MLFlow

Model Structure

MLflow models have the following structure.

# https://mlflow.org/docs/latest/models.html#mlmodel-file
my_model/
├── MLmodel
├── model.pkl
├── conda.yaml
├── python_env.yaml
└── requirements.txt

For more information about models, please refer to the following page: https://mlflow.org/docs/latest/models.html

Environment

In the BentoML deployment environment, Python 3.10 is used based on buildpack-deps:jammy (Ubuntu 22.04). The uncompressed model will be located in /user/model.

Example Projects

MLflow provides various example projects, which you can refer to for creating models.

• https://github.com/mlflow/mlflow-example
• https://mlflow.org/docs/latest/tutorials-and-examples/index.html
• https://github.com/mlflow/mlflow/tree/master/examples/docker
• https://mlserver.readthedocs.io/en/latest/examples/mlflow/README.html