User guide¶
The user guide explains what pyTomoAO computes and how to control it. Read it in order the first time; afterwards the configuration reference is the page you will return to most.
Tomographic reconstruction, the MMSE estimator, and the objects pyTomoAO builds.
Every section and key of the YAML file, with units and validation rules.
Model-based versus interaction-matrix-based reconstruction, and regularisation.
Influence functions, slope ordering, scaling and actuator masking.
How the CuPy path is selected, precision options and how to force the CPU.
The workflow in one picture¶
flowchart TD
A[YAML configuration] --> B[tomographicReconstructor]
B --> C{Interaction matrix supplied?}
C -->|No| D[Model-based reconstructor R<br/>slopes to phase]
C -->|Yes| E[IM-based reconstructor R<br/>slopes to DM commands]
D --> F[assemble_reconstructor_and_fitting]
F --> G[FR: slopes to DM commands]
E --> G
The two branches differ in where the DM enters the problem. The model-based path estimates the phase over the pupil and then fits the DM influence functions to it; the IM-based path folds a measured interaction matrix into the estimator so that the output is already in command space.