From bd28335d5dc3ff2cab2476f0821a481f77f9814b Mon Sep 17 00:00:00 2001
From: HuangHai <10402852@qq.com>
Date: Mon, 23 Jun 2025 19:09:45 +0800
Subject: [PATCH] 'commit'

---
 dsRag/Config/Config.py                        |   4 +++-
 .../Config/__pycache__/Config.cpython-310.pyc | Bin 311 -> 408 bytes
 dsRag/T2_Txt2Vec.py                           |   8 +++++++-
 dsRag/T4_VectorSave.py                        |  14 +++++++-------
 dsRag/Util/EmbeddingUtil.py                   |   5 ++---
 .../__pycache__/EmbeddingUtil.cpython-310.pyc | Bin 1261 -> 1224 bytes
 6 files changed, 19 insertions(+), 12 deletions(-)

diff --git a/dsRag/Config/Config.py b/dsRag/Config/Config.py
index fec10fc7..7c522ab4 100644
--- a/dsRag/Config/Config.py
+++ b/dsRag/Config/Config.py
@@ -5,4 +5,6 @@ ES_CONFIG = {
     "verify_certs": False,
     "ssl_show_warn": False,
     "default_index": "knowledge_base"
-}
\ No newline at end of file
+}
+
+WORD2VEC_MODEL_PATH = r"D:\Tencent_AILab_ChineseEmbedding\Tencent_AILab_ChineseEmbedding.txt"
diff --git a/dsRag/Config/__pycache__/Config.cpython-310.pyc b/dsRag/Config/__pycache__/Config.cpython-310.pyc
index 79c3fc9013d422c137185d041b75d05871232f64..b01f98ae74452dfa0ee8c02b7fa93823836d66b8 100644
GIT binary patch
delta 250
zcmdnaG=rHhpO=@50SIJFBQty%C-TY2NdS2%3@MB$OhCw-!ZL?7iXnwHiZO*Pm_d_$
zqDxqP6jxGWab|LSVrfZ66i-=dQD$0Yd~#}0NpTc!adA$3aYlZ5e0gF~-Ywpg)U?FX
zoRav=yp+_6Di;^4n2^-G<kY;9ct=m4#H4uVjLf{$;#Ak%q|}s@%)E30(t0HoC4SnP
zOi`S!!ST-ier}%bw}iv}gItWlT%F^6{asvr;sYE*JXSIku>ieM#4_=voe&?8#l*zK
K2!U+BIXD1zph`;s

delta 153
zcmbQiyq$?JpO=@50SJnWBQstxOyrZ15&`m37*ZHhn1GNug=G$F6hjJYFoP!BM2|54
zTU<$r#hJ<RiKQhOw|L4@i!#$H<C9a1N{Vms78mEl7iZ*`$CoD-<wfzPq^2d7=9I)|
t=B1=o_-QiU;&cs;clP&l^K@UyP{anZlx?yHqn!W`kjccv#0Y_0zX9GcDhL1o

diff --git a/dsRag/T2_Txt2Vec.py b/dsRag/T2_Txt2Vec.py
index 9e45c34f..98454d96 100644
--- a/dsRag/T2_Txt2Vec.py
+++ b/dsRag/T2_Txt2Vec.py
@@ -5,10 +5,16 @@ pip install jieba
 '''
 
 from Util.EmbeddingUtil import *
+from Config.Config import *
+
 # 在加载模型后添加自定义词典
 jieba.load_userdict("./Dict/custom_dict.txt")  # 文件中包含: 花呗
 
 sentences = ['如何更换花呗绑定银行卡', '花呗更改绑定银行卡']
 
 for sentence in sentences:
-    x = text_to_embedding(sentence)
\ No newline at end of file
+    x = text_to_embedding(sentence)
+
+# 词汇数+向量维度
+with open(WORD2VEC_MODEL_PATH, 'r', encoding='utf-8') as f:
+    print("词汇数和向量维度：" + f.readline())  # 第一行为词汇数和向量维度，在这里不予展示
diff --git a/dsRag/T4_VectorSave.py b/dsRag/T4_VectorSave.py
index e66fb5e9..4d49baf2 100644
--- a/dsRag/T4_VectorSave.py
+++ b/dsRag/T4_VectorSave.py
@@ -1,8 +1,9 @@
-from Config.Config import ES_CONFIG
+import datetime
+
 from elasticsearch import Elasticsearch
+
+from Config.Config import ES_CONFIG
 from T2_Txt2Vec import text_to_embedding
-import datetime
-import warnings
 
