{"id":388,"date":"2023-08-29T14:34:28","date_gmt":"2023-10-27T10:23:31","guid":{"rendered":"https:\/\/www.silicloud.com\/zh\/blog\/index.php\/2023\/11\/30\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/"},"modified":"2025-08-02T03:44:52","modified_gmt":"2025-08-01T19:44:52","slug":"%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f","status":"publish","type":"post","link":"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/","title":{"rendered":"Python\u8ba1\u7b97BLEU\u5206\u6570\u5b8c\u5168\u6307\u5357\uff1aNLP\u6a21\u578b\u8bc4\u4f30\u6559\u7a0b"},"content":{"rendered":"<p>\u5728Python\u4e2d\uff0cBLEU\u5206\u6570\u662f\u4e00\u79cd\u7528\u4e8e\u8861\u91cf\u673a\u5668\u7ffb\u8bd1\u6a21\u578b\u4f18\u52a3\u7684\u6307\u6807\u3002\u5c3d\u7ba1\u6700\u521d\u5b83\u53ea\u8bbe\u8ba1\u7528\u4e8e\u7ffb\u8bd1\u6a21\u578b\uff0c\u4f46\u73b0\u5728\u5b83\u4e5f\u88ab\u7528\u4e8e\u5176\u4ed6\u81ea\u7136\u8bed\u8a00\u5904\u7406\u5e94\u7528\u3002<\/p>\n<p>BLEU\u5206\u6570\u662f\u5c06\u4e00\u4e2a\u53e5\u5b50\u4e0e\u4e00\u4e2a\u6216\u591a\u4e2a\u53c2\u8003\u53e5\u5b50\u8fdb\u884c\u6bd4\u8f83\uff0c\u544a\u8bc9\u6211\u4eec\u5019\u9009\u53e5\u5b50\u4e0e\u53c2\u8003\u53e5\u5b50\u5217\u8868\u7684\u5339\u914d\u7a0b\u5ea6\u3002\u5b83\u7ed9\u51fa\u4e00\u4e2a\u4ecb\u4e8e0\u52301\u4e4b\u95f4\u7684\u5206\u6570\u3002<\/p>\n<p>BLEU\u5f97\u5206\u4e3a1\u8868\u793a\u5019\u9009\u53e5\u5b50\u4e0e\u53c2\u8003\u53e5\u5b50\u4e2d\u7684\u4e00\u4e2a\u5b8c\u5168\u5339\u914d\u3002<\/p>\n<p>\u8fd9\u4e2a\u5206\u6570\u662f\u5bf9\u56fe\u50cf\u63cf\u8ff0\u6a21\u578b\u5e38\u89c1\u7684\u5ea6\u91cf\u65b9\u6cd5\u3002<\/p>\n<p>\u5728\u672c\u6559\u7a0b\u4e2d\uff0c\u6211\u4eec\u5c06\u4f7f\u7528nltk\u5e93\u4e2d\u7684sentence_bleu()\u51fd\u6570\u3002\u8ba9\u6211\u4eec\u5f00\u59cb\u5427\u3002<\/p>\n<h2>\u5728Python\u4e2d\u8ba1\u7b97BLEU\u5206\u6570<\/h2>\n<p>\u4e3a\u4e86\u8ba1\u7b97BLEU\u5206\u6570\uff0c\u6211\u4eec\u9700\u8981\u4ee5\u6807\u8bb0\uff08token\uff09\u7684\u5f62\u5f0f\u63d0\u4f9b\u53c2\u8003\u53e5\u5b50\u548c\u5019\u9009\u53e5\u5b50\u3002<\/p>\n<p>\u5728\u672c\u8282\u4e2d\uff0c\u6211\u4eec\u5c06\u5b66\u4e60\u5982\u4f55\u505a\u5230\u8fd9\u4e00\u70b9\u5e76\u8ba1\u7b97\u5f97\u5206\u3002\u8ba9\u6211\u4eec\u4ece\u5bfc\u5165\u5fc5\u8981\u7684\u6a21\u5757\u5f00\u59cb\u3002<\/p>\n<pre class=\"post-pre\"><code><span class=\"token keyword\">from<\/span> nltk<span class=\"token punctuation\">.<\/span>translate<span class=\"token punctuation\">.<\/span>bleu_score <span class=\"token keyword\">import<\/span> sentence_bleu\r\n<\/code><\/pre>\n<p>\u73b0\u5728\u6211\u4eec\u53ef\u4ee5\u4ee5\u5217\u8868\u5f62\u5f0f\u8f93\u5165\u53c2\u8003\u53e5\u5b50\u3002\u5728\u5c06\u53e5\u5b50\u4f20\u9012\u5230sentence_bleu()\u51fd\u6570\u4e4b\u524d\uff0c\u8fd8\u9700\u8981\u5c06\u5176\u8f6c\u6362\u4e3a\u6807\u8bb0\u3002<\/p>\n<h3>1. \u8f93\u5165\u5e76\u62c6\u5206\u53e5\u5b50<\/h3>\n<p>\u6211\u4eec\u7684\u53c2\u8003\u6587\u732e\u6e05\u5355\u4e2d\u7684\u53e5\u5b50\u4e3a\uff1a<\/p>\n<pre class=\"post-pre\"><code>    'this is a dog'\r\n    'it is dog'\r\n    'dog it is'\r\n    'a dog, it is'\r\n<\/code><\/pre>\n<p>\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528\u5206\u5272\u51fd\u6570\u5c06\u5b83\u4eec\u5206\u6210\u51e0\u4e2a\u7247\u6bb5\u3002<\/p>\n<pre