A Fiery approach to R web services
summary
Claims a repeatable path from R model to realtime endpoint using ProjectTemplate, Fiery and Docker image. Shows a minimal pipeline: project skeleton auto-loads packages and data, a Fiery HTTP app plus router exposes a POST /predict that accepts JSON and returns k‑NN predictions on the iris dataset, and a Dockerfile installs OS deps (libxml2, libssl), runs the project installer and builds a portable container. The deployed container starts quickly and reproduces the prediction API as demonstrated.
AI-written summary · about
chapters (4)
related talks
22mPush straight to prod: API development with R and Tensorflow
01h 18mBuild Data Apps by Deploying ML Models as API Services
21mDeploying End-To-End Data Science with Shiny, Plumber, and Pins
37mBuild a Data-Driven Web App That Everyone Can Use
01h 02mDocker for Data Science: Reproducibility and Deployment
14mRobust R Deployments: Building a Pipeline from RStudio to Production
29mHow to automate the deployment of ...
01h 45mBuilding a Model Prediction Server