'commit'

dsRag/ElasticSearch/T6_XiangLiangQuery.py

@@ -2,9 +2,7 @@ import logging
 import os
 from logging.handlers import RotatingFileHandler
 
-from gensim.models import KeyedVectors
-
-from Config.Config import ES_CONFIG, MS_MODEL_PATH, MS_MODEL_LIMIT
+from Config.Config import ES_CONFIG
 from ElasticSearch.Utils.ElasticsearchConnectionPool import ElasticsearchConnectionPool
 
 # 初始化日志
@@ -16,11 +14,6 @@ handler = RotatingFileHandler('Logs/start.log', maxBytes=1024 * 1024, backupCoun
 handler.setFormatter(logging.Formatter('%(asctime)s - %(levelname)s - %(message)s'))
 logger.addHandler(handler)
 
-# 1. 加载预训练的 Word2Vec 模型
-model = KeyedVectors.load_word2vec_format(MS_MODEL_PATH, binary=False, limit=MS_MODEL_LIMIT)
-logger.info(f"模型加载成功，词向量维度: {model.vector_size}")
-
-
 
 def init_es_pool():
     # 初始化Elasticsearch连接池

dsRag/Util/EsSearchUtil.py

@@ -1,11 +1,21 @@
+import logging
+import os
+from logging.handlers import RotatingFileHandler
+
 import jieba
 from elasticsearch import Elasticsearch
-from gensim.models import KeyedVectors
-
-from Config.Config import MS_MODEL_PATH, MS_MODEL_LIMIT
 
+# 初始化日志
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.INFO)
+# 确保日志目录存在
+os.makedirs('Logs', exist_ok=True)
+handler = RotatingFileHandler('Logs/start.log', maxBytes=1024 * 1024, backupCount=5)
+handler.setFormatter(logging.Formatter('%(asctime)s - %(levelname)s - %(message)s'))
+logger.addHandler(handler)
 
 class EsSearchUtil:
+
     def __init__(self, es_config):
         """
         初始化Elasticsearch搜索工具
@@ -18,15 +28,30 @@ class EsSearchUtil:
             verify_certs=False
         )
 
-    def text_to_embedding(self, text):
+    def __init__(self, es_config):
+        from gensim.models import KeyedVectors
+        from Config.Config import MS_MODEL_PATH, MS_MODEL_LIMIT
+        
         # 加载预训练模型
-        model = KeyedVectors.load_word2vec_format(MS_MODEL_PATH, binary=False, limit=MS_MODEL_LIMIT)
+        self.model = KeyedVectors.load_word2vec_format(MS_MODEL_PATH, binary=False, limit=MS_MODEL_LIMIT)
+        logger.info(f"模型加载成功，词向量维度: {self.model.vector_size}")
+        
+        # 初始化Elasticsearch连接
+        self.es = Elasticsearch(
+            hosts=es_config['hosts'],
+            basic_auth=es_config['basic_auth'],
+            verify_certs=False
+        )
+        self.index_name = es_config['index_name']
+        
+    def text_to_embedding(self, text):
+        # 使用已加载的模型
         # 对文本分词并计算平均向量
         words = jieba.lcut(text)
-        vectors = [model[word] for word in words if word in model]
+        vectors = [self.model[word] for word in words if word in self.model]
         
         if not vectors:
-            return [0.0] * model.vector_size
+            return [0.0] * self.model.vector_size
             
         # 计算平均向量
         avg_vector = [sum(dim)/len(vectors) for dim in zip(*vectors)]