Introduction

This is an example of using plumber to create an API for an R program.

The service is exposing several endpoints as following:

• GET /status - return the status of the service
• GET /samples - 1 parameter: spec - if provided, return this species cardinality, otherwise return all samples cardinality
• GET /plot - 1 parameter: spec - if provided, plot only this species samples, otherwise plot all samples
• POST /prediction_class - 2 parameters: sepal_length & petal_length, predict the class (model is randomForest)

Source files

The following files are included in the project:

• train_model.R - script to train the model; a RandomForest model is trained to predict the species of iris
• plumber.R - API using plumber; to predict the species, a pretrained model is load
• run_it.R - script to start the API using plumber

Usage

To start the service, run run_it.R