class=\"post-pre\"><code>reference <span class=\"token operator\">=<\/span> <span class=\"token punctuation\">[<\/span>\r\n    <span class=\"token string\">'this is a dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span>\r\n    <span class=\"token string\">'it is dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span>\r\n    <span class=\"token string\">'dog it is'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span>\r\n    <span class=\"token string\">'a dog, it is'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span> \r\n<span class=\"token punctuation\">]<\/span>\r\n<span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span>reference<span class=\"token punctuation\">)<\/span>\r\n<\/code><\/pre>\n<p>\u8f93\u51fa\uff1a<\/p>\n<pre class=\"post-pre\"><code>[['this', 'is', 'a', 'dog'], ['it', 'is', 'dog'], ['dog', 'it', 'is'], ['a', 'dog,', 'it', 'is']]\r\n<\/code><\/pre>\n<p>\u8fd9\u662f\u53e5\u5b50\u4ee5\u6807\u8bb0\u5f62\u5f0f\u7684\u6837\u5b50\u3002\u73b0\u5728\u6211\u4eec\u53ef\u4ee5\u8c03\u7528sentence_bleu()\u51fd\u6570\u6765\u8ba1\u7b97\u5f97\u5206\u3002<\/p>\n<h3>2. \u5728Python\u4e2d\u8ba1\u7b97BLEU\u5206\u6570<\/h3>\n<p>\u4f7f\u7528\u4e0b\u9762\u7684\u4ee3\u7801\u6765\u8ba1\u7b97\u5206\u6570\u3002<\/p>\n<pre class=\"post-pre\"><code>candidate <span class=\"token operator\">=<\/span> <span class=\"token string\">'it is dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span>\r\n<span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string\">'BLEU score -&gt; {}'<\/span><span class=\"token punctuation\">.<\/span><span class=\"token builtin\">format<\/span><span class=\"token punctuation\">(<\/span>sentence_bleu<span class=\"token punctuation\">(<\/span>reference<span class=\"token punctuation\">,<\/span> candidate<span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span>\r\n<\/code><\/pre>\n<p>\u8f93\u51fa\uff1a<\/p>\n<pre class=\"post-pre\"><code>BLEU score -&gt; 1.0\r\n<\/code><\/pre>\n<p>\u6211\u4eec\u7ed9\u8fd9\u53e5\u8bdd\u6253\u51fa\u4e86\u6ee1\u52061\u5206\uff0c\u56e0\u4e3a\u5019\u9009\u53e5\u5b50\u5c5e\u4e8e\u53c2\u8003\u96c6\u5408\u3002\u6211\u4eec\u518d\u8bd5\u4e00\u53e5\u3002<\/p>\n<pre class=\"post-pre\"><code>candidate <span class=\"token operator\">=<\/span> <span class=\"token string\">'it is a dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span>\r\n<span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string\">'BLEU score -&gt; {}'<\/span><span class=\"token punctuation\">.<\/span><span class=\"token builtin\">format<\/span><span class=\"token punctuation\">(<\/span>sentence_bleu<span class=\"token punctuation\">(<\/span>reference<span class=\"token punctuation\">,<\/span> candidate<span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span>\r\n<\/code><\/pre>\n<p>\u8f93\u51fa\uff1a<\/p>\n<pre class=\"post-pre\"><code>BLEU score -&gt; 0.8408964152537145\r\n<\/code><\/pre>\n<p>\u6211\u4eec\u5728\u53c2\u8003\u96c6\u4e2d\u6709\u8fd9\u4e2a\u53e5\u5b50\uff0c\u4f46\u5b83\u5e76\u4e0d\u5b8c\u5168\u5339\u914d\u3002\u8fd9\u5c31\u662f\u4e3a\u4ec0\u4e48\u6211\u4eec\u5f97\u5230\u4e860.84\u5206\u7684\u539f\u56e0\u3002<\/p>\n<h3>3. Python\u5b8c\u6574\u4ee3\u7801\u7528\u4e8e\u5b9e\u73b0BLEU\u5206\u6570<\/h3>\n<h3>4. \u8ba1\u7b97BLEU\u5206\u6570\u7684n-gram\u5f97\u5206<\/h3>\n<p>\u4ee5\u4e0b\u662f\u8ba1\u7b97BLEU\u5206\u6570\u7684\u5b8c\u6574\u4ee3\u7801\u793a\u4f8b\uff1a<\/p>\n<pre class=\"post-pre\"><code><span class=\"token keyword\">from<\/span> nltk<span class=\"token punctuation\">.