!https://github.com/burakince/mlflow/workflows/Build%20and%20Publish%20Mlflow%20Docker%20Image/badge.svg
!Docker Pulls
!Docker Image Size (tag)
镜像概述
欢迎使用MLflow Docker镜像仓库!所有Docker文件均托管在https://github.com/burakince/mlflow%E3%80%82%E5%AF%B9%E4%BA%8EKubernetes%E7%94%A8%E6%88%B7%EF%BC%8C%E5%8F%AF%E9%80%9A%E8%BF%87https://community-charts.github.io/docs/charts/mlflow/usage Helm chart实现无缝部署。
核心功能与特性
- 多版本支持:提供从1.24.0到3.5.0的多个MLflow版本标签,满足不同项目需求
- 便捷部署:通过简单Docker命令即可快速启动MLflow服务,无需复杂配置
- Kubernetes兼容:支持通过Helm chart在Kubernetes集群中部署,适合容器化生产环境
- 开源可靠:基于官方MLflow项目构建,源码透明,持续集成确保镜像质量
适用场景
- 机器学习实验跟踪与结果对比
- 模型版本管理与迭代记录
- 团队协作中的机器学习工作流标准化
- 需要快速搭建MLflow服务的开发、测试与生产环境
- 个人开发者与企业团队的机器学习项目管理
版本标签
提供以下主要版本标签(每个标签链接至对应Dockerfile):
- https://github.com/burakince/mlflow/blob/3.5.0/Dockerfile
- https://github.com/burakince/mlflow/blob/3.4.0/Dockerfile
- https://github.com/burakince/mlflow/blob/3.3.2/Dockerfile
- https://github.com/burakince/mlflow/blob/3.3.1/Dockerfile
- https://github.com/burakince/mlflow/blob/3.2.0/Dockerfile
- https://github.com/burakince/mlflow/blob/3.1.4/Dockerfile
- https://github.com/burakince/mlflow/blob/3.1.3/Dockerfile
- https://github.com/burakince/mlflow/blob/3.1.1/Dockerfile
- https://github.com/burakince/mlflow/blob/3.1.0/Dockerfile
- https://github.com/burakince/mlflow/blob/2.22.1/Dockerfile
- https://github.com/burakince/mlflow/blob/2.22.0/Dockerfile
- https://github.com/burakince/mlflow/blob/2.21.3/Dockerfile
- https://github.com/burakince/mlflow/blob/2.21.2/Dockerfile
- https://github.com/burakince/mlflow/blob/2.21.1/Dockerfile
- https://github.com/burakince/mlflow/blob/2.21.0/Dockerfile
- https://github.com/burakince/mlflow/blob/2.20.4/Dockerfile
- https://github.com/burakince/mlflow/blob/2.20.3/Dockerfile
- https://github.com/burakince/mlflow/blob/2.20.2/Dockerfile
- https://github.com/burakince/mlflow/blob/2.20.1/Dockerfile
- https://github.com/burakince/mlflow/blob/2.20.0/Dockerfile
- https://github.com/burakince/mlflow/blob/2.19.0/Dockerfile
- https://github.com/burakince/mlflow/blob/2.1.1/Dockerfile
- https://github.com/burakince/mlflow/blob/2.1.0/Dockerfile
- https://github.com/burakince/mlflow/blob/2.0.1/Dockerfile
- https://github.com/burakince/mlflow/blob/1.30.0/Dockerfile
- https://github.com/burakince/mlflow/blob/1.29.0/Dockerfile
- https://github.com/burakince/mlflow/blob/1.28.0/Dockerfile
- https://github.com/burakince/mlflow/blob/1.27.0/Dockerfile
- https://github.com/burakince/mlflow/blob/1.26.1/Dockerfile
- https://github.com/burakince/mlflow/blob/1.26.0/Dockerfile
- https://github.com/burakince/mlflow/blob/1.25.1/Dockerfile
- https://github.com/burakince/mlflow/blob/1.25.0/Dockerfile
- https://github.com/burakince/mlflow/blob/1.24.0/Dockerfile
使用方法
Docker部署
通过以下命令快速启动MLflow服务:
bash
docker run -d -p 5000:5000 burakince/mlflow
参数说明
-d: 后台运行容器
-p 5000:5000: 将容器的5000端口映射到主机的5000端口
burakince/mlflow: 默认使用latest标签,可指定具体版本如burakince/mlflow:3.5.0
服务启动后,通过http://localhost:5000访问MLflow Web界面。
Kubernetes部署
对于Kubernetes环境,推荐使用社区维护的Helm chart:
bash
helm repo add community-charts https://community-charts.github.io/docs/
helm install mlflow community-charts/mlflow
详细配置请参考https://community-charts.github.io/docs/charts/mlflow/usage%E3%80%82