 # 初始化ES连接
 es = Elasticsearch(
@@ -12,16 +13,15 @@ es = Elasticsearch(
     ssl_show_warn=ES_CONFIG["ssl_show_warn"]
 )
 
-# 修改create_vector_index和save_to_es函数中使用ES_CONFIG["default_index"]
 def create_vector_index(index_name="knowledge_base"):
-    """创建带有向量字段的索引"""
+    """创建带有向量字段的索引(适配200维腾讯词向量)"""
     mapping = {
         "mappings": {
             "properties": {
                 "text": {"type": "text", "analyzer": "ik_max_word"},
                 "vector": {
                     "type": "dense_vector",
-                    "dims": 768,  # 需与text2vec模型维度一致
+                    "dims": 200,  # 修改为腾讯词向量实际维度
                     "index": True,
                     "similarity": "cosine"
                 },
@@ -34,7 +34,7 @@ def create_vector_index(index_name="knowledge_base"):
         if es.indices.exists(index=index_name):
             es.indices.delete(index=index_name)
         es.indices.create(index=index_name, body=mapping)
-        print(f"索引 {index_name} 创建成功")
+        print(f"索引 {index_name} 创建成功(200维)")
     except Exception as e:
         print(f"创建索引失败: {str(e)}")
         raise
diff --git a/dsRag/Util/EmbeddingUtil.py b/dsRag/Util/EmbeddingUtil.py
index fde367c0..8a0031b4 100644
--- a/dsRag/Util/EmbeddingUtil.py
+++ b/dsRag/Util/EmbeddingUtil.py
@@ -1,14 +1,13 @@
 import logging
 import jieba
 from gensim.models import KeyedVectors
-
+from Config.Config import *
 # 配置日志
 logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")
 logger = logging.getLogger(__name__)
 
 # 初始化 Word2Vec 模型
-model_path = r"D:\Tencent_AILab_ChineseEmbedding\Tencent_AILab_ChineseEmbedding.txt"
-model = KeyedVectors.load_word2vec_format(model_path, binary=False, limit=10000)
+model = KeyedVectors.load_word2vec_format(WORD2VEC_MODEL_PATH, binary=False, limit=10000)
 logger.info(f"模型加载成功，词向量维度: {model.vector_size}")
 
 # 将文本转换为嵌入向量
diff --git a/dsRag/Util/__pycache__/EmbeddingUtil.cpython-310.pyc b/dsRag/Util/__pycache__/EmbeddingUtil.cpython-310.pyc
index a42cd0ed856963b63a2f8ac1aa9235437867c449..899fae43233515494ba441b092b7d1ce145e87b3 100644
GIT binary patch
delta 409
zcmYLD%SyvQ6rG#&k<_FQ6<moT6q<@61#zK>qJrRCOT}zL?Y+j9qyv-qL~!G_i;{&4
z5%ddO_yaEe0+IfK3;)4MyKrF6opaB*bMCMHshiw1GfbZ4`IGZf*Bl40J3Voa3{@{t
zWyD%Zc~h9cWJdwXU(%$#f~}L<%Fqy|FpXMAwGGl=Ow`F}WuMry@(RE(gF_F@&LJEz
z%lL^yb}i@`TXaKMqejR@AUTw8tzxMjG{UA&Rxmj;SuAxw%I|Yfhm<CVgcdz2D^Q5O
zl`qhvH+6Ap(2g}jLaz|gb?AiLA^(&Rn{7868Hw?QF)3PkJMD4Tl}(Y8+%{cm2Tgz0
z=SW^4l=uU4jknL5?(CpM2MOygAA3hVUPtGO3#?pH!#U%^uZO*n<GsW6N^N7!+1Xp)
z*mm|;tD8cXeo=^FG=?}@{IA)6^>@{lWblPY%OD-Wump{1V1NdBB?EbN7!36fmIq;h

delta 368
zcmX@X`IeJ6pO=@50SG>5M`lD#<dtR2n5ZqElFE?90>rE-%qc9rOpFYv3=7y&*%vZK
zaip+Dai%h7aiwymu%)o4aP%@qai?-D;7Q?3<<8=r%`lf~AtNJ03Renu3qurN3S%&X
zCeKTd1%8^0w|Kl$D^pX#Qj<&ai;5?1w{fj<aj}XCNzF@6%`1s_^z=zgig(V)%u6j!
zb<IsmO-aejOD7<$S5mR@QxYQ^JJ5<E_Q_F94va#RXER-6<d|H<Y^?+l2FY;)iCcmt
zsTC#hCHe8GFyr|rpJ$d86az|X@<*}f<fo?t?JVM({FhmLvNDUN06$PBiYqriB{e6$
wAh9H4asZ2%Y7qy>L;(;X3?e|<ix5_US=^Hsu*mQTFmo}oFbgmWG4e120874JZU6uP