<\/span>translate<span class=\"token punctuation\">.<\/span>bleu_score <span class=\"token keyword\">import<\/span> sentence_bleu\r\nreference <span class=\"token operator\">=<\/span> <span class=\"token punctuation\">[<\/span>\r\n    <span class=\"token string\">'this is a dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span>\r\n    <span class=\"token string\">'it is dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span>\r\n    <span class=\"token string\">'dog it is'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span>\r\n    <span class=\"token string\">'a dog, it is'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span> \r\n<span class=\"token punctuation\">]<\/span>\r\ncandidate <span class=\"token operator\">=<\/span> <span class=\"token string\">'it is dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span>\r\n<span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string\">'BLEU score -&gt; {}'<\/span><span class=\"token punctuation\">.<\/span><span class=\"token builtin\">format<\/span><span class=\"token punctuation\">(<\/span>sentence_bleu<span class=\"token punctuation\">(<\/span>reference<span class=\"token punctuation\">,<\/span> candidate <span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span>\r\n\r\ncandidate <span class=\"token operator\">=<\/span> <span class=\"token string\">'it is a dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span>\r\n<span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string\">'BLEU score -&gt; {}'<\/span><span class=\"token punctuation\">.<\/span><span class=\"token builtin\">format<\/span><span class=\"token punctuation\">(<\/span>sentence_bleu<span class=\"token punctuation\">(<\/span>reference<span class=\"token punctuation\">,<\/span> candidate<span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span>\r\n<\/code><\/pre>\n<p>\u5728\u5339\u914d\u53e5\u5b50\u65f6\uff0c\u60a8\u53ef\u4ee5\u9009\u62e9\u6a21\u578b\u4e00\u6b21\u5339\u914d\u7684\u8bcd\u8bed\u6570\u91cf\u3002\u4f8b\u5982\uff0c\u60a8\u53ef\u4ee5\u9009\u62e9\u4e00\u6b21\u5339\u914d\u4e00\u4e2a\u8bcd\u8bed\uff081-gram\uff09\u3002\u6216\u8005\uff0c\u60a8\u4e5f\u53ef\u4ee5\u9009\u62e9\u4e00\u6b21\u5339\u914d\u8bcd\u8bed\u5bf9\uff082-gram\uff09\u6216\u4e09\u4e2a\u8bcd\u8bed\uff083-gram\uff09\u3002<\/p>\n<p>\u5728\u672c\u8282\u4e2d\uff0c\u6211\u4eec\u5c06\u5b66\u4e60\u5982\u4f55\u8ba1\u7b97\u8fd9\u4e9bn-gram\u5206\u6570\u3002<\/p>\n<p>\u5728sentence_bleu()\u51fd\u6570\u4e2d\uff0c\u60a8\u53ef\u4ee5\u4f20\u9012\u4e00\u4e2a\u53c2\u6570\uff0c\u5176\u4e2d\u5305\u542b\u4e0e\u5404\u4e2an-gram\u5bf9\u5e94\u7684\u6743\u91cd\u3002<\/p>\n<p>\u4f8b\u5982\uff0c\u8981\u5355\u72ec\u8ba1\u7b97n-gram\u5206\u6570\uff0c\u60a8\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u6743\u91cd\u3002<\/p>\n<pre class=\"post-pre\"><code>\u5355\u4e2a1-gram: (1, 0, 0, 0)\r\n\u5355\u4e2a2-gram: (0, 1, 0, 0). \r\n\u5355\u4e2a3-gram: (0, 0, 1, 0). \r\n\u5355\u4e2a4-gram: (0, 0, 0, 1). \r\n<\/code><\/pre>\n<p>\u4ee5\u4e0b\u662f\u5bf9\u5e94\u7684Python\u4ee3\u7801\uff1a<\/p>\n<pre class=\"post-pre\"><code><span class=\"token keyword\">from<\/span> nltk<span class=\"token punctuation\">.