向量模型的部署，知识库搭建milvus+PostgreSQL

教程中所使用到的脚本右边下载

激活虚拟环境

source qwen_env/bin/activate

# 安装 PostgreSQL
sudo apt install -y postgresql postgresql-contrib

sudo apt install -y postgresql postgresql-contrib

# 启动服务
sudo service postgresql start

# 创建数据库和用户
sudo -u postgres psql -c “CREATE DATABASE knowledge_db;”
sudo -u postgres psql -c “CREATE USER kb_admin WITH PASSWORD ‘your_password’;”
sudo -u postgres psql -c “GRANT ALL PRIVILEGES ON DATABASE knowledge_db TO kb_admin;”

# 配置远程访问 (可选)
sudo sed -i “s/#listen_addresses = ‘localhost’/listen_addresses = ‘*’/” /etc/postgresql/*/main/postgresql.conf
echo “host all all 0.0.0.0/0 md5” | sudo tee -a /etc/postgresql/*/main/pg_hba.conf
sudo service postgresql restart

pip install pymilvus
pip install modelscope

启动—————-
win系统要安装Docker 启动电脑时记得启动一下这个！！！

启动向量数据库docker-compose up -d

启动数据库：sudo service postgresql start

CD到数据写入脚本目录：

source qwen_env/bin/activate
cd /home/dministrator/gte_Qwen2-7B-instruct/Script

创建表：python data.py

写入数据：python process_knowledge.py

清理旧表：python cleanup_databases.py

重建约束：python fix_database.py

查询数据：python cxdata.py

查询向量库python check_milvus_count.py

启动向量7B+通用模型7B

source qwen_env/bin/activate
cd /home/dministrator/gte_Qwen2-7B-instruct
python api_server.py