Analyse and Export Results

Once FindMuonWorkChain finishes you have several ways to inspect and export the results.

---

Workflow outputs overview

Output	Type	Always present?	Description
all_index_uuid	orm.Dict	Yes	Mapping from site index to AiiDA node UUID
all_sites	orm.Dict	Yes	All relaxed sites (before clustering)
unique_sites	orm.Dict	Yes	Unique candidate sites after symmetry clustering
unique_sites_hyperfine	orm.Dict	Magnetic only	Contact hyperfine field (T) per unique site
unique_sites_dipolar	orm.List	Magnetic only	Classical dipolar field (T) per unique site

---

Inspecting outputs in Python

from aiida import load_profile, orm
load_profile()

node = orm.load_node(<PK>)

# All relaxed structures and their energies
all_sites = node.outputs.all_sites.get_dict()
for label, data in all_sites.items():
    print(f"{label}: position={data['position']}, energy_diff={data['energy']} eV")

# Unique sites
unique = node.outputs.unique_sites.get_dict()
for label, data in unique.items():
    print(f"{label}: position={data['position']}")

---

Exporting to a pandas DataFrame

The aiida_muon.utils.export_findmuon module provides a high-level helper that assembles a human-readable summary table:

from aiida_muon.utils.export_findmuon import get_clustering_after_run, get_distortions

# Summary table (unique sites, energies, cluster mapping)
df = get_clustering_after_run(node)
print(df.to_string())

The DataFrame includes columns such as:

• muon_index — index of the relaxed supercell
• label — cluster label
• position — fractional coordinates of the relaxed muon
• energy_diff — energy relative to the lowest-energy site (eV)
• multiplicity — number of symmetry-equivalent sites in the unit cell

Per-site distortions

from aiida.orm import StructureData

# Load the unrelaxed and relaxed supercell nodes
unrelaxed = orm.load_node(uuid_unrelaxed).get_ase()
relaxed   = orm.load_node(uuid_relaxed).get_ase()

distortions = get_distortions(unrelaxed, relaxed)
print("Distortion norms (Å):", distortions['distortion_norm'])
print("Δ distances from muon (Å):", distortions['delta_distance'])

---

Accessing individual relaxation calculations

The mapping between site index and AiiDA UUID stored in all_index_uuid lets you drill into any individual sub-calculation:

uuid_map = node.outputs.all_index_uuid.get_dict()
# e.g. {'0': 'xxxxxxxx-...', '1': 'yyyyyyyy-...', ...}

# Load one of the PwRelaxWorkChain nodes
pw_node = orm.load_node(uuid_map['0'])
print(pw_node.outputs.output_parameters.get_dict())

# Get the relaxed structure as ASE Atoms
ase_struct = pw_node.outputs.output_structure.get_ase()

---

Visualising structures

# Use ASE's built-in viewer
from ase.visualize import view

for label, data in unique.items():
    uuid = uuid_map.get(str(data['index']))
    if uuid:
        rlx_node = orm.load_node(uuid)
        view(rlx_node.outputs.output_structure.get_ase())

---

Querying the database

For large studies, use the AiiDA QueryBuilder to find all finished FindMuonWorkChain nodes:

from aiida.orm import QueryBuilder, WorkChainNode

qb = QueryBuilder()
qb.append(WorkChainNode, filters={
    'attributes.process_label': 'FindMuonWorkChain',
    'attributes.process_state': 'finished',
    'attributes.exit_status': 0,
})

results = qb.all(flat=True)
print(f"Found {len(results)} finished FindMuonWorkChain runs.")