<\/span>translate<span class=\"token punctuation\">.<\/span>bleu_score <span class=\"token keyword\">import<\/span> sentence_bleu\r\nreference <span class=\"token operator\">=<\/span> <span class=\"token punctuation\">[<\/span>\r\n    <span class=\"token string\">'this is a dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span>\r\n    <span class=\"token string\">'it is dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span>\r\n    <span class=\"token string\">'dog it is'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">,<\/span>\r\n    <span class=\"token string\">'a dog, it is'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span> \r\n<span class=\"token punctuation\">]<\/span>\r\ncandidate <span class=\"token operator\">=<\/span> <span class=\"token string\">'it is a dog'<\/span><span class=\"token punctuation\">.<\/span>split<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span>\r\n\r\n<span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string\">'\u5355\u4e2a1-gram: %f'<\/span> <span class=\"token operator\">%<\/span> sentence_bleu<span class=\"token punctuation\">(<\/span>reference<span class=\"token punctuation\">,<\/span> candidate<span class=\"token punctuation\">,<\/span> weights<span class=\"token operator\">=<\/span><span class=\"token punctuation\">(<\/span><span class=\"token number\">1<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span>\r\n<span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string\">'\u5355\u4e2a2-gram: %f'<\/span> <span class=\"token operator\">%<\/span> sentence_bleu<span class=\"token punctuation\">(<\/span>reference<span class=\"token punctuation\">,<\/span> candidate<span class=\"token punctuation\">,<\/span> weights<span class=\"token operator\">=<\/span><span class=\"token punctuation\">(<\/span><span class=\"token number\">0<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">1<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span>\r\n<span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string\">'\u5355\u4e2a3-gram: %f'<\/span> <span class=\"token operator\">%<\/span> sentence_bleu<span class=\"token punctuation\">(<\/span>reference<span class=\"token punctuation\">,<\/span> candidate<span class=\"token punctuation\">,<\/span> weights<span class=\"token operator\">=<\/span><span class=\"token punctuation\">(<\/span><span class=\"token number\">0<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">1<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span>\r\n<span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span><span class=\"token string\">'\u5355\u4e2a4-gram: %f'<\/span> <span class=\"token operator\">%<\/span> sentence_bleu<span class=\"token punctuation\">(<\/span>reference<span class=\"token punctuation\">,<\/span> candidate<span class=\"token punctuation\">,<\/span> weights<span class=\"token