Projects

scikit learn

Python module integrating various machine learning algorithms under a common interface. It offers a wide range of methods such as Support Vector Machines, linear models (L1, L2 penalized), logistic regression, gaussian mixture models and more. The large number of algorithms aleady implemented allows ... [More] for easy comparison of accuracy and performance of various algorithms. [Less]

SHOGUN

The SHOGUN machine learning toolbox's focus is on large scale kernel methods and especially on Support Vector Machines (SVM). It comes with a generic interface for SVMs, features several SVM and kernel implementations, includes LinAdd optimizations and also Multiple Kernel Learning algorithms. ... [More] SHOGUN also implements a number of linear methods. It allows the input feature-objects to be dense, sparse or strings and of type int/short/double/char. It provides efficient implementations several kernels but also linear methods, hidden markov models etc. and interfaces to matlab,octave,python,R and has a cmdline interface and allows C++ extensions via a library. [Less]

Accord.NET Framework

Accord.NET Framework is a C# framework which extends the excellent AForge.NET Framework with new tools and libraries. The framework is comprised by libraries and sample applications demonstrating their features. Some of the libraries include: Accord.Statistics - library with statistical ... [More] analysis and other tools; Accord.Imaging - extension to the AForge.NET Imaging library with new filters and routines; Accord.Neuro - extension to the AForge.NET Neuro library with other learning algorithms; Accord.MachineLearning - extension to AForge's machine learning library with Support Vector Machines; Accord.Audio - experimental library with filters and audio processing routines. [Less]

PyBrain

PyBrain is a modular Machine Learning Library for Python. It's goal is to offer flexible, easy-to-use yet still powerful algorithms for Machine Learning Tasks and a variety of predefined environments to test and compare your algorithms. It's the Swiss army knife for machine learning and neural networking.

corpusevaluator

Coeval is a free Corpus Evaluation software written in Java.It allows you to create, manage and customize your own corpus of documents. Coeval can be used to train classifiers, evaluate performance and cross-compare classifiers on the same corpus. A Support Vector Machine classifier (LIBSVM -- ... [More] A Library for Support Vector Machines) is provided with this release and it also allows you easily add and test out your classifiers. RequirementsApache Tomcat 6.x MySQL 5.1 Java EE 5 Eclipse 3.x Instructionsopen mysql command line, type and execute "source COEVAL_HOME/db/dump.sql" edit COEVAL_HOME/WEB-INF/mysql.properties with your mysql account data edit COEVAL_HOME/localhost.properties with your Apache Tomcat account data load with ant "build.xml" and deploy Coeval Contact MePlease feel free to contact me if you have any questions or comments. [Less]

support-vector-machine-bioinformatics

hgig'pi[oi[pihoij;,k;'

torchit

The aim of this project is to implement a realistic application of artificial neural networks. Our work will be based on wisconsin breast cancer database (http://mlearn.ics.uci.edu/databases/breast-cancer-wisconsin/) and use the C++-based Torch machine learning library to implement BackPropagation ... [More] , EM with GMMs (initialized with the K-means algorithm) and an SVM algorithm. [Less]