“第二十九章 AI诗人”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
高级读者
 
(相同用户的9个中间修订版本未显示)
第7行: 第7行:
 
*百度百科:诗歌 [https://baike.baidu.com/item/%E8%AF%97%E6%AD%8C/5555]
 
*百度百科:诗歌 [https://baike.baidu.com/item/%E8%AF%97%E6%AD%8C/5555]
 
*百度百科:绝句律诗的格律 [https://baike.baidu.com/item/%E7%BB%9D%E5%8F%A5%E5%BE%8B%E8%AF%97%E6%A0%BC%E5%BE%8B/4503398?fr=aladdin]
 
*百度百科:绝句律诗的格律 [https://baike.baidu.com/item/%E7%BB%9D%E5%8F%A5%E5%BE%8B%E8%AF%97%E6%A0%BC%E5%BE%8B/4503398?fr=aladdin]
*AI100问:AI如何成为诗人 [http://166.111.134.44:7777/caiyq/ai100/pdf/AI-100-70-%E6%9C%BA%E5%99%A8%E5%A6%82%E4%BD%95%E6%88%90%E4%B8%BA%E8%AF%97%E4%BA%BA.pdf]
+
*AI100问:AI如何成为诗人 [http://aigraph.cslt.org/ai100/pdf/AI-100-70-%E6%9C%BA%E5%99%A8%E5%A6%82%E4%BD%95%E6%88%90%E4%B8%BA%E8%AF%97%E4%BA%BA.pdf]
 
*以画入诗原文:[https://ojs.aaai.org/index.php/AAAI/article/view/12001/11860  How images inspire poems: Generating classical chinese poetry from images with memory networks]
 
*以画入诗原文:[https://ojs.aaai.org/index.php/AAAI/article/view/12001/11860  How images inspire poems: Generating classical chinese poetry from images with memory networks]
 
*百度百科:诗学含英 [https://baike.baidu.com/item/%E8%AF%97%E5%AD%A6%E5%90%AB%E8%8B%B1/4888951?fr=aladdin]
 
*百度百科:诗学含英 [https://baike.baidu.com/item/%E8%AF%97%E5%AD%A6%E5%90%AB%E8%8B%B1/4888951?fr=aladdin]
 
*《笠翁对韵》全文阅读 [http://xh.5156edu.com/page/z7550m1720j20012.html]
 
*《笠翁对韵》全文阅读 [http://xh.5156edu.com/page/z7550m1720j20012.html]
  
==高级读者==
+
==演示链接==
*拼凑法:[http://www.jos.org.cn/jos/article/pdf/3596?st=article_issue 一种宋词自动生成的遗传算法及其机器实现]
+
*概率法:[https://aclanthology.org/C08-1048.pdf Generating Chinese couplets using a statistical MT approach]
+
*神经网络法:[https://aclanthology.org/D14-1074.pdf Chinese poetry generation with recurrent neural networks]
+
*宋词的神经网络生成:[https://arxiv.org/pdf/1604.06274.pdf Chinese song iambics generation with neural attention-based model]
+
*灵活创新性诗歌生成:[https://arxiv.org/pdf/1705.03773.pdf Flexible and Creative Chinese Poetry Generation Using Neural Memory]
+
*清华大学CSLT AI诗人薇薇:[https://arxiv.org/pdf/1606.05829.pdf Can Machine Generate Traditional Chinese Poetry? A Feigenbaum Test]
+
 
+
==链接==
+
 
*2011年非神经网络诗人 [http://tssc.sinaapp.com/ ]
 
*2011年非神经网络诗人 [http://tssc.sinaapp.com/ ]
 
*诗三百人工智能诗人 [https://www.aichpoem.net/#/shisanbai/poem]
 
*诗三百人工智能诗人 [https://www.aichpoem.net/#/shisanbai/poem]
*微软小冰 [http://www.xiaoice.com/]
+
*微软小冰(关注其公众号后,输入“作诗”,即可唤起小冰为你作诗) [http://www.xiaoice.com/]
  
 
==开发者资源==
 
==开发者资源==
 
*中国诗歌数据库 [https://github.com/chinese-poetry/chinese-poetry]
 
*中国诗歌数据库 [https://github.com/chinese-poetry/chinese-poetry]
 
*Pytorch book中一个简单的RNN-CHAR生成古诗的github代码库 [https://github.com/chenyuntc/pytorch-book/tree/master/chapter09-neural_poet_RNN ]
 
*Pytorch book中一个简单的RNN-CHAR生成古诗的github代码库 [https://github.com/chenyuntc/pytorch-book/tree/master/chapter09-neural_poet_RNN ]
 +
*一个基于Keras的notebook程序,应用起来更简单[https://github.com/youyuge34/Poems_generator_Keras]
 +
*用UER[https://github.com/dbiir/UER-py/]训练出的transformer模型,转成hugging face transformers模式[https://github.com/huggingface/transformers],可直接用预训练模型测试性能。注意,需要装transformers。实测TF模型可用,pytorch模型有问题。[https://huggingface.co/uer/gpt2-chinese-poem]
 +
 +
==高级读者==
 +
*拼凑法: 一种宋词自动生成的遗传算法及其机器实现:[http://www.jos.org.cn/jos/article/pdf/3596?st=article_issue]
 +
*概率法: Generating Chinese couplets using a statistical MT approach:[https://aclanthology.org/C08-1048.pdf]
 +
*神经网络法: Chinese poetry generation with recurrent neural networks:[https://aclanthology.org/D14-1074.pdf]
 +
*宋词的神经网络生成: Chinese song iambics generation with neural attention-based model[https://arxiv.org/pdf/1604.06274.pdf]
 +
*灵活创新性诗歌生成:Flexible and Creative Chinese Poetry Generation Using Neural Memory[https://arxiv.org/pdf/1705.03773.pdf]
 +
*清华大学CSLT AI诗人薇薇:Can Machine Generate Traditional Chinese Poetry? A Feigenbaum Test[https://arxiv.org/pdf/1606.05829.pdf]
 +
*图片流作诗:Images2Poem : Generating Chinese Poetry from Image Streams[https://dl.acm.org/doi/pdf/10.1145/3240508.3241910 ]

2023年8月13日 (日) 02:18的最后版本

教学资料

扩展阅读

演示链接

  • 2011年非神经网络诗人 [6]
  • 诗三百人工智能诗人 [7]
  • 微软小冰(关注其公众号后,输入“作诗”,即可唤起小冰为你作诗) [8]

开发者资源

  • 中国诗歌数据库 [9]
  • Pytorch book中一个简单的RNN-CHAR生成古诗的github代码库 [10]
  • 一个基于Keras的notebook程序,应用起来更简单[11]
  • 用UER[12]训练出的transformer模型,转成hugging face transformers模式[13],可直接用预训练模型测试性能。注意,需要装transformers。实测TF模型可用,pytorch模型有问题。[14]

高级读者

  • 拼凑法: 一种宋词自动生成的遗传算法及其机器实现:[15]
  • 概率法: Generating Chinese couplets using a statistical MT approach:[16]
  • 神经网络法: Chinese poetry generation with recurrent neural networks:[17]
  • 宋词的神经网络生成: Chinese song iambics generation with neural attention-based model[18]
  • 灵活创新性诗歌生成:Flexible and Creative Chinese Poetry Generation Using Neural Memory[19]
  • 清华大学CSLT AI诗人薇薇:Can Machine Generate Traditional Chinese Poetry? A Feigenbaum Test[20]
  • 图片流作诗:Images2Poem : Generating Chinese Poetry from Image Streams[21]