Machine Learning - Enhancements for Python
Exponential Smoothing Method (ESM) for Time Series Forecasting in OML4Py
Exponential Smoothing is a moving average method with a single parameter which models an exponentially decreasing effect of past values. This in-database algorithm is exposed through the Python API of Oracle Machine Learning for Python.
Exponential Smoothing Methods have been widely used in forecasting for over half a century. It has applications at the strategic, tactical, and operational levels. Being exposed as part of the Python API, you have native Python access to this in-database algorithm.
Non-Negative Matrix Factorization Support for Dimensionality Reduction in OML4Py
Non-Negative Matrix Factorization (NMF) is a state-of-the-art feature extraction algorithm. You can now use this in-database algorithm through the Python API of Oracle Machine Learning for Python.
NMF is useful when there are many attributes, and those attributes are ambiguous or have weak predictability. By combining attributes through linear combinations, NMF can produce meaningful patterns, topics, or themes.
Support for Date, Time, and Integer Data Types in OML4Py
OML4Py introduces support for date, time, and Integer data types.
The OML4Py support for date, time, and integer data types enables you to create pandas DataFrame proxy objects and operate on database tables and views that contain those data types. This enables you to explore and prepare data at scale in the database.
XGBoost for Classification and Regression in OML4Py
XGBoost is a scalable gradient tree boosting algorithm that supports both classification and regression. The in-database implementation makes available the XGBoost Gradient Boosting open source package.
XGBoost is a popular classification and regression algorithm due to its high predictive accuracy. Being exposed as part of the Python API, you have native Python access to this in-database algorithm.