Python API reference¶
Curated pages for the public surface. Internal helpers are deliberately not documented here: importability is not a contract.
The stable surface is openauc and openauc.api. Model types are also
importable from openauc.models, and the subpackages openauc.formats,
openauc.plotting and openauc.synthetic carry their own facades.
| Page | Covers |
|---|---|
| Top-level API | load, available_formats, AUCX entry points |
| Experiments and metadata | AUCExperiment and the metadata models |
| Observations | The numeric layer and both radius modes |
| Validation and readiness | Reports, findings, readiness types |
| Summaries | ExperimentSummary and its components |
| Plotting | plot_scans, plot_scan |
| Synthetic generation | Config, generator and writers |
| Generic-delimited ingestion | Manifest and parser types |
| AUCX archives | Archive functions and records |
| Exceptions | The full hierarchy |
Import conventions¶
import openauc # light: no matplotlib
from openauc.models import ... # canonical model types
from openauc.plotting import ... # imports matplotlib on first draw
from openauc.synthetic import ... # generator
import openauc.api as api # the curated facade
openauc.api.plot_scans resolves lazily, so importing the facade does not pull
in matplotlib.
A note on what these types promise¶
Nothing in this API assesses scientific validity. validate() reports
structural consistency, assess_readiness() reports metadata presence, and
summary_data() reports counts and ranges. See
Scientific boundaries.