HamGNN User Guide

Contents

• HamGNN Introduction
  • Key Features
  • Application Areas
  • Latest Developments
• Environment Configuration and Installation
  • Python Environment
  • Third-party DFT Tool Support
  • HamGNN Installation Steps
• Construction of graph_data.npz File
  • General Process
  • OpenMX Process
  • SIESTA/HONPAS Process
  • ABACUS Process
• Model Training Process
  • Training Mode Classification
  • Training Command
  • Key Configuration Items
  • Training Monitoring
• Model Prediction Operation
  • Preparation for Prediction
  • Execute Prediction
  • Output Results
• Band Structure Calculation
  • Operation Process
  • Band Calculation for Large Systems
• Uni-HamGNN Universal Model
  • Model Introduction
  • Input Requirements
  • Usage Process
  • Output Results
• HamGNN Parameter Details
  • setup (Basic Settings)
  • profiler_params (Profiler Parameters)
  • dataset_params (Dataset Parameters)
  • losses_metrics (Loss Functions and Evaluation Metrics)
  • optim_params (Optimizer Parameters)
  • output_nets.HamGNN_out (Output Network Parameters)
  • representation_nets.HamGNN_pre (Representation Network Parameters)
  • Parameter Adjustment Recommendations

This section provides a complete guide for using and configuring the HamGNN software, including installation, data preparation, model training and prediction, band structure calculation, and other full workflow operations.