dopingflow

ML-Driven High-Throughput Doping Workflow for Oxide Materials

The workflow integrates:

• Structure generation and symmetry-aware enumeration

• ML-based relaxation and screening (M3GNet, ALIGNN)

• Formation energy calculations using configurable thermodynamic references

• Automated database collection

• Fully reproducible, stage-isolated execution

All stages are controlled through a single input.toml file.

User Guide

• Workflow Overview
  • Conceptual Pipeline
  • Pipeline Structure
  • Stages
• Installation, Usage, and Outputs
  • Installation
  • Required Inputs
  • Running the Workflow
  • Outputs Overview
  • Tips
• Required Input Files
  • Overview
  • Directory Layout Example
  • ALIGNN Model Directory (Environment Variable)
  • Important Notes
  • Summary
• Input File Specification (input.toml)
  • Execution Model
  • Execution Behavior
  • [references]
  • [structure]
  • [doping]
  • [generate]
  • [scan]
  • [relax]
  • [filter]
  • [bandgap]
  • [formation]
  • [database]
  • Design Principles

Workflow Stages

The workflow is organized into modular stages. Each stage can be executed independently.

• 0. Reference Energy Construction
• 1. Structure Generation
• 2. Symmetry-Reduced Energy Pre-screening (Scanning)
• 3. Candidate Relaxation
• 4. Relaxed-Candidate Filtering
• 5. Bandgap Prediction
• 6. Formation Energy Evaluation
• 7. Database Collection

Examples

• Explicit Example — Single Target Composition
• Explicit Example — Single Target Composition using Oxide References
• Explicit Example — Multiple Target Compositions
• Enumerate Example — Systematic Dopant Screening
• Smoke Test — Minimal Fast Run

API Reference

• dopingflow

===== END =====