predictive
relationalai.semantics.reasoners
RelationalAI Predictive Reasoner using Graph Neural Networks (GNNs).
The Predictive Reasoner generates meaningful predictions from your relational data, supporting decision-intelligence tasks like identifying potential churn, detecting fraud, and more.
It provides a declarative Python API for formulating prediction problems — node classification, regression, or link prediction — and solving them with Graph Neural Networks.
Functions
Functions exposed by this module.
RelFMTabular()Predict in-context with RelFM on a single flat table — no relational joins. Re-exported from
relationalai.semantics.reasoners.predictive.estimators.relfm_tabular.factory.Classes
Classes exposed by this module.
GNNTrain, load, register and predict with a Graph Neural Network. Re-exported from
relationalai.semantics.reasoners.predictive.estimators.relational.gnn.RelFMPredict in-context with RelFM — a foundation model that requires no training. Re-exported from
relationalai.semantics.reasoners.predictive.estimators.relational.relfm.RelFMTabularPredictorClassification/regression flat-table RelFM — see
RelFMTabular. Re-exported from relationalai.semantics.reasoners.predictive.estimators.relfm_tabular.predictor.RelFMTabularForecasterForecasting flat-table RelFM — see
RelFMTabular. Re-exported from relationalai.semantics.reasoners.predictive.estimators.relfm_tabular.forecaster.PropertyTransformerAnnotate concept fields with semantic types for GNN data preparation. Re-exported from
relationalai.semantics.reasoners.predictive.core.property_transformer.Modules and Subpackages
Submodules and subpackages available under this namespace.
estimatorsPredictive estimators: GNN, RelFM, and RelFMTabular.