Skip to content

Modelling Performance Enhancement

Past due by 9 months 47% complete

Objective Overview:
Enhance modeling performance in WaveDiff by addressing centroiding issues, introducing a new PSF model class, refactoring validation test datasets, input SEDs, handling software deprecations, resolving bugs, and renaming modules and classes.
This milestone also includes establishing the WaveDiff workflow with guidelines on workflow, co…

Objective Overview:
Enhance modeling performance in WaveDiff by addressing centroiding issues, introducing a new PSF model class, refactoring validation test datasets, input SEDs, handling software deprecations, resolving bugs, and renaming modules and classes.
This milestone also includes establishing the WaveDiff workflow with guidelines on workflow, contributions, and code of conduct.

Key Objectives:
Centroiding Problem: Improve accuracy and reliability of centroid estimation.
New PSF Model Class: Reintroduce a refactored version of the PSF semi-parametric model class with a physical layer from v1.0.3.
Validation Dataset: Refactor the script to generate simulated datasets for validation
Software Deprecations and Upgrades: Update TensorFlow, Keras, or change optimisers to maintain compatibility.
Bug Fixes: Resolve known bugs from previous releases.
Module and Class Renaming: Mitigate naming conflicts and improve code clarity.

Timeline and Deliverables:
Duration: ~1 month

Loading