recipe deeplift

DeepLIFT (Deep Learning Important FeaTures)

Homepage:

https://github.com/kundajelab/deeplift

Documentation:

https://github.com/kundajelab/deeplift/blob/master/README.md

License:

OTHER / MIT License

Recipe:

/deeplift/meta.yaml

Algorithms for computing importance scores in deep neural networks.

Implements the methods in "Learning Important Features Through Propagating Activation Differences" by Shrikumar, Greenside & Kundaje, as well as other commonly-used methods such as gradients, guided backprop and integrated gradients. See https://github.com/kundajelab/deeplift for documentation and FAQ.

package deeplift

(downloads) docker_deeplift

versions:

0.6.13.0-00.6.12.0-00.6.10.0-00.6.9.3-00.6.9.1-00.6.9.0-0

depends numpy:

>=1.9

depends python:

requirements:

Installation

You need a conda-compatible package manager (currently either micromamba, mamba, or conda) and the Bioconda channel already activated (see set-up-channels).

While any of above package managers is fine, it is currently recommended to use either micromamba or mamba (see here for installation instructions). We will show all commands using mamba below, but the arguments are the same for the two others.

Given that you already have a conda environment in which you want to have this package, install with:

   mamba install deeplift

and update with::

   mamba update deeplift

To create a new environment, run:

mamba create --name myenvname deeplift

with myenvname being a reasonable name for the environment (see e.g. the mamba docs for details and further options).

Alternatively, use the docker container:

   docker pull quay.io/biocontainers/deeplift:<tag>

(see `deeplift/tags`_ for valid values for ``<tag>``)

Download stats