Open-source .NET neural network library? [closed]

Answer 1

Download Weka and convert from the jar file (Java bytecode) to .NET Framework (managed) dll through IKVM. Weka is aknowledged by many as a very good open-source library for machine learning. It includes neural networks.

Answer 2

From what I can see: FANN fits both of the categories.

Answer 3

NeuronDotNet It is not as broad in its scope as AForge, but its exclusive focus on NN makes it maybe more readily accessible, and possibly more featured in this area.

Update: (December 2012)
NeuronDotNet's old repository at http://neurondotnet.freehostia.com is defunct.
The most recent source code, Version 3.0 is available on sourceforge, but this project is effectively inactive. Apparently, its original developer, and sole contributor, Vijeth Dinesha, has stopped updating and improving this code base. Thank you, Vijeth! Maybe someone will, some day, pick-up from where you left. Indeed, NeuronDotNet is a relatively mature framework and its simplicity and single focus make it attractive, compared with other frameworks where Neural Nets are just one feature/module therein.

Answer 4

https://dotnet.github.io/infer/ Infer.NET is a framework for running Bayesian inference in graphical models.

Answer 5

Best .NET neural network library is AForge

Links:

MainPage

CodeProject-Article

Answer 6

Here is a list of F# related resources for open source .NET machine learning.

http://fsharp.org/machine-learning/

Frameworks available on NuGet: (The content below was taken directly from the above referenced URL for persistence purposes.)

• Accord.MachineLearning - Contains Support Vector Machines, Decision Trees, Naive Bayesian models, K-means, Gaussian Mixture models and general algorithms such as Ransac, Cross-validation and Grid-Search for machine-learning applications. This package is part of the Accord.NET Framework.

• Encog Machine Learning Framework - An advanced neural network and machine learning framework. Encog contains classes to create a wide variety of networks, as well as support classes to normalize and process data for these neural networks. Encog trains using multithreaded resilient propagation. Encog can also make use of a GPU to further speed processing time. A GUI based workbench is also provided to help model and train neural networks.

• Numl - A machine learning library intended to ease the use of using standard modeling techniques for both prediction and clustering