datarobot-mlflow 0.1.dev1

datarobot-mlflow client to synchronize an MLFlow model with DataRobot model

mlflow-integration

This script and library exports a model from an MLflow model registry, and uploads key-values to an existing DataRobot model package.

Setup

• Python 3.7 or later
• pip install datarobot-mlflow
  • if using Azure: pip install "datarobot-mlflow[azure]"

DataRobot information

• URL of DataRobot instance, example https://app.datarobot.com
• ID of DataRobot model package to post key-values; example: 64227b4bf82db411c90c3209
• export MLOPS_API_TOKEN=<API token from DataRobot Developer Tools>

Local MLflow information

• MLflow tracking URI; example "file:///Users/me/mlflow/examples/mlruns"
• Model name; example "cost-model"
• Model version; example "2"

Azure DataBricks MLFlow with Service Principal information

• MLflow tracking URI; example "azureml://region.api.azureml.ms/mlflow/v1.0/subscriptions/subscription-id/resourceGroups/resource-group-name/providers/Microsoft.MachineLearningServices/workspaces/azure-ml-workspace-name'"
• Model name; example "cost-model"
• Model version; example "2"
• Provide service principal details in environment:
  • export AZURE_TENANT_ID="<tenant-id>"
  • export AZURE_CLIENT_ID="<client-id>"
  • export AZURE_CLIENT_SECRET="<secret>"

Invocation

ACTION=sync  # other actions: delete-all-dr-keys, list-mlflow-keys

DR_MODEL_ID="<MODEL_PACKAGE_ID>"

drflow-cli \
	--mlflow-url http://localhost:8080 \
 	--mlflow-model diabetes-1  \
        --mlflow-model-version 1 \
        --dr-model $DR_MODEL_ID \
        --dr-url https://app.datarobot.com \
        --with-artifacts \
        --verbose \
	--action $ACTION

Reference scripts

• type run.sh as an example, to see installed location
• run.sh uses the drflow CLI to sync key-values
• validate-auth.sh uses the CLI to validate Azure credentials only