Skip to content

Using JATIC tools

Dataset evaluation

  • Dataeval: Bias

    The Dataeval bias detection tools identify biases or correlations present in a dataset that may lead to shortcut learning in models.

  • Dataeval: Feasibility

    The Dataeval feasiblity tools measure the irreducible error of a dataset.

  • Dataeval: Cleaning

    The Dataeval cleaning tools analyze datasets to identify and remove invalid, irrelevant, or low-quality data, such as duplicates and statistical outliers.

  • Dataeval: Shift

    The Dataeval shift tool identifies statistical differences between operational data and training data

Model evaluation

  • NRTK

    The Natural Robustness Toolkit (NRTK) is an open source toolkit for generating operationally realistic perturbations to evaluate the natural robustness of computer vision algorithms.

  • XAITK

    The Explainable AI Toolkit for saliency (XAITK - Saliency) is an open source framework for visual saliency algorithm interfaces and implementations.

Results & analysis

  • Analytics Store

    The analytics store persists capability results as queryable records, enabling SQL-based comparison across runs, datasets, and capabilities.

Job submission

  • Ray Simple Job Submission

    Run a capability asynchronously with the lightweight process-local Ray task backend.

  • Ray Job Submission

    Run a capability asynchronously with the registry-backed Ray backend for tracked and reattachable jobs.