Neural nets prototyping framework
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
data
examples
protonn
tests
.gitignore
.travis.yml
LICENSE
MANIFEST.in
README.rst
fabfile.py
requirements.txt
setup.py
setup_boilerplate.py
test_requirements.txt

README.rst

[protoNN logo]

A framework for code-agnostic, interactive prototyping of DNNs.

build status from Travis CI https://coveralls.io/repos/github/undertherain/protoNN/badge.svg?branch=master pypi version

Features

  • Transparent and elastic scheduling of DNN training jobs on modern HPC systems.
  • Monitoring and visualizing model parameters and computational performance statistics.
  • Perform semi-automatic hyperparameter tuning/optimization and architecture search using evolutionary algorithms.
  • A user-defined interactive interface to drive the framework/ design process, not bound to any particular framework.
  • Scaling the functionality and performance of the model as the resources increase.

How do I get set up?

  • pip3 install protonn for latest stable release
  • pip3 install git+https://github.com/undertherain/protoNN.git for recent development version
  • Python 3.4 or later is required

Contributors

Aleksandr Drozd

Mohamed Wahib

Mateusz Bysiek

Maxim Shpakovich

For licensing information, please see LICENSE