Harry is a small tool for comparing strings and measuring their similarity. It implements several common distance and kernel functions for strings, as well as some exotic similarity measures. For example, Harry supports the Levenshtein (edit) distance, the Jaro-Winkler distance, and the compression distance. Harry is implemented using OpenMP, so its runtime scales linearly with the number of available CPU cores. Efficient implementations and effective caching speed comparison of strings.
Erudite is an application for training and testing back propogation neural networks using the ANNeML (Artifical Neural Network Markup Language) XML format. It supports testing and training neural nets with CSV files and has support for randomized training sets, optional adapting learning rate, sigmoid or hyperbolic tangent transfer functions, optional bias and weight adjustment locking, and more.
Salad (short for Letter Salad) is an efficient and flexible implementation of the well-known anomaly detection method Anagram by Wang et al. (RAID 2006). Salad is based on n-gram models, that is, data is represented as all of its substrings of length n. During training these n-grams are stored in a Bloom filter. This enables the detector to represent a large number of n-grams in little memory and still being able to efficiently access the data. Salad extends Anagram by allowing various n-gram types, a 2-class version of the detector for classification, and various model analysis modes.
LibBi is used for state-space modelling and Bayesian inference on high-performance computer hardware, including multi-core CPUs, many-core GPUs (graphics processing units), and distributed-memory clusters. The staple methods of LibBi are based on sequential Monte Carlo (SMC), also known as particle filtering. These methods include particle Markov chain Monte Carlo (PMCMC) and SMC2. Other methods include the extended Kalman filter and some parameter optimization routines. LibBi consists of a C++ template library and a parser and compiler, written in Perl, for its own modelling language.
Biomechanical ToolKit (BTK) is a cross-platform library for biomechanical analysis. It can read and write a large variety of file formats used in biomechanics, and can modify them. All these operations can be done with the C++ API or with the wrappers included (Python, Octave, and Matlab). The goal of this project is to help the community share data without the restriction of the file format or the biomecanical model provided by the manufacturer of the acquisition system.
iNA is a computational tool for quantitative analysis of fluctuations in biochemical reaction networks. Such fluctuations, also known as intrinsic noise, arise due to the stochastic nature of chemical reactions and cannot be ignored for when some molecules are present only in very low copy numbers as is the case in living cells. The SBML-based software computes statistical measures such as means and standard deviations of concentrations within a given accuracy using the analytical system size expansion. The result of iNA’s analysis can be tested against the computationally much more expensive stochastic simulation algorithm.
The Common Pipeline Library provides a highly robust set of functions for manipulating signals and images. It is primarily intended for the building of VLT instrument pipelines, but is also useful for generic data handling. It includes a number of useful low-level data types, medium-level data access methods, standard implementations of commonly-used signal processing and data reduction tasks, and dynamic loading of "recipes" for data processing.
SHTns is a high-performance Spherical Harmonic Transform library. It was designed for numerical simulation (fluid flows, mhd, etc.) in spherical geometries, but can be used for any kind of problem involving scalar or vector spherical harmonics. It is very fast, thanks to careful vectorization and runtime tuning. It supports multi-threaded transforms via OpenMP. It features scalar and vector transforms, synthesis and analysis, and flexible truncation and normalization. A Python interface is included.