Fast C++ library for linear algebra (matrix maths) and scientific computing. Easy to use functions and syntax, deliberately similar to Matlab. Uses template meta-programming techniques. Also provides efficient wrappers for LAPACK, BLAS, ATLAS, ARPACK and SuperLU libraries, including high-performance versions such as OpenBLAS and Intel MKL. Useful for machine learning, pattern recognition, signal processing, bioinformatics, statistics, finance, etc. For more details, see http://arma.sourceforge.net
Java Neural Network Framework
Neuroph is lightweight Java Neural Network Framework which can be used to develop common neural network architectures. Small number of basic classes which correspond to basic NN concepts, and GUI editor makes it easy to learn and use.
Speech recognition research toolkit
lightweight GPU-based sparse matrix-vector multiplication (SpMV)
LightSpMV is a novel CUDA-compatible sparse matrix-vector multiplication (SpMv) algorithm using the standard compressed sparse row (CSR) storage format. We have evaluated LightSpMV using various sparse matrices and further compared it to the CSR-based SpMV subprograms in the state-of-the-art CUSP and cuSPARSE. Performance evaluation reveals that on a single Tesla K40c GPU, LightSpMV is superior to both CUSP and cuSPARSE, with a speedup of up to 2.60 and 2.63 over CUSP, and up to 1.93 and 1.79 over cuSPARSE for single and double precision, respectively.
Workflow Designer, Hive Editor, Pig Editor, File System Browser
Flamingo is a open-source Big Data Platform that combine a Ajax Rich Web Interface + Workflow Engine + Workflow Designer + MapReduce + Hive Editor + Pig Editor. 1. Easy Tool for big data 2. Use comfortable in Hadoop EcoSystem projects 3. Based GPL V3 License Supporting Pig IDE, Hive IDE, HDFS Browser, Scheduler, Hadoop Job Monitoring, Workflow Engine, Workflow Designer, MapReduce.
This is a C++ implementation of the original C-IL2P system, invented by Artur D'Avila Garcez and Gerson Zaverucha. C-IL2P is a neural-symbolic learning system which uses a propositional logic program to create a three-layer recursive neural network and uses back-propagation to learn from examples.
Neural Networks Collection
This project implements in C++ a bunch of known Neural Networks. So far the project implements: LVQ in several variants, SOM in several variants, Hopfield network and Perceptron. Other neural network types are planned, but not implemented yet. The project can run in two modes: command line tool and Python 7.2 extension. Currently, Python version appears more functional, as it allows easy interaction with algorithms developed by other people.
A forensic file identification tool using neural networks
Just carved a bunch of bytes and have no idea what they could be? Maybe ANNFiD can help. ANNFiD uses neural network to identify byte patterns. It can be trained and has a GUI to help in the process. The tool is still on a very early stage, but could improve exponentially with the help of the developer community
experimental C++ library for GPU based linear algebra
In-development neural-symbolic system, CILP++
This is the project of the extension of the original C-IL2P neural-symbolic system, CILP++, for reasoning, knowledge extraction and theory revision from propositional and first-order logics. *** ATTENTION! The CILP++ project has been discontinued from SVN! It has been migrated to Github on https://github.com/manoelfranca/cilppp!
A category-based approach to exploring film data.
Cinefile is a prototype of a category-based method of database exploration. It allows the user to identify abstract categories of films by providing examples of category members, learns to classify films as belonging or not belonging to those categories, and provides a graphical interface for exploring and comparing categories. Cinefile is designed to work with data retrieved from the Internet Movie Database (imdb.com). This data is used for classification and is the subject of the category-based analysis. Cinefile was developed by the University of Mary Washington's Computer Science department (http://cas.umw.edu/computerscience).
End-to-end big data in a massively scalable supercomputing platform.
HPCC Systems® (www.hpccsystems.com) from LexisNexis® Risk Solutions is a proven, open source solution for Big Data insights that can be implemented by businesses of all sizes. With HPCC Systems, developers can design applications with Big Data at their core, enabling businesses to better analyze and understand data at scale, improving business time to results and decisions. HPCC Systems offers a consistent data-centric programming language, two processing platforms and a single, complete end-to-end architecture for efficient processing. Read our blog (http://hpccsystems.com/blog ), or connect with us on Twitter (@hpccsystems), Facebook (https://www.facebook.com/hpccsystems ) and LinkedIn (http://www.linkedin.com/company/hpcc-systems) HPCC Systems is available on AWS & can be configured through the Instant Cloud Solution. The download here is a VM.
SOM - Self-Organizing Maps of Teuvo Kohonen
It's a "Hello World" implementation of SOM (Self-Organizing Map) of Teuvo Kohonen, otherwise called as the Kohonen map or Kohonen artificial neural networks.
Java implementation of some document classification algorithms.
Analogical Modeling module for Java
Analogical Modeling is an exemplar-based approach to machine learning which imitates human behavior in outcome prediction. Its design has been applied to many natural language and other phenomena which exhibit variable behavior. A Perl XS implementation is available from http://humanities.byu.edu/am/ . This project is a Java implementation of the same. For more information on Analogical Modeling, see http://en.wikipedia.org/wiki/Analogical_modeling .
A Lexical Substitution Framework
Lexical substitution framework for supervised all-words lexical substitution using delexicalized features. For a runnable (but GPL-licensed) version of LexSub, see LexSub-GPL (sf.net/p/lexsub/lexsub-gpl)
Parallel pairwise correlation computation on Intel Xeon Phi clusters
The first parallel and distributed library for pairwise correlation/dependence computation on Intel Xeon Phi clusters. This library is written in C++ template classes and achieves high speed by exploring the SIMD-instruction-level and thread-level parallelism within Xeon Phis as well as accelerator-level parallelism among multiple Xeon Phis. To facilitate balanced workload distribution, we have proposed a general framework for symmetric all-pairs computation by building provable bijective functions between job identifier and coordinate space for the first time.
Onyx is for rapid prototyping and large-scale experimentation on advanced machine-learning algorithms with an emphasis on algorithms for online or streaming analysis, modeling, and classification.
This project aims to provide a simple python interface for CBR
Case base reasoninig is one of the primitive AI techniques in existance. Infact it's one of the laziest. Implementation however takes some effort. Python is one of the most to used languages that is becoming popular in every community for its simplicity & ease of learning. It has an interface for wordnet (through nltk tools) which brings us why pyCBR exists. This script is still in its infant stage of CBR & script matching. Nontheless you could use it to do many crazy stuff. Enjoy!!!
A slim cluster framework with a wide range of features
Stochastico is an implementation of stochastic discrimination for pattern recognition, predictive modeling and data mining applications.
A tool that converts CCGBank to PTB
Conversion between different grammar frameworks is of great importance to comparative performance analysis of the parsers developed on them. This tool can convert CCG derivations to PTB trees by using Max Entropy models as well as visualizing the tree graphs. The main technical innovation presented here is the effective conversion method which achieves a F score over 95%.
A neural network library for Java.
Cognity is an object-oriented neural network library for Java. It's goal is to provide easy-to-use, high level architecture for neural network computations along with reasonable performance.