neural-python 0.0.2

NeuralPy is the Artificial Neural Network library implemented in Python.

NeuralPy is a Python library for Artificial Neural Networks. You can run and test different Neural Network algorithms.

Installation

$ pip install neural-python

Links

• Documentation

• Issues

• Tutorials

• Available algorithms

Dependence

• Python 2.7, 3.2, 3.3, 3.4

• NumPy >= 1.9.0

• SciPy >= 0.14.0

• Matplotlib >= 1.4.0

Next steps

• Bug fixing and version stabilization

• Speeding up algorithms

• Adding more algorithms

Library support

• Radial Basis Functions Networks (RBFN)

• Backpropagation and different optimization for it

• Neural Network Ensembles

• Associative and Autoasociative Memory

• Competitive Networks

• Step update algorithms for backpropagation

• Weight control algorithms for backpropagation

• Basic Linear Networks

Algorithms

• Backpropagation

  • Classic Gradient Descent

  • Mini-batch Gradient Descent

  • Conjugate Gradient

  • quasi-Newton

  • Levenberg-Marquardt

  • Hessian diagonal

  • Momentum

  • RPROP

  • iRPROP+

  • QuickProp

• Weight update rules

  • Weight Decay

  • Weight Elimination

• Learning rate update rules

  • Adaptive Learning Rate

  • Error difference Update

  • Linear search by Golden Search or Brent

  • Search than converge

  • Simple Step Minimization

• Ensembles

  • Mixture of Experts

  • Dynamically Averaged Network (DAN)

• Radial Basis Functions Networks (RBFN)

  • Generalized Regression Neural Network (GRNN)

  • Probabilistic Neural Network (PNN)

  • Radial basis function K-means

• Autoasociative Memory

  • Discrete BAM Network

  • CMAC Network

  • Discrete Hopfield Network

• Competitive Networks

  • Adaptive Resonance Theory (ART1) Network

  • Self-Organizing Feature Map (SOFM or SOM)

• Linear networks

  • Perceptron

  • LMS Network

  • Modified Relaxation Network

• Associative

  • OJA

  • Kohonen

  • Instar

  • Hebb

Tests

$ pip install tox
$ tox