operator\">=<\/span><span class=\"token punctuation\">(<\/span><span class=\"token number\">0<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">1<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span>\r\n\r\n<\/code><\/pre>\n<p>\u8f93\u51fa\u7ed3\u679c\uff1a<\/p>\n<pre class=\"post-pre\"><code>\u5355\u4e2a1-gram: 1.000000\r\n\u5355\u4e2a2-gram: 1.000000\r\n\u5355\u4e2a3-gram: 0.500000\r\n\u5355\u4e2a4-gram: 1.000000\r\n<\/code><\/pre>\n<p>\u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0csentence_bleu()\u51fd\u6570\u8ba1\u7b97\u7d2f\u79ef4-gram BLEU\u5206\u6570\uff0c\u4e5f\u79f0\u4e3aBLEU-4\u3002BLEU-4\u7684\u6743\u91cd\u5982\u4e0b\u6240\u793a\u3002<\/p>\n<pre class=\"post-pre\"><code>(0.25, 0.25, 0.25, 0.25)\r\n<\/code><\/pre>\n<p>\u6211\u4eec\u6765\u770b\u4e00\u4e0bBLEU-4\u7684\u4ee3\u7801\uff1a<\/p>\n<pre class=\"post-pre\"><code>score <span class=\"token operator\">=<\/span> sentence_bleu<span class=\"token punctuation\">(<\/span>reference<span class=\"token punctuation\">,<\/span> candidate<span class=\"token punctuation\">,<\/span> weights<span class=\"token operator\">=<\/span><span class=\"token punctuation\">(<\/span><span class=\"token number\">0.25<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0.25<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0.25<\/span><span class=\"token punctuation\">,<\/span> <span class=\"token number\">0.25<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">)<\/span>\r\n<span class=\"token keyword\">print<\/span><span class=\"token punctuation\">(<\/span>score<span class=\"token punctuation\">)<\/span>\r\n<\/code><\/pre>\n<p>\u8f93\u51fa\u7ed3\u679c\uff1a<\/p>\n<pre class=\"post-pre\"><code>0.8408964152537145\r\n<\/code><\/pre>\n<p>\u8fd9\u5c31\u662f\u6211\u4eec\u6ca1\u6709\u6307\u5b9an-gram\u6743\u91cd\u65f6\u5f97\u5230\u7684\u51c6\u786e\u5206\u6570\u3002<\/p>\n<h2>\u7ed3\u8bba<\/h2>\n<p>\u672c\u6559\u7a0b\u8be6\u7ec6\u4ecb\u7ecd\u4e86\u5982\u4f55\u5728Python\u4e2d\u8ba1\u7b97BLEU\u5206\u6570\uff08\u53cc\u8bed\u8bc4\u4f30\u66ff\u6362\u5206\u6570\uff09\u3002\u6211\u4eec\u5b66\u4e60\u4e86BLEU\u5206\u6570\u7684\u57fa\u672c\u6982\u5ff5\uff0c\u4ee5\u53ca\u5982\u4f55\u8ba1\u7b97\u5355\u4e2an\u5143\u8bed\u6cd5\uff08n-gram\uff09\u548c\u7d2f\u79ef\u7684n\u5143\u8bed\u6cd5BLEU\u5206\u6570\u3002\u8fd9\u4e9b\u77e5\u8bc6\u5bf9\u4e8e\u81ea\u7136\u8bed\u8a00\u5904\u7406\u548c\u673a\u5668\u7ffb\u8bd1\u9886\u57df\u7684\u8bc4\u4f30\u5de5\u4f5c\u975e\u5e38\u91cd\u8981\u3002\u5e0c\u671b\u60a8\u5728\u5b66\u4e60\u672c\u6559\u7a0b\u7684\u8fc7\u7a0b\u4e2d\u6536\u83b7\u6ee1\u6ee1\uff0c\u80fd\u591f\u5c06\u8fd9\u4e9b\u6280\u672f\u5e94\u7528\u5230\u5b9e\u9645\u9879\u76ee\u4e2d\uff01<\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u5728Python\u4e2d\uff0cBLEU\u5206\u6570\u662f\u4e00\u79cd\u7528\u4e8e\u8861\u91cf\u673a\u5668\u7ffb\u8bd1\u6a21\u578b\u4f18\u52a3\u7684\u6307\u6807\u3002\u5c3d\u7ba1\u6700\u521d\u5b83\u53ea\u8bbe\u8ba1\u7528\u4e8e\u7ffb\u8bd1\u6a21\u578b\uff0c\u4f46\u73b0\u5728\u5b83\u4e5f\u88ab [&hellip;]<\/p>\n","protected":false},"author":6,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[1260,1263,32,1262,1261],"class_list":["post-388","post","type-post","status-publish","format-standard","hentry","category-uncategorized","tag-bleu","tag-nlp","tag-python","tag-1262","tag-1261"],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v21.5 (Yoast SEO v21.5) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Python\u8ba1\u7b97BLEU\u5206\u6570\u5b8c\u5168\u6307\u5357\uff1aNLP\u6a21\u578b\u8bc4\u4f30\u6559\u7a0b - Blog - Silicon Cloud<\/title>\n<meta name=\"description\" content=\"\u672c\u6587\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528Python\u8ba1\u7b97BLEU\u5206\u6570\uff0c\u5305\u62ecBLEU\u7b97\u6cd5\u539f\u7406\u3001NLTK\u5e93\u5b9e\u73b0\u65b9\u6cd5\u53ca\u4ee3\u7801\u793a\u4f8b\u3002\u901a\u8fc7\u672c\u6559\u7a0b\uff0c\u60a8\u5c06\u638c\u63e1NLP\u6a21\u578b\u8bc4\u4f30\u7684\u5173\u952e\u6280\u672f\uff0c\u8f7b\u677e\u8bc4\u4f30\u673a\u5668\u7ffb\u8bd1\u8d28\u91cf\u3002\u9002\u5408\u81ea\u7136\u8bed\u8a00\u5904\u7406\u521d\u5b66\u8005\u548c\u7814\u7a76\u4eba\u5458\u5b66\u4e60\u53c2\u8003\u3002\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.silicloud.com\/zh\/blog\/\u5982\u4f55\u4f7f\u7528python\u8ba1\u7b97bleu\u5206\u6570\uff1f\/\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Python\u8ba1\u7b97BLEU\u5206\u6570\u5b8c\u5168\u6307\u5357\uff1aNLP\u6a21\u578b\u8bc4\u4f30\u6559\u7a0b\" \/>\n<meta property=\"og:description\" content=\"\u672c\u6587\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528Python\u8ba1\u7b97BLEU\u5206\u6570\uff0c\u5305\u62ecBLEU\u7b97\u6cd5\u539f\u7406\u3001NLTK\u5e93\u5b9e\u73b0\u65b9\u6cd5\u53ca\u4ee3\u7801\u793a\u4f8b\u3002\u901a\u8fc7\u672c\u6559\u7a0b\uff0c\u60a8\u5c06\u638c\u63e1NLP\u6a21\u578b\u8bc4\u4f30\u7684\u5173\u952e\u6280\u672f\uff0c\u8f7b\u677e\u8bc4\u4f30\u673a\u5668\u7ffb\u8bd1\u8d28\u91cf\u3002\u9002\u5408\u81ea\u7136\u8bed\u8a00\u5904\u7406\u521d\u5b66\u8005\u548c\u7814\u7a76\u4eba\u5458\u5b66\u4e60\u53c2\u8003\u3002\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.silicloud.com\/zh\/blog\/\u5982\u4f55\u4f7f\u7528python\u8ba1\u7b97bleu\u5206\u6570\uff1f\/\" \/>\n<meta property=\"og:site_name\" content=\"Blog - Silicon Cloud\" \/>\n<meta property=\"article:published_time\" content=\"2023-10-27T10:23:31+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-08-01T19:44:52+00:00\" \/>\n<meta name=\"author\" content=\"\u6587, \u7fd4\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"\u6587, \u7fd4\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"1 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/\",\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/\",\"name\":\"Python\u8ba1\u7b97BLEU\u5206\u6570\u5b8c\u5168\u6307\u5357\uff1aNLP\u6a21\u578b\u8bc4\u4f30\u6559\u7a0b - Blog - Silicon Cloud\",\"isPartOf\":{\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#website\"},\"datePublished\":\"2023-10-27T10:23:31+00:00\",\"dateModified\":\"2025-08-01T19:44:52+00:00\",\"author\":{\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/64d5cc7727fffbff2f9a2a8da1de3e5c\"},\"description\":\"\u672c\u6587\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528Python\u8ba1\u7b97BLEU\u5206\u6570\uff0c\u5305\u62ecBLEU\u7b97\u6cd5\u539f\u7406\u3001NLTK\u5e93\u5b9e\u73b0\u65b9\u6cd5\u53ca\u4ee3\u7801\u793a\u4f8b\u3002\u901a\u8fc7\u672c\u6559