Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
Metatrain
Light Logo Dark Logo
Metatrain on GitHub
  • Installation
  • Getting started
    • Quickstart
    • Training YAML Reference
    • Override Architecture’s Default Parameters
    • Restarting and Checkpoints
    • Units
  • Available Architectures
    • Classifier (Experimental)
    • Composition
    • DPA3 (Experimental)
    • FlashMD (Experimental)
    • Symplectic FlashMD (Experimental)
    • GAP
    • LLPR
    • MACE (Experimental)
    • PET
    • Scaler
    • SOAP-BPNN
    • SPACE (Experimental)
  • Tutorials
    • Beginner Tutorials
      • Basic Usage
      • How to prepare data for training
      • Fine-tuning a pre-trained model
      • Training a model from scratch
      • Model validation with parity plots
      • Running molecular dynamics with ASE
    • Advanced Tutorials
      • Transfer Learning (experimental)
      • Computing LLPR uncertainties
      • Training a model with ZBL corrections
      • Fitting generic targets
      • Training or fine-tuning a FlashMD model
      • Multi-GPU training
      • Training with Mixed Stress Structures
      • Training a DOS model
      • Generating and training an LLPR-derived shallow ensemble model
      • Training a Classifier Model
      • Training a Symplectic FlashMD Model
      • Extracting Hidden Layer Features from PET
  • Concepts and Design
    • Dataset formats
    • Output naming
    • Fine-tune a pre-trained model
    • Loss functions
    • Target scaling
    • Auxiliary outputs
    • Training with atom-count-based batching
  • Frequently Asked Questions
  • Citing metatrain
  • Developer documentation
    • Contributing
    • Life Cycle of an Architecture
    • Adding a new architecture
    • Dataset Information
    • Adding a new loss function
    • CLI API
      • Train
      • Eval
      • Export
    • Utility API
      • Additive models
        • Removing additive contributions
        • ZBL short-range potential
      • Scaler models
        • Scaler model
        • Removing the scale from targets
      • Data
        • Combining dataloaders
        • Dataset
        • Reading a dataset
        • Readers
        • Target data Writers
        • Converting Systems to ASE
      • Abstract base classes
      • Architectures
      • Augmentation
      • Device
      • Dtype
      • Errors
      • Evaluating a model
      • External Naming
      • Hyperparameters
      • IO
      • Logging
      • Long-range
      • Loss
      • Metrics
      • Neighbor lists
      • Custom omegaconf functions
      • Output gradient
      • Averaging predictions per atom
      • Pydantic utilities
      • Summing over atoms
      • Testing Utilities
      • Data type and device transfers
      • Unit handling
    • Base hyperparameters
    • Changelog
Back to top
View this page
Edit this page

Available Architectures¶

This is a list of all architectures available in metatrain.

  • Classifier (Experimental)
  • Composition
  • DPA3 (Experimental)
  • FlashMD (Experimental)
  • Symplectic FlashMD (Experimental)
  • GAP
  • LLPR
  • MACE (Experimental)
  • PET
  • Scaler
  • SOAP-BPNN
  • SPACE (Experimental)
Next
Classifier (Experimental)
Previous
Units
Copyright © 2026, metatrain developers
Made with Sphinx and @pradyunsg's Furo