6\">\n<li><strong>\u4ece GitHub \u5b89\u88c5\u6700\u65b0\u7248\uff08\u53ef\u9009\uff09<\/strong><br \/>\n\u83b7\u53d6\u5f00\u53d1\u7248\uff1a<\/li>\n<\/ol>\n<pre><code>git clone https:\/\/github.com\/DataScienceUIBK\/Rankify.git\r\ncd Rankify\r\npip install -e \".[all]\"\r\n<\/code><\/pre>\n<ol start=\"7\">\n<li><strong>\u5b89\u88c5 ColBERT \u68c0\u7d22\u5668\uff08\u53ef\u9009\uff09<\/strong><br \/>\n\u9700\u8981\u989d\u5916\u914d\u7f6e\uff1a<\/li>\n<\/ol>\n<pre><code>conda install -c conda-forge gcc=9.4.0 gxx=9.4.0\r\nconda install -c conda-forge libstdcxx-ng\r\nexport LD_LIBRARY_PATH=$CONDA_PREFIX\/lib:$LD_LIBRARY_PATH\r\nexport CC=gcc\r\nexport CXX=g++\r\nrm -rf ~\/.cache\/torch_extensions\/*\r\n<\/code><\/pre>\n<p>\u5b89\u88c5\u5b8c\u6210\u540e\u5373\u53ef\u4f7f\u7528 Rankify\u3002<\/p>\n<h3>\u529f\u80fd\u64cd\u4f5c\u6307\u5357<\/h3>\n<h4>1. \u4f7f\u7528\u9884\u68c0\u7d22\u6570\u636e\u96c6<\/h4>\n<p>Rankify \u63d0\u4f9b 40 \u4e2a\u9884\u68c0\u7d22\u6570\u636e\u96c6\uff0c\u53ef\u4ece Hugging Face \u4e0b\u8f7d\u3002<\/p>\n<ul>\n<li><strong>\u6b65\u9aa4<\/strong>\uff1a<\/li>\n<\/ul>\n<ol>\n<li>\u5bfc\u5165\u6570\u636e\u96c6\u6a21\u5757\u3002<\/li>\n<li>\u9009\u62e9\u68c0\u7d22\u5668\u548c\u6570\u636e\u96c6\u3002<\/li>\n<li>\u4e0b\u8f7d\u6216\u52a0\u8f7d\u6570\u636e\u3002<\/li>\n<\/ol>\n<ul>\n<li><strong>\u793a\u4f8b\u4ee3\u7801<\/strong>\uff1a<\/li>\n<\/ul>\n<pre><code>from rankify.dataset.dataset import Dataset\r\n# \u67e5\u770b\u53ef\u7528\u6570\u636e\u96c6\r\nDataset.available_dataset()\r\n# \u4e0b\u8f7d <a href=\"https:\/\/www.kdjingpai.com\/pt\/bm25\/\">BM25<\/a> \u7684 nq-dev \u6570\u636e\u96c6\r\ndataset = Dataset(retriever=\"bm25\", dataset_name=\"nq-dev\", n_docs=100)\r\ndocuments = dataset.download(force_download=False)\r\n# \u52a0\u8f7d\u672c\u5730\u6570\u636e\u96c6\r\ndocuments = Dataset.load_dataset('.\/bm25_nq_dev.json', 100)\r\n<\/code><\/pre>\n<h4>2. \u4f7f\u7528\u68c0\u7d22\u529f\u80fd<\/h4>\n<p>\u652f\u6301\u591a\u79cd\u68c0\u7d22\u65b9\u6cd5\uff0c\u5982 BM25\u3001DPR \u7b49\u3002<\/p>\n<ul>\n<li><strong>\u6b65\u9aa4<\/strong>\uff1a\n<ol>\n<li>\u521d\u59cb\u5316\u68c0\u7d22\u5668\u3002<\/li>\n<li>\u8f93\u5165\u6587\u6863\u6216\u95ee\u9898\u3002<\/li>\n<li>\u83b7\u53d6\u68c0\u7d22\u7ed3\u679c\u3002<\/li>\n<\/ol>\n<\/li>\n<li><strong>\u793a\u4f8b\u4ee3\u7801<\/strong>\uff1a\n<pre><code>from rankify.retrievers.retriever import Retriever\r\n# \u4f7f\u7528 BM25 \u68c0\u7d22 Wikipedia\r\nretriever = Retriever(method=\"bm25\", n_docs=5, index_type=\"wiki\")\r\ndocs = [{\"question\": \"\u592a\u9633\u662f\u4ec0\u4e48\uff1f\"}]\r\nresults = retriever.retrieve(docs)\r\nprint(results)\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<h4>3. \u4f7f\u7528\u91cd\u6392\u5e8f\u529f\u80fd<\/h4>\n<p>\u91cd\u6392\u5e8f\u4f18\u5316\u68c0\u7d22\u7ed3\u679c\uff0c\u652f\u6301\u591a\u79cd\u6a21\u578b\u3002<\/p>\n<ul>\n<li><strong>\u6b65\u9aa4<\/strong>\uff1a\n<ol>\n<li>\u51c6\u5907\u521d\u59cb\u68c0\u7d22\u7ed3\u679c\u3002<\/li>\n<li>\u521d\u59cb\u5316\u91cd\u6392\u5e8f\u6a21\u578b\u3002<\/li>\n<li>\u91cd\u65b0\u6392\u5e8f\u3002<\/li>\n<\/ol>\n<\/li>\n<li><strong>\u793a\u4f8b\u4ee3\u7801<\/strong>\uff1a\n<pre><code>from rankify.models.reranking import Reranking\r\nfrom rankify.dataset.dataset import Document, Question, Context\r\n# \u51c6\u5907\u6570\u636e\r\nquestion = Question(\"\u592a\u9633\u662f\u4ec0\u4e48\uff1f\")\r\ncontexts = [Context(text=\"\u592a\u9633\u662f\u6052\u661f\u3002\", id=1), Context(text=\"\u6708\u4eae\u4e0d\u662f\u6052\u661f\u3002\", id=2)]\r\ndoc = Document(question=question, contexts=contexts)\r\n# \u91cd\u6392\u5e8f\r\nreranker = Reranking(method=\"monot5\", model_name=\"monot5-base-msmarco\")\r\nreranker.rank([doc])\r\nfor ctx in doc.reorder_contexts:\r\nprint(ctx.text)\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<h4>4. \u4f7f\u7528 RAG \u529f\u80fd<\/h4>\n<p>RAG \u7ed3\u5408\u68c0\u7d22\u548c\u751f\u6210\uff0c\u751f\u6210\u51c6\u786e\u56de\u7b54\u3002<\/p>\n<ul>\n<li><strong>\u6b65\u9aa4<\/strong>\uff1a\n<ol>\n<li>\u51c6\u5907\u6587\u6863\u548c\u95ee\u9898\u3002<\/li>\n<li>\u521d\u59cb\u5316\u751f\u6210\u5668\u3002<\/li>\n<li>\u751f\u6210\u56de\u7b54\u3002<\/li>\n<\/ol>\n<\/li>\n<li><strong>\u793a\u4f8b\u4ee3\u7801<\/strong>\uff1a\n<pre><code>from rankify.generator.generator import Generator\r\ndoc = Document(question=Question(\"\u6cd5\u56fd\u9996\u90fd\u662f\u4ec0\u4e48\uff1f\"), contexts=[Context(text=\"\u6cd5\u56fd\u9996\u90fd\u662f\u5df4\u9ece\u3002\", id=1)])\r\ngenerator = Generator(method=\"in-context-ralm\", model_name=\"meta-llama\/Llama-3.1-8B\")\r\nanswers = generator.generate([doc])\r\nprint(answers)  # \u8f93\u51fa\uff1a[\"\u5df4\u9ece\"]\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<h4>5. \u8bc4\u4f30\u7ed3\u679c<\/h4>\n<p>\u5185\u7f6e\u8bc4\u4f30\u5de5\u5177\uff0c\u68c0\u67e5\u6027\u80fd\u3002<\/p>\n<ul>\n<li><strong>\u793a\u4f8b\u4ee3\u7801<\/strong>\uff1a\n<pre><code>from rankify.metrics.metrics import Metrics\r\nmetrics = Metrics(documents)\r\nretrieval_metrics = metrics.calculate_retrieval_metrics(ks=[1, 5, 10])\r\nprint(retrieval_metrics)\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<h3>\u6ce8\u610f\u4e8b\u9879<\/h3>\n<ul>\n<li>GPU \u7528\u6237\u9700\u786e\u4fdd PyTorch \u652f\u6301 CUDA\u3002<\/li>\n<li>\u5927\u6570\u636e\u96c6\u5efa\u8bae\u4f7f\u7528\u9ad8\u5185\u5b58\u8bbe\u5907\u3002<\/li>\n<li>\u66f4\u591a\u8be6\u60c5\u89c1\u5b98\u65b9\u6587\u6863\uff1ahttp:\/\/rankify.readthedocs.io\/\u3002<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2>\u5e94\u7528\u573a\u666f<\/h2>\n<ol>\n<li><strong>\u5b66\u672f\u7814\u7a76<\/strong><br \/>\n\u7814\u7a76\u4eba\u5458\u53ef\u4ee5\u7528 Rankify \u6d4b\u8bd5\u68c0\u7d22\u548c\u91cd\u6392\u5e8f\u7b97\u6cd5\uff0c\u5206\u6790\u6027\u80fd\u3002<\/li>\n<li><strong>\u667a\u80fd\u95ee\u7b54<\/strong><br \/>\n\u5f00\u53d1\u8005\u53ef\u5229\u7528 RAG \u6784\u5efa\u95ee\u7b54\u7cfb\u7edf\uff0c\u56de\u7b54\u7528\u6237\u63d0\u95ee\u3002<\/li>\n<li><strong>\u641c\u7d22\u4f18\u5316<\/strong><br \/>\n\u91cd\u6392\u5e8f\u529f\u80fd\u53ef\u63d0\u5347\u641c\u7d22\u7ed3\u679c\u76f8\u5173\u6027\uff0c\u9002\u5408\u6539\u8fdb\u641c\u7d22\u5f15\u64ce\u3002<\/li>\n<\/ol>\n<p>&nbsp;<\/p>\n<h2>QA<\/h2>\n<ol>\n<li><strong>Rankify \u652f\u6301\u54ea\u4e9b\u7cfb\u7edf\uff1f<\/strong><br \/>\n\u652f\u6301 Windows\u3001Linux \u548c macOS\uff0c\u53ea\u8981\u5b89\u88c5 Python 3.10+ \u5373\u53ef\u3002<\/li>\n<li><strong>\u9700\u8981\u8054\u7f51\u5417\uff1f<\/strong><br \/>\n\u6838\u5fc3\u529f\u80fd\u79bb\u7ebf\u53ef\u7528\uff0c\u4f46\u6570\u636e\u96c6\u548c\u90e8\u5206\u6a21\u578b\u9700\u4e0b\u8f7d\u3002<\/li>\n<li><strong>\u652f\u6301\u4e2d\u6587\u5417\uff1f<\/strong><br \/>\n\u652f\u6301\uff0c\u4f46\u9884\u6784\u5efa\u7d22\u5f15\u4e3b\u8981\u4e3a\u82f1\u6587\uff08Wikipedia \u548c MS MARCO\uff09\u3002<\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"<p>Rankify \u662f\u7531\u5965\u5730\u5229\u56e0\u65af\u5e03\u9c81\u514b\u5927\u5b66\u6570\u636e\u79d1\u5b66\u5c0f\u7ec4\u5f00\u53d1\u7684\u5f00\u6e90 Python \u5de5\u5177\u5305\u3002\u5b83\u4e13\u6ce8\u4e8e\u4fe1\u606f\u68c0\u7d22\u3001\u91cd\u6392\u5e8f\u548c\u68c0\u7d22\u589e\u5f3a\u751f\u6210\uff08RAG\uff09\uff0c\u63d0\u4f9b\u4e86\u4e00\u4e2a\u7edf\u4e00\u7684\u6846\u67b6\u3002\u5de5\u5177\u5305\u5185\u7f6e 40 \u4e2a\u9884\u68c0\u7d22\u57fa\u51c6\u6570\u636e\u96c6\uff0c\u652f\u6301 7 \u79cd\u68c0\u7d22\u6280\u672f\u548c 24 \u79cd\u91cd\u6392\u5e8f\u6a21\u578b\uff0c&#8230;<\/p>\n","protected":false},"author":1,"featured_media":62114,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[425,20,459],"tags":[230,243],"class_list":["post-29401","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-professional","category-tool","category-rag-project","tag-aikaiyuanxiangmu","tag-aizhishikuyukefu"],"_links":{"self":[{"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/posts\/29401","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/comments?post=29401"}],"version-history":[{"count":0,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/posts\/29401\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/media\/62114"}],"wp:attachment":[{"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/media?parent=29401"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/categories?post=29401"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/ja\/wp-json\/wp\/v2\/tags?post=29401"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}