getml.feature_learning

This module contains relational learning algorithms to learn features from relational data or time series.

Note:

All feature learners need to be passed to Pipeline.

Classes

FastProp(aggregation, delta_t, ...)

Generates simple features based on propositionalization.

Multirel(aggregation, allow_sets, delta_t, ...)

Feature learning based on Multi-Relational Decision Tree Learning.

Relboost(allow_null_weights, delta_t, gamma, ...)

Feature learning based on Gradient Boosting.

Fastboost([gamma, loss_function, max_depth, ...])

Feature learning based on Gradient Boosting.

RelMT(allow_avg, delta_t, gamma, ...)

Feature learning based on relational linear model trees.

Submodules

aggregations

This module contains all possible aggregations to be used with Multirel, FastProp, Mapping.