This service has two modes, a model training (train.R) and an API polling and prediction service (prediction_service.R).
To run prediction service
- Turn on turbine data API.
- In the project directory type command
./run.sh API_SERVICE
To retrain model
- In the project directory type command
./run.sh TRAIN
NOTE: API polling rates and training params can be edited in config.yml