MAIA provides a track/cluster workflow for a muon-collider detector study. Conversion, TFDS building, and local CPU training are implemented, but data validation retains failed gates and incomplete detector metadata.
Physics sample¶
| Physics sample | TFDS dataset | Schema version | Model recipe |
|---|---|---|---|
| Top-quark pairs | maia_edm_ttbar_pf | 1.1.0 | pyg-maia-v1 |
The recipe configures partitions 1--10. Public availability must be checked independently; MAIA is not part of the verified Hub selection.
Event view¶
These views show the same simulated ttbar event (index 0) in the transverse plane. The particle panel shows the Pandora reference reconstruction from the local MAIA sample.
| Tracks and clusters | Pandora reconstructed particles |
|---|---|
 |  |
All-component overlay, vector input view and vector particle view are available for reuse. See event-view conventions for particle sources, sampling and scale.
Production and validation¶
Review productions.maia in particleflow_spec.yaml before following dataset generation. The builder is under mlpf/heptfds/maia_pf_edm4hep/. The detector-comparison runbook records source assumptions, validation failures, and local smoke coverage. Passing a short training run does not resolve the detector-data limitations.