Neural Network Package

This package provides an easy and modular way to build and train simple or complex neural networks using Torch:

• Modules are the bricks used to build neural networks. Each are themselves neural networks, but can be combined with other networks using containers to create complex neural networks:
  • Module: abstract class inherited by all modules;
  • Containers: composite and decorator classes like Sequential, Parallel, Concat and NaN;
  • Transfer functions: non-linear functions like Tanh and Sigmoid;
  • Simple layers: like Linear, Mean, Max and Reshape;
  • Table layers: layers for manipulating tables like SplitTable, ConcatTable and JoinTable;
  • Convolution layers: Temporal, Spatial and Volumetric convolutions;
• Criterions compute a gradient according to a given loss function given an input and a target:
  • Criterions: a list of all criterions, including Criterion, the abstract class;
  • MSECriterion: the Mean Squared Error criterion used for regression;
  • ClassNLLCriterion: the Negative Log Likelihood criterion used for classification;
• Additional documentation:
  • Overview of the package essentials including modules, containers and training;
  • Training: how to train a neural network using StochasticGradient;
  • Testing: how to test your modules.
  • Experimental Modules: a package containing experimental modules and criteria.