# Examples

The examples are placeholders, to be written: each page states what it will show. The sections follow the workflow in order: BART’s applications, encoding operators, an application assembled from operators, and reconstructions.

For the operator inventory and advanced methods, see {doc}`/gallery/encoding`, {doc}`/gallery/research` and {doc}`/gallery/learning_resources`.

## Applications

Phantoms, FFTs and calibration, then coil preparation by prewhitening and compression. BART’s commands keep the coil axis where BART puts it.

Phantom, Fourier transform and calibration

Phantom, Fourier transform and calibration

Prewhitening, coil compression and sensitivity maps

Prewhitening, coil compression and sensitivity maps

## Encoding operators

Operators with explicit domain and codomain shapes. Check the adjoint before using an encoding inside a solver.

A Cartesian encoding and its adjoint

A Cartesian encoding and its adjoint

Radial encoding and a direct Fourier reference

Radial encoding and a direct Fourier reference

An idealized multi-shot EPI encoding

An idealized multi-shot EPI encoding

Wave encoding and temporal subspaces

Wave encoding and temporal subspaces

## An application from operators

BART operators and a Python callback composed into an encoding, solved without shell commands or CFL files.

A reconstruction with a Python operator

A reconstruction with a Python operator

## Reconstructions

Parallel imaging, model-based parameter mapping, and a small DeepInverse learning example. Measured-data reproductions are described in the {doc}`research recipes </gallery/research>`.

Parallel imaging with ESPIRiT and PICS

Parallel imaging with ESPIRiT and PICS

Model-based multi-echo reconstruction

Model-based multi-echo reconstruction

DeepInverse physics and a learned reconstruction

DeepInverse physics and a learned reconstruction