The GLE node fits generalized linear models (GLMs), which includes a broad class of models that are defined by their error probability distributions (all of which fall into the exponential family of distributions) and link or transformation functions. This class of models includes linear regression for normally distributed responses, logistic and probit models for binary data, loglinear models for count data, complementary log-log models for interval-censored survival data, plus many others. GLE will also fit multinomial response models for categorical targets with more than two levels, with either ordered or ordinal responses or nominal or unordered responses. In the latter case, the model is technically no longer a generalized linear model, but shares many characteristics with GLMs, particularly in practice.
- For more information about this node, see GLE Overview.
- For more information about the visualizations for this node, see GLE Visualizations.
Like your visualization? Why not deploy it? For more information, see Deploy a model.