Welcome to PG-SUI’s documentation!

Contents:

• About PG-SUI
  • PG-SUI Philosophy
  • Key Design (at a glance)
  • References
• Installation
  • Command-line installation (pip)
  • Anaconda installation
  • MacOS GUI add-on (Electron)
  • Development Installation
• PG-SUI Tutorial
  • Overview
  • What’s new
  • How to run PG-SUI
  • Model Families
  • Installation
  • Data Input, Encodings, and Conventions
  • CLI Quickstart (copy/paste)
  • Loading best parameters from a previous run
  • Simulated missingness and evaluation
  • Quick Start (End-to-End VAE Example)
  • Using Presets Programmatically
  • YAML Configuration Files
  • Command-Line Interface (CLI)
  • Minimal API Reference
  • References
• Simulated Missingness and Evaluation
  • Overview
  • When simulated missingness is used
  • Simulation process
  • Simulation strategies
  • Evaluation workflow
  • Configuration and CLI controls
  • Notes and best practices
• Config Reference
  • Common Blocks (Unsupervised NN)
  • Field-by-field notes
  • Unsupervised NN presets
  • Deterministic Imputers
  • Supervised Wrappers (RF / HistGB)
• PG-SUI Desktop GUI
  • Install & launch
  • First run walkthrough
  • GUI to CLI Mapping
  • Troubleshooting & Tips
• Unsupervised Imputers
  • Overview
  • Workflow Highlights
  • Configuration
  • Config Dataclasses
  • Model Summaries
  • Usage Examples
• ImputeAutoencoder
  • Overview
  • Model formulation
  • Algorithm summary
  • Configuration highlights
  • Usage
  • References
• ImputeVAE
  • Overview
  • Model formulation
  • Algorithm summary
  • Configuration highlights
  • Usage
  • References
• ImputeNLPCA
  • Overview
  • Model formulation
  • Algorithm summary
  • Configuration highlights
  • Usage
• ImputeUBP
  • Overview
  • Model formulation
  • Training phases
  • Configuration highlights
  • Usage
  • References
• Optuna Hyperparameter Tuning
  • Overview
  • Workflow
  • Configuration controls
  • Artifacts and outputs
  • Examples
  • Notes
• Deterministic (non-Machine Learning) Imputers
  • Overview
  • What’s included
  • Shared behavior & outputs
  • Quick start (Python)
  • YAML configuration
  • CLI usage (concept)
  • Configuration dataclasses
  • API reference
• Supervised Imputers
  • Overview
  • Shared Arguments
  • Config Dataclasses (API)
  • Supervised Imputer Models
  • CLI Examples
  • Outputs
• Imputation Methods Implemented in PG-SUI
  • Autoencoder Model for Genotype Data Imputation
  • Variational Autoencoder (VAE) Model for Genotype Data Imputation
  • Non-linear PCA (NLPCA) Model for Genotype Data Imputation
  • Unsupervised Backpropagation (UBP) Model for Genotype Data Imputation
  • IterativeImputer and the MICE Algorithm
  • Further Reading
  • Conclusion
• Tutorial: Implementing New Imputation Models
  • Prerequisites
  • Model Development Overview
  • Step 0 — Create a Config Dataclass
  • Step 1 — Implement the PyTorch Module
  • Step 2 — Write the Imputer Wrapper
  • Step 3 — (Optional) Hyperparameter Tuning
  • Registering the Model (Top-Level API & CLI)
  • Usage Examples
  • Design Notes & Best Practices
  • FAQ
• pgsui
  • pg-sui package
• Changelog
  • v1.7.8 - 2026-02-15
  • v1.7.7 - 2026-02-03
  • v1.7.6 - 2026-02-02
  • v1.7.4 - 2026-01-20
  • v1.7.3 - 2026-01-19
  • v1.7.2 - 2026-01-19
  • v1.7.1 - 2026-01-17
  • v1.7.0 - 2026-01-08
  • v1.6.28 - 2025-12-24
  • v1.6.27 - 2025-12-23
  • v1.6.26 - 2025-12-22
  • v1.6.25 - 2025-12-21
  • v1.6.24 - 2025-12-17
  • v1.6.23 - 2025-12-17
  • v1.6.22 - 2025-12-14
  • 1.6.21 - 2025-12-09
  • v1.6.14 to v1.6.20
  • 1.6.14-alpha - 2025-12-05
  • 1.6.12 - 2025-11-28
  • 1.6.11 - 2025-11-21
  • 1.6.10 - 2025-11-20
  • 1.6.9 - 2025-10-29
  • 1.6.8 - 2025-10-27
  • 1.6.4 - 2025-10-26
  • 1.6.3 - 2025-10-25
  • 1.6.2 — 2025-10-04
  • 1.6.1 — 2025-10-04
  • 1.6.0 — 2025-09-23
  • 1.5.2 — 2025-03-01
  • 1.5.1 — 2025-02-07
  • 1.5 — 2025-01-28
  • 1.0.2.1 — 2023-09-11
  • 1.0.2 — 2023-08-28
  • 1.0 — 2023-07-29
  • 0.3.0 — 2023-07-26
  • 0.2.4 — 2023-07-24

Indices and tables

• Index

• Module Index

• Search Page