\u7a0b\uff0c\u60a8\u5c06\u638c\u63e1NLP\u6a21\u578b\u8bc4\u4f30\u7684\u5173\u952e\u6280\u672f\uff0c\u8f7b\u677e\u8bc4\u4f30\u673a\u5668\u7ffb\u8bd1\u8d28\u91cf\u3002\u9002\u5408\u81ea\u7136\u8bed\u8a00\u5904\u7406\u521d\u5b66\u8005\u548c\u7814\u7a76\u4eba\u5458\u5b66\u4e60\u53c2\u8003\u3002\",\"breadcrumb\":{\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\/\/www.silicloud.com\/zh\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Python\u8ba1\u7b97BLEU\u5206\u6570\u5b8c\u5168\u6307\u5357\uff1aNLP\u6a21\u578b\u8bc4\u4f30\u6559\u7a0b\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#website\",\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/\",\"name\":\"Blog - Silicon Cloud\",\"description\":\"\",\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/64d5cc7727fffbff2f9a2a8da1de3e5c\",\"name\":\"\u6587, \u7fd4\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/920c3d673e0bccacc98e5e6b7149bb3c22edd8d39cb753e5d7d7e471498118a1?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/920c3d673e0bccacc98e5e6b7149bb3c22edd8d39cb753e5d7d7e471498118a1?s=96&d=mm&r=g\",\"caption\":\"\u6587, \u7fd4\"},\"url\":\"https:\/\/www.silicloud.com\/zh\/blog\/author\/wenxiang\/\"},{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/#local-main-organization-logo\",\"url\":\"\",\"contentUrl\":\"\",\"caption\":\"Blog - Silicon Cloud\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Python\u8ba1\u7b97BLEU\u5206\u6570\u5b8c\u5168\u6307\u5357\uff1aNLP\u6a21\u578b\u8bc4\u4f30\u6559\u7a0b - Blog - Silicon Cloud","description":"\u672c\u6587\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528Python\u8ba1\u7b97BLEU\u5206\u6570\uff0c\u5305\u62ecBLEU\u7b97\u6cd5\u539f\u7406\u3001NLTK\u5e93\u5b9e\u73b0\u65b9\u6cd5\u53ca\u4ee3\u7801\u793a\u4f8b\u3002\u901a\u8fc7\u672c\u6559\u7a0b\uff0c\u60a8\u5c06\u638c\u63e1NLP\u6a21\u578b\u8bc4\u4f30\u7684\u5173\u952e\u6280\u672f\uff0c\u8f7b\u677e\u8bc4\u4f30\u673a\u5668\u7ffb\u8bd1\u8d28\u91cf\u3002\u9002\u5408\u81ea\u7136\u8bed\u8a00\u5904\u7406\u521d\u5b66\u8005\u548c\u7814\u7a76\u4eba\u5458\u5b66\u4e60\u53c2\u8003\u3002","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:\/\/www.silicloud.com\/zh\/blog\/\u5982\u4f55\u4f7f\u7528python\u8ba1\u7b97bleu\u5206\u6570\uff1f\/","og_locale":"zh_CN","og_type":"article","og_title":"Python\u8ba1\u7b97BLEU\u5206\u6570\u5b8c\u5168\u6307\u5357\uff1aNLP\u6a21\u578b\u8bc4\u4f30\u6559\u7a0b","og_description":"\u672c\u6587\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528Python\u8ba1\u7b97BLEU\u5206\u6570\uff0c\u5305\u62ecBLEU\u7b97\u6cd5\u539f\u7406\u3001NLTK\u5e93\u5b9e\u73b0\u65b9\u6cd5\u53ca\u4ee3\u7801\u793a\u4f8b\u3002\u901a\u8fc7\u672c\u6559\u7a0b\uff0c\u60a8\u5c06\u638c\u63e1NLP\u6a21\u578b\u8bc4\u4f30\u7684\u5173\u952e\u6280\u672f\uff0c\u8f7b\u677e\u8bc4\u4f30\u673a\u5668\u7ffb\u8bd1\u8d28\u91cf\u3002\u9002\u5408\u81ea\u7136\u8bed\u8a00\u5904\u7406\u521d\u5b66\u8005\u548c\u7814\u7a76\u4eba\u5458\u5b66\u4e60\u53c2\u8003\u3002","og_url":"https:\/\/www.silicloud.com\/zh\/blog\/\u5982\u4f55\u4f7f\u7528python\u8ba1\u7b97bleu\u5206\u6570\uff1f\/","og_site_name":"Blog - Silicon