Skip to content

Latest commit

 

History

History
34 lines (23 loc) · 914 Bytes

README.md

File metadata and controls

34 lines (23 loc) · 914 Bytes

Example: Prediction Service

See “A poor man’s orchestration of predictive models, or do it yourself” for further details.

Usage

In order to build a Docker image and upload it to Container Registry, run

make build

In order to (1) start a Compute Engine instance for training a model, (2) wait until completion, and (3) ensure that everything went well, run

make training-start training-wait training-check

In order to (1) start a Compute Engine instance for applying a trained model, (2) wait until completion, and (3) ensure that everything went well, run

make application-start application-wait application-check

In order to schedule training and application using Airflow, run

git clone https://github.com/chain-rule/example-prediction-service.git "${AIRFLOW_HOME}/dags"