APIs

OML4Py

Oracle Machine Learning for Python (OML4Py) enables you to run Python code for data transformations and for statistical, machine learning, and graphical analysis on data stored in or accessible through Oracle Autonomous Database and Oracle Database instances using a Python API. OML4Py is a proprietary Python library that enables Python users to manipulate data in database tables and views using Python syntax. OML4Py provides a select set of Python functions and methods that transparently generate SQL and PL/SQL to perform requested functionality with in-database processing.

OML4Py users can use Automated Machine Learning (AutoML) to enhance user productivity and machine learning results through automated algorithm and feature selection, as well as automated model tuning and selection. Users can use Embedded Python Execution to run user-defined Python functions in Python engines spawned by the Oracle Autonomous Database and Oracle Database instances.

OML4Py is included with Oracle Database on-premises, Base Database Service (BDBS), and Oracle Autonomous Database. To learn more, see Machine Learning System Requirements.

OML4R

Oracle Machine Learning for R (OML4R) enables you to run R code for data transformations, statistical analysis, machine learning, and graphical analysis on data stored in or accessible through Oracle Autonomous Database and Oracle Database instances using an R API. OML4R is a proprietary set of R packages that allows R users to manipulate data in database tables and views using R syntax and run user-defined R functions. OML4R functions and methods translate a select set of R functions that transparently generate SQL and PL/SQL to perform requested functionality with in-database processing.

OML4R supports in-database machine learning model building, scoring, and evaluation using Oracle Machine Learning algorithms. However, users can leverage Embedded R Execution to run user-defined R functions inside R engines managed by Oracle Autonomous Database and Oracle Database instances.

OML4R is included with Oracle Database on-premises, Base Database Service (BDBS), and Oracle Autonomous Database.

OML4SQL

Oracle Machine Learning for SQL (OML4SQL) provides PL/SQL access to powerful, in-database machine learning algorithms and SQL access to corresponding models. You can use OML4SQL to build and deploy predictive and descriptive machine learning models that can be used to add intelligent capabilities to applications and dashboards. OML4SQL is included across Oracle Autonomous Database and Oracle Database instances.

To learn more, see Machine Learning System Requirements.

OML Services

Oracle Machine Learning Services (OML Services) provides MLOps support on Autonomous Database Serverless and Dedicated Region, where you can manage and use machine learning models from REST endpoints.

OML Services supports model management, deployment, data and model monitoring, as well as data bias detection. OML Services provides lightweight scoring using REST endpoints, making it appropriate for real-time and streaming applications.

Users have the option to deploy in-database models or "bring your own model" in Open Neural Networks Exchange (ONNX) format. These ONNX-format models can be created in third-party environments and imported into the database for use through the same REST API. OML Services supports various types of models, including classification, regression, clustering, and feature extraction.

Unlike other machine learning model deployment methods, which require you to provision, configure, manage, and pay for a Virtual Machine around the clock, OML Services only charges for the actual scoring. The Autonomous Database takes care of provisioning and managing the Virtual Machine environment.

To learn more, see What is OML Services.