Cloud","article_published_time":"2023-10-27T10:23:31+00:00","article_modified_time":"2025-08-01T19:44:52+00:00","author":"\u6587, \u7fd4","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"\u6587, \u7fd4","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"1 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/","url":"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/","name":"Python\u8ba1\u7b97BLEU\u5206\u6570\u5b8c\u5168\u6307\u5357\uff1aNLP\u6a21\u578b\u8bc4\u4f30\u6559\u7a0b - Blog - Silicon Cloud","isPartOf":{"@id":"https:\/\/www.silicloud.com\/zh\/blog\/#website"},"datePublished":"2023-10-27T10:23:31+00:00","dateModified":"2025-08-01T19:44:52+00:00","author":{"@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/64d5cc7727fffbff2f9a2a8da1de3e5c"},"description":"\u672c\u6587\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528Python\u8ba1\u7b97BLEU\u5206\u6570\uff0c\u5305\u62ecBLEU\u7b97\u6cd5\u539f\u7406\u3001NLTK\u5e93\u5b9e\u73b0\u65b9\u6cd5\u53ca\u4ee3\u7801\u793a\u4f8b\u3002\u901a\u8fc7\u672c\u6559\u7a0b\uff0c\u60a8\u5c06\u638c\u63e1NLP\u6a21\u578b\u8bc4\u4f30\u7684\u5173\u952e\u6280\u672f\uff0c\u8f7b\u677e\u8bc4\u4f30\u673a\u5668\u7ffb\u8bd1\u8d28\u91cf\u3002\u9002\u5408\u81ea\u7136\u8bed\u8a00\u5904\u7406\u521d\u5b66\u8005\u548c\u7814\u7a76\u4eba\u5458\u5b66\u4e60\u53c2\u8003\u3002","breadcrumb":{"@id":"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.silicloud.com\/zh\/blog\/"},{"@type":"ListItem","position":2,"name":"Python\u8ba1\u7b97BLEU\u5206\u6570\u5b8c\u5168\u6307\u5357\uff1aNLP\u6a21\u578b\u8bc4\u4f30\u6559\u7a0b"}]},{"@type":"WebSite","@id":"https:\/\/www.silicloud.com\/zh\/blog\/#website","url":"https:\/\/www.silicloud.com\/zh\/blog\/","name":"Blog - Silicon Cloud","description":"","inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/64d5cc7727fffbff2f9a2a8da1de3e5c","name":"\u6587, \u7fd4","image":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.silicloud.com\/zh\/blog\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/920c3d673e0bccacc98e5e6b7149bb3c22edd8d39cb753e5d7d7e471498118a1?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/920c3d673e0bccacc98e5e6b7149bb3c22edd8d39cb753e5d7d7e471498118a1?s=96&d=mm&r=g","caption":"\u6587, \u7fd4"},"url":"https:\/\/www.silicloud.com\/zh\/blog\/author\/wenxiang\/"},{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.silicloud.com\/zh\/blog\/%e5%a6%82%e4%bd%95%e4%bd%bf%e7%94%a8python%e8%ae%a1%e7%ae%97bleu%e5%88%86%e6%95%b0%ef%bc%9f\/#local-main-organization-logo","url":"","contentUrl":"","caption":"Blog - Silicon Cloud"}]}},"_links":{"self":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/388","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/comments?post=388"}],"version-history":[{"count":4,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/388\/revisions"}],"predecessor-version":[{"id":109952,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/posts\/388\/revisions\/109952"}],"wp:attachment":[{"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/media?parent=388"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/categories?post=388"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.silicloud.com\/zh\/blog\/wp-json\/wp\/v2\/tags?post=388"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}