From 439fc1ebc980fea603b9d450d13fa05a78f9eebe Mon Sep 17 00:00:00 2001 From: loopsaaage Date: Tue, 10 Sep 2024 15:22:31 +0800 Subject: [PATCH] Refactor and edit readme --- README.CN.md | 13 ++++++++++++- infra_ai_service/api/ai_enhance/embedding.py | 5 +++-- infra_ai_service/api/ai_enhance/text_process.py | 2 +- infra_ai_service/api/ai_enhance/vector_search.py | 7 +++---- infra_ai_service/api/common/utils.py | 4 ++-- tests/test_health.py | 15 --------------- 6 files changed, 21 insertions(+), 25 deletions(-) delete mode 100644 tests/test_health.py diff --git a/README.CN.md b/README.CN.md index 03b0b35..b236a0f 100644 --- a/README.CN.md +++ b/README.CN.md @@ -1,6 +1,5 @@ # 注意事项 - ## 前置条件 - 需要安装本地psql数据库 ```shell @@ -12,6 +11,18 @@ ALTER USER postgres PASSWORD 'postgres'; # 建立数据库（db） CREATE DATABASE db; ``` +- 需要安装qdrant本地服务 +```shell +# docker环境准备 +sudo apt update +sudo apt install docker.io +sudo systemctl start docker +sudo systemctl enable docker +# 安装 +sudo docker run -p 6333:6333 -d --name qdrant qdrant/qdrant +# 测试 +curl http://localhost:6333 +``` ## 本地启动 ```shell diff --git a/infra_ai_service/api/ai_enhance/embedding.py b/infra_ai_service/api/ai_enhance/embedding.py index 52c13c2..12f5910 100644 --- a/infra_ai_service/api/ai_enhance/embedding.py +++ b/infra_ai_service/api/ai_enhance/embedding.py @@ -30,6 +30,7 @@ async def embed_text(input_data: TextInput): # 生成嵌入 embeddings = list(fastembed_model.embed([input_data.content])) if not embeddings: + logger.error(f"Failed to generate embedding", exc_info=True) raise ValueError("Failed to generate embedding") embedding_vector = embeddings[0] @@ -51,7 +52,7 @@ async def embed_text(input_data: TextInput): return EmbeddingOutput(id=point_id, embedding=embedding_vector.tolist()) except Exception as e: - logger.error(f"Error in vector search: {str(e)}", exc_info=True) + logger.error(f"Error processing embedding: {str(e)}", exc_info=True) raise HTTPException(status_code=400, detail=f"Error processing embedding: {str(e)}") @@ -66,6 +67,6 @@ async def get_collection_status(): "status": "ready" if collection_info.status == "green" else "not ready" } except Exception as e: - logger.error(f"Error in vector search: {str(e)}", exc_info=True) + logger.error(f"Error getting collection status: {str(e)}", exc_info=True) raise HTTPException(status_code=400, detail=f"Error getting collection status: {str(e)}") diff --git a/infra_ai_service/api/ai_enhance/text_process.py b/infra_ai_service/api/ai_enhance/text_process.py index bdc07b5..8b044d5 100644 --- a/infra_ai_service/api/ai_enhance/text_process.py +++ b/infra_ai_service/api/ai_enhance/text_process.py @@ -28,5 +28,5 @@ async def process_text(input_data: TextInput): modified_text = clean_text(input_data.content) return TextOutput(modified_content=modified_text) except Exception as e: - logger.error(f"Error in vector search: {str(e)}", exc_info=True) + logger.error(f"Error processing text: {str(e)}", exc_info=True) raise HTTPException(status_code=400, detail=f"Error processing text: {str(e)}") diff --git a/infra_ai_service/api/ai_enhance/vector_search.py b/infra_ai_service/api/ai_enhance/vector_search.py index 75bb4b9..8dcb54b 100644 --- a/infra_ai_service/api/ai_enhance/vector_search.py +++ b/infra_ai_service/api/ai_enhance/vector_search.py @@ -1,9 +1,6 @@ from fastapi import APIRouter, HTTPException from pydantic import BaseModel from typing import List -from fastembed.embedding import DefaultEmbedding -from qdrant_client import QdrantClient -from qdrant_client.http.models import Distance, VectorParams from infra_ai_service.api.common.utils import setup_qdrant_environment import logging @@ -35,11 +32,13 @@ async def vector_search(input_data: SearchInput): try: # 检查集合是否存在 if not qdrant_client.get_collection(collection_name): + logger.error(f"Collection {collection_name} does not exist", exc_info=True) raise ValueError(f"Collection {collection_name} does not exist") # 生成查询文本的嵌入 query_vector = list(fastembed_model.embed([input_data.query_text])) if not query_vector: + logger.error(f"Failed to generate query embedding", exc_info=True) raise ValueError("Failed to generate query embedding") # 执行向量搜索 @@ -62,5 +61,5 @@ async def vector_search(input_data: SearchInput): return SearchOutput(results=results) except Exception as e: - logger.error(f"Error in vector search: {str(e)}", exc_info=True) + logger.error(f"Error performing vector search: {str(e)}", exc_info=True) raise HTTPException(status_code=500, detail=f"Error performing vector search: {str(e)}") diff --git a/infra_ai_service/api/common/utils.py b/infra_ai_service/api/common/utils.py index aa88512..796477d 100644 --- a/infra_ai_service/api/common/utils.py +++ b/infra_ai_service/api/common/utils.py @@ -1,7 +1,7 @@ -from fastapi import APIRouter, HTTPException +from fastapi import HTTPException from fastembed.embedding import DefaultEmbedding from qdrant_client import QdrantClient -from qdrant_client.http.models import Distance, VectorParams, PointStruct +from qdrant_client.http.models import Distance, VectorParams def setup_qdrant_environment(): diff --git a/tests/test_health.py b/tests/test_health.py deleted file mode 100644 index 8fbf5a8..0000000 --- a/tests/test_health.py +++ /dev/null @@ -1,15 +0,0 @@ -import pytest -from fastapi import FastAPI, status -from httpx import AsyncClient - -@pytest.mark.asyncio -async def test_health(client: AsyncClient, app: FastAPI) -> None: - """ - Checks the health endpoint. - - :param client: client for the infra_ai_service. - :param app: current FastAPI application. - """ - url = app.url_path_for("health") - response = await client.get(url) - assert response.status_code == status.HTTP_200_OK