Skip to main content

BackPACK: Packing more into backprop

Project description

BackPACK BackPACK: Packing more into backprop

Travis Coveralls Python 3.7+

BackPACK is built on top of PyTorch. It efficiently computes quantities other than the gradient.

Provided quantities include:

  • Individual gradients from a mini-batch
  • Estimates of the gradient variance or second moment
  • Approximate second-order information (diagonal and Kronecker approximations)

Motivation: Computation of most quantities is not necessarily expensive (often just a small modification of the existing backward pass where backpropagated information can be reused). But it is difficult to do in the current software environment.


pip install backpack-for-pytorch



BackPACK is actively being developed. We are appreciating any help. If you are considering to contribute, do not hesitate to contact us. An overview of the development procedure is provided in the developer README.

How to cite

If you are using BackPACK, consider citing the paper

    title     = {Back{PACK}: Packing more into Backprop},
    author    = {Felix Dangel and Frederik Kunstner and Philipp Hennig},
    booktitle = {International Conference on Learning Representations},
    year      = {2020},
    url       = {}
BackPACK is not endorsed by or affiliated with Facebook, Inc. PyTorch, the PyTorch logo and any related marks are trademarks of Facebook, Inc.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

backpack-for-pytorch-1.5.2.tar.gz (1.2 MB view hashes)

Uploaded source

Built Distribution

backpack_for_pytorch-1.5.2-py3-none-any.whl (183.2 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page