README.md

Set up an MLflow instance on your local machine using docker and terraform, with minio as artifact storage.

Setup

1. Clone the repo

git clone https://github.com/n-traore/mlflow-terraform-docker.git && \
cd mlflow-terraform-docker

2. Source the setup_teraform script

source setup_terraform

3. Run terraform commands to create the instance

This will setup a MLflow tracking server, a postgres database and a S3-like bucket in minio for artifact storage.

terraform init

# to check the resources that will be created
terraform plan

# create
terraform apply --auto-approve

• Visit http://localhost:8090 for the MLflow UI
• Visit http://localhost:9100 for the minio UI
• You can find the minio bucket credentials at modules/mlflow/mounts/shares/creds_YOUR_BUCKET_NAME.txt (default bucket name is mlflow)

4. Clean-up

Delete the instance by running the following command:

terraform destroy --auto-approve