VROOM-SBI Trained Models

Trained neural posterior estimators for Rotation Measure (RM) synthesis.

Model Information

• Model Types: burn, external, faraday, internal
• Max Components: 1
• Classifier: No
• Upload Date: 2026-02-17

Files

File	Description
posterior_burn_slab_n1.pt	Posterior model (.pt)
posterior_external_dispersion_n1.pt	Posterior model (.pt)
posterior_faraday_thin_n1.pt	Posterior model (.pt)
posterior_internal_dispersion_n1.pt	Posterior model (.pt)
training_burn_slab_n1.png	Training plot
training_external_dispersion_n1.png	Training plot
training_faraday_thin_n1.png	Training plot
training_internal_dispersion_n1.png	Training plot
training_summary.txt	Summary file

Usage

from vroom_sbi.inference import InferenceEngine

# Load models
engine = InferenceEngine(model_dir="path/to/downloaded/models")
engine.load_models()

# Run inference
result, all_results = engine.infer(qu_obs)
print(f"Best model: {result.n_components} components")

Citation

If you use these models, please cite:

[Citation information to be added]