Arch Linux User Repository

Search Criteria

Enter search criteria

Search by

Keywords

Out of Date

Sort by

Sort order

Per page

Package Details: nemesyst-git 2.0.6.r6.68bebbc-1

Package Actions

Git Clone URL:	https://aur.archlinux.org/nemesyst-git.git (read-only, click to copy)
Package Base:	nemesyst-git
Description:	Practical, distributed, hybrid-parallelism, deep learning framework.
Upstream URL:	https://github.com/DreamingRaven/nemesyst
Keywords:	deep-learning framework machine-learning neural-network
Licenses:	OSL-3.0
Provides:	nemesyst
Submitter:	GeorgeRaven
Maintainer:	GeorgeRaven
Last Packager:	GeorgeRaven
Votes:	1
Popularity:	0.000000
First Submitted:	2018-09-20 14:43 (UTC)
Last Updated:	2020-04-30 10:58 (UTC)

Dependencies (15)

python (python37^AUR, python311^AUR, python310^AUR)
python-configargparse (python-configargparse-git^AUR)
python-ezdb-git^AUR
python-future
python-pymongo
python-setuptools
git (git-git^AUR, git-gl^AUR) (make)
python-sphinx (python-sphinx-git^AUR) (make)
python-sphinx-argparse (make)
python-sphinx_rtd_theme (make)
mongodb^AUR (mongodb32-bin^AUR, percona-server-mongodb-bin^AUR, mongodb36-bin^AUR, mongodb34-bin^AUR, mongodb40-bin^AUR, mongodb44-bin^AUR, mongodb42-bin^AUR, mongodb60-bin^AUR, mongodb50-bin^AUR, mongodb50^AUR, ferretdb^AUR, mongodb70-bin^AUR, ferretdb-bin^AUR, mongodb^AUR, mongodb-bin^AUR, mongodb44^AUR) (optional) – local database creation
python-keras (python-keras-git^AUR) (optional) – example neural networks
python-keras-applications (optional) – example neural networks
python-keras-preprocessing (python-keras) (optional) – example neural networks
python-scikit-learn (python-scikit-learn-git^AUR) (optional) – mnist example auto data fetcher

Required by (0)

Sources (1)

nemesyst

Pinned Comments

GeorgeRaven commented on 2018-09-24 14:47 (UTC) (edited on 2018-09-24 14:51 (UTC) by GeorgeRaven)

Important Installation Note:

This package does not install a Keras back-end for you which is a requirement for Nemesyst to function at all. And does not include any back ends in optional dependencies. This is a decision made on two grounds:

Keras can use a choice selection of back-ends and we do not want to force usage of one or another by putting one in the dependency array, e.g python-tensorflow-cuda vs python-tensorflow, a user may require one over another lets say on gpu-less server systems.
While we could put these in optdepends we feel this is inappropriate, illustrated by the wording of the pkgbuild wiki: "[optdepends] An array of packages that are not needed for the software to function, but provide additional features." Since you most certainly do require a back-end of your choice we have opted not to include them in opt-depends either.

Latest Comments

GeorgeRaven commented on 2018-09-24 14:47 (UTC) (edited on 2018-09-24 14:51 (UTC) by GeorgeRaven)

Important Installation Note:

Keras can use a choice selection of back-ends and we do not want to force usage of one or another by putting one in the dependency array, e.g python-tensorflow-cuda vs python-tensorflow, a user may require one over another lets say on gpu-less server systems.
While we could put these in optdepends we feel this is inappropriate, illustrated by the wording of the pkgbuild wiki: "[optdepends] An array of packages that are not needed for the software to function, but provide additional features." Since you most certainly do require a back-end of your choice we have opted not to include them in opt-depends either.