Skip to content

Validate OASIS sparsity parameter against Python package #42

Description

@MarcelMB

Summary

The penalty (sparsity) parameter in OasisConfig (config.py, lines 16-40) is passed to oasisAR2() as lam in neural.py (lines 61-128). This parameter may have slightly different semantics between the browser-based KTune tool (used to tune parameters) and the Python oasis package.

Action

  • Run identical deconvolution parameters through the Python oasis package directly and compare outputs to KTune results
  • Document the exact parameter mapping between KTune and oasisAR2(g1, g2, lam, s_min)
  • If there is a discrepancy, adjust the config or add a conversion step

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions