Comments (1)
To address the issue of memory not being released and leading to an OutOfMemory (OOM) error when consecutively uploading documents with self.milvus.add_documents
, consider implementing the following strategies:
-
Batch Processing: Break down the document uploads into smaller batches. This approach limits the amount of memory used at any given time and allows for periodic memory release.
-
Garbage Collection: Manually invoke Python's garbage collector using
import gc
followed bygc.collect()
after processing each batch or a significant operation. This can help in reclaiming memory that's no longer in use. -
Optimize Document Handling: Ensure that the operations on documents before uploading do not unnecessarily increase memory usage. This includes careful management of any modifications or additions to document metadata.
-
Memory Profiling: Utilize memory profiling tools to identify specific areas in your code where memory usage is high or not being released as expected. This can provide insights into potential optimizations or issues.
-
Review Milvus Configuration: Check the configurations used for Milvus to ensure they are optimized for your specific use case, as certain settings might impact memory usage.
By implementing these strategies, you should be able to mitigate the memory issues and prevent the OutOfMemory error during document upload processes.
Details
Sources
- server/knowledge_base/kb_service/milvus_kb_service.py
- server/db/repository/knowledge_file_repository.py
- server/knowledge_base/kb_service/es_kb_service.py
About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.
Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.
To continue the conversation, mention @dosu.
from langchain-chatchat.
Related Issues (20)
- 报错stderr之后没有反应
- 如何在知识库问答时,让模型回答内容中的每句话添加上角标引用来源及链接,像秘塔AI搜索问答结果那样? HOT 1
- 在知识库问答模式如何同时使用多个知识库 HOT 1
- 关于reranker重排的使用方式 HOT 1
- vllm支持qwen1.5-32B和Yi1.5-34B HOT 2
- 在知识库问答模式如何同时调用多个知识库 HOT 1
- 怎么通过url的方式调用知识库问答 HOT 1
- inotify watch limit reached
- [BUG] search_knowledgebase_complex.py如何加载本地模型进行Rag?model_container.MODEL应该如何修改成自己的本地模型?
- 读取知识库的介绍
- 初始化数据库执行python init_database.py --recreate-vs 在服务器资源完全足够的情况下报错:Bus error (core dumped) HOT 2
- 与excel交互时,数据经常查不全 HOT 1
- 如何在同一个知识库下,做多级目录管理,
- 知识库问答报错API通信遇到错误:peer closed connection without sending complete message body (incomplete chunked read)
- Tesla M40不支持吗? HOT 2
- 调用api接口知识库问答怎么设置历史对话轮数? HOT 4
- 如何使加载rerank模型时默认使用trust_remote_code=true HOT 1
- 如何同时启用多个llm模型
- 文档内容关联标题
- 添加PDF文件到知识库时报错: HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from langchain-chatchat.