Readme
AttriBench is a Pytorch <https://pytorch.org/>_-based implementation of
several metrics for the evaluation of feature attribution maps and methods.
AttriBench provides a functional and an object-oriented API for the computation
of these metrics, along with a set of utility functions for the necessary
preparations (e.g. computing attribution maps) as well as for the visualization
of the results.
The functional API is generally easier to use, and can be used to get started quickly if the scale of the evaluation is not too large. The object-oriented API is more flexible and can use multiple GPUs for evaluation of large datasets.
For more information, see the documentation <https://attribench.readthedocs.io/>_.
AttriBench can be installed from PyPI using pip:
.. code-block:: bash
pip install attribench
FAQs
A benchmark for feature attribution techniques
We found that attribench demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
CISA launched a new project called Vulnrichment to enrich CVEs with details that help prioritize patching and mitigation efforts, as the NVD backlog of unenriched CVEs awaiting analysis surpasses 10,000.
Security News
Socket is joining forces with CISA and other industry leaders at the RSA Conference to sign the Secure by Design pledge, committing to uphold the highest security standards in our products.
Research
The Socket research team breaks down a sampling of malicious packages that download and execute files, among other suspicious behaviors, targeting the popular Discord platform.