API Reference

This section documents the key classes and functions of Modeva, reconstructed from the public API. Each model has its own page so it can be found directly from the sidebar, and the pages are grouped into families — generalized linear and additive models, trees and ensembles, neural networks, rule-based/kernel/ mixture models, and the model framework — rather than one long flat list. The TestSuite explainability, interpretation, diagnostic and comparison tests are likewise split by area, so every part of the reference is grouped consistently and easy to browse.

Data

• DataSet
• Feature Engineering

Generalized Linear & Additive Models

• Linear Models
• GLM Tree
• GLM Tree Boost

Trees & Ensembles

• Decision Tree
• Random Forest
• Gradient Boosting
• XGBoost
• LightGBM
• CatBoost

Neural Networks

• ReLU-DNN
• Neural Tree
• GAMINet

Rule-based, Kernel & Mixture Models

• DirectRS
• ICL-MoE
• GBDT Leaf Kernel
• FuseKernel
• Mixture of Experts

Model Framework

• Model Zoo
• Model Wrappers
• Base Model Classes
• Hyperparameter Tuning

Explainability & Interpretation

• Test Suite
• Post-hoc Explainability
• Inherent Interpretation

Diagnostics

• Performance
• Weakness Detection
• Residual Analysis
• Overfit
• Reliability
• Robustness
• Resilience
• Fairness

Model Comparison

• Model Comparison

Results & Utilities

• Validation Result
• Pipeline
• Utilities