The Data Model component of the Intel® oneAPI Data Analytics Library (oneDAL) provides classes for model representation. The model mimics the actual data and represents it in a compact way so that you can use the library when the actual data is missing, incomplete, noisy or unavailable.
There are two categories of models in the library: Regression models and Classification models. Regression models are used to predict the values of dependent variables (responses) by observing independent variables. Classification models are used to predict to which sub-population (class) a given observation belongs.
A set of parameters characterizes each model. oneDAL model classes provide interfaces to access these parameters. It also provides the corresponding classes to train models, that is, to estimate model parameters using training data sets. As soon as a model is trained, it can be used for prediction and cross-validation. For this purpose, the library provides the corresponding prediction classes.