Examples

This section contains focused examples demonstrating specific pyAVS functionality. Each example page shows the real, runnable script from the examples/ directory alongside explanatory prose.

For the core AVSComposer workflow (the recommended entry point for most analyses), see the dedicated AVSComposer Guide instead of this section.

Tip

New to pyAVS? pyavs_colab_quickstart.ipynb is a ~15 minute, no-install Colab notebook covering raw MEG + eye tracking, plotting gaze on a scene image, and a fixation-locked MEG comparison – a gentler starting point than the scripts below, which assume a local data path is already configured.

Available Examples

• Source Reconstruction Examples
  • Quickstart: Synthetic Data
  • Real-Data, Config-Driven Workflow
  • ROI-Based Extraction
  • Population Codes
  • See Also
• Cross-Session Beamformer Filters
  • See Also
• Configuration System
• COCO-Stuff Object Detection
  • See Also
• Object Detection on Real Data
  • See Also

Source Reconstruction

From forward modeling to source localization and population code extraction, see Source Reconstruction Examples.

Cross-Session Beamformer Filters

Computing a single set of LCMV beamformer filters shared across a subject’s sessions, see Cross-Session Beamformer Filters.

Configuration

Setting up and inspecting the pyAVS data path / configuration system, see Configuration System.

Object Detection on Scenes

Mapping fixations to MS-COCO / COCO-Stuff object categories, both on synthetic (COCO-Stuff Object Detection) and real (Object Detection on Real Data) data.

Complete Example Scripts

All working example scripts live in the examples/ directory at the repository root. We recommend starting with avs_composer_example.py (covered in AVSComposer Guide) to understand the core pyAVS workflow.