https://github.com/torch/nn

https://github.com/torch/nn

Last synced: about 2 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/torch/nn
• Owner: torch
• License: other
• Created: 2013-10-18T12:27:25.000Z (almost 11 years ago)
• Default Branch: master
• Last Pushed: 2021-01-12T19:56:45.000Z (over 3 years ago)
• Last Synced: 2024-05-12T22:36:03.163Z (4 months ago)
• Language: Lua
• Homepage:
• Size: 3.58 MB
• Stars: 1,328
• Watchers: 109
• Forks: 970
• Open Issues: 171
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md

Awesome Lists containing this project

README

        
[![Build Status](https://travis-ci.org/torch/nn.svg?branch=master)](https://travis-ci.org/torch/nn)



# Neural Network Package #

This package provides an easy and modular way to build and train simple or complex neural networks using [Torch](https://github.com/torch/torch7/blob/master/README.md):

 * 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](doc/module.md#nn.Module): abstract class inherited by all modules;

   * [Containers](doc/containers.md#nn.Containers): composite and decorator classes like [`Sequential`](doc/containers.md#nn.Sequential), [`Parallel`](doc/containers.md#nn.Parallel), [`Concat`](doc/containers.md#nn.Concat) and [`NaN`](doc/containers.md#nn.NaN);

   * [Transfer functions](doc/transfer.md#nn.transfer.dok): non-linear functions like [`Tanh`](doc/transfer.md#nn.Tanh) and [`Sigmoid`](doc/transfer.md#nn.Sigmoid);

   * [Simple layers](doc/simple.md#nn.simplelayers.dok): like [`Linear`](doc/simple.md#nn.Linear), [`Mean`](doc/simple.md#nn.Mean), [`Max`](doc/simple.md#nn.Max) and [`Reshape`](doc/simple.md#nn.Reshape);

   * [Table layers](doc/table.md#nn.TableLayers): layers for manipulating `table`s like [`SplitTable`](doc/table.md#nn.SplitTable), [`ConcatTable`](doc/table.md#nn.ConcatTable) and [`JoinTable`](doc/table.md#nn.JoinTable);

   * [Convolution layers](doc/convolution.md#nn.convlayers.dok): [`Temporal`](doc/convolution.md#nn.TemporalModules),  [`Spatial`](doc/convolution.md#nn.SpatialModules) and [`Volumetric`](doc/convolution.md#nn.VolumetricModules) convolutions;

 * Criterions compute a gradient according to a given loss function given an input and a target:

   * [Criterions](doc/criterion.md#nn.Criterions): a list of all criterions, including [`Criterion`](doc/criterion.md#nn.Criterion), the abstract class;

   * [`MSECriterion`](doc/criterion.md#nn.MSECriterion): the Mean Squared Error criterion used for regression;

   * [`ClassNLLCriterion`](doc/criterion.md#nn.ClassNLLCriterion): the Negative Log Likelihood criterion used for classification;

 * Additional documentation:

   * [Overview](doc/overview.md#nn.overview.dok) of the package essentials including modules, containers and training;

   * [Training](doc/training.md#nn.traningneuralnet.dok): how to train a neural network using [`StochasticGradient`](doc/training.md#nn.StochasticGradient);

   * [Testing](doc/testing.md): how to test your modules.

   * [Experimental Modules](https://github.com/clementfarabet/lua---nnx/blob/master/README.md): a package containing experimental modules and criteria.