Setting up predictive analytics services with Palladium
summary
Palladium is an open-source Python framework for building predictive analytics services that can fit, evaluate, store, distribute and update models, with core ML processes compatible with scikit-learn. The talk demonstrates using Palladium with other languages (R, Julia), shows an example classification service, and explains deployment and scaling using Docker plus Mesos/Marathon, including an automated Docker image creation script.
AI-written summary · about
related talks
34mBuild ML Enhanced Event Streaming Apps with Python Microservices
39mPyData in Production: Lesson Learned from Various PyData Deployment Strategies
32mDeploying Python models to production
17mProductionizing Data Science Workloads
19mDepend on Docker for Self-Managed Predictive Modeling
48mDeploying Machine Learning Models in Practice
17mR & Python playing nice, in Production
22mOperationalising Data Science on Cloud Foundry