tree2retriever:面向RAG场景的递归摘要树检索器实现

文章预览

tree2retriever 面向RAG场景的递归摘要树检索器实现 Recursive Abstractive Processing for Tree-Organized Retrieval Github:https://github.com/yanqiangmiffy/tree2retriever Example import  logging import  pickle from  tree2retriever.cluster_tree_builder  import  ClusterTreeBuilder, ClusterTreeConfig from  tree2retriever.embedding  import  SBertEmbeddingModel from  tree2retriever.summary  import  GLMSummarizationModel from  tree2retriever.tree_retriever  import  TreeRetriever, TreeRetrieverConfig if  __name__ ==  '__main__' :     tree_builder_type =  'cluster'     logging.basicConfig(format= "%(asctime)s - %(message)s" , level=logging.INFO)     supported_tree_builders = { "cluster" : (ClusterTreeBuilder, ClusterTreeConfig)}     tree_builder_class, tree_builder_config_class = supported_tree_builders[         tree_builder_type     ]     embedding_model = SBertEmbeddingModel(model_name= r"I:\pretrained_models\bert\eng ………………………………