Oracle Machine Learning: what's new and what's next?
summary
Overview of concrete capabilities and roadmap: OML runs in-database ML via OML4Py AutoML UI and OML4SQL with Python, R and SQL APIs, plus no-code Oracle Data Miner and OML Services REST deployment. New: OML4R 2.0, ONNX clustering support, Italian text analytics, and SQL API for embedded Python that stores models in the DB. Roadmap items include data and model drift detection and bias monitoring, batch/asynchronous scoring, AutoML for unstructured text, conda package support, time series forecasting and Survival Analysis, and a 4,000 column limit.
AI-written summary · about
chapters (18)
- 0:00Oracle Machine Learning intro
- 1:45OML product family overview
- 3:36OML4SQL benefits
- 5:00OML4Py and OML4R explained
- 6:49AutoML UI overview
- 8:18OML licensing
- 9:44New OML features
- 10:29OML4R 2.0 on Autonomous DB
- 11:30OML4Py SQL API
- 13:16OML Services for models
- 14:43OML Monitoring lifecycle
- 17:41AutoML UI for text
- 18:45Oracle 23c roadmap
- 21:42Time series regression
- 22:25Integration with other Oracle products
- 23:58OAC AutoML integration
- 24:43OML success stories
- 25:25OML resources
related talks
31mIn-database machine learning: Automate, scale, deploy
22mPredict the future (or don't) with machine learning in applications
54mUnderstand & Where to use AI & Machine Learning 101
12mOpenMLDB: An Open-Source Real-Time Feature Platform
12mTackle Any Workload with Oracle Database, AI, and Multicloud
04mBreaking down the buzzwords to get value from AI, ML, and deep learning
14mMulticloud Analytics Made Simple
13mOperationalizing machine learning for the enterprise