mlflow
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseMLflow
MLflow
MLflow is the standard for tracking experiments. v3.0 (2025) pivots to GenAI, adding LLM Tracing, Prompt Management, and "LLM-as-a-Judge".
MLflow是实验追踪的标准工具。其2025年发布的v3.0版本转向GenAI方向,新增了LLM Tracing、Prompt Management以及"LLM-as-a-Judge"功能。
When to Use
适用场景
- Experiment Tracking: Logging hyperparameters () and metrics (
lr=0.01).accuracy=0.98 - GenAI Tracing: Visualizing the full chain of a RAG application.
- Model Registry: Versioning models () for deployment.
my-model/v3
- 实验追踪:记录超参数(如)和指标(如
lr=0.01)。accuracy=0.98 - GenAI追踪:可视化RAG应用的完整链路。
- 模型注册表:为待部署的模型(如)进行版本管理。
my-model/v3
Core Concepts
核心概念
Tracking URI
追踪URI(Tracking URI)
Where logs are stored (local or remote ).
./mlrunshttp://mlflow-server日志存储的位置,可本地存储至或远程存储至。
./mlrunshttp://mlflow-serverAutologging
自动日志记录(Autologging)
mlflow.autolog()mlflow.autolog()LLM Tracing
LLM追踪(LLM Tracing)
OpenTelemetry-based tracing to debug prompt chains.
基于OpenTelemetry的追踪功能,用于调试提示词链路。
Best Practices (2025)
2025年最佳实践
Do:
- Use : To run "LLM-as-a-Judge" metrics on your RAG pipeline.
mlflow.evaluate() - Use Prompt Engineering UI: MLflow 3.0 has a UI to iterate on prompts.
Don't:
- Don't use it for data storage: Log artifacts (models), not datasets. Log metadata about datasets instead.
建议:
- 使用:在你的RAG管道上运行"LLM-as-a-Judge"指标评估。
mlflow.evaluate() - 使用提示词工程UI:MLflow 3.0提供了UI界面用于迭代优化提示词。
不建议:
- 不要用于数据存储:仅记录模型等工件,不要记录数据集,可改为记录数据集的元数据。