diff --git a/math/R-cran-combinat/pkg-descr b/math/R-cran-combinat/pkg-descr index 322223dcabdd..e27b37a777d3 100644 --- a/math/R-cran-combinat/pkg-descr +++ b/math/R-cran-combinat/pkg-descr @@ -1,17 +1 @@ -The R-cran-combinat package provides a collection of essential routines -for combinatorial mathematics within the R environment. Combinatorics is -a branch of mathematics concerning the study of finite or countable -discrete structures. - -This package offers functions to generate and manipulate various combinatorial -objects, including permutations, combinations, and partitions. It is -invaluable for researchers, statisticians, and data scientists who need -to perform tasks such as: - -- Generating all possible orderings of a set of items. -- Selecting subsets of items without regard to their order. -- Enumerating ways to divide a set into non-empty subsets. - -By providing these fundamental combinatorial tools, R-cran-combinat -facilitates a wide range of applications in probability, statistics, -computer science, and experimental design. +Routines for combinatorics. diff --git a/math/R-cran-conf.design/pkg-descr b/math/R-cran-conf.design/pkg-descr index de393137be7b..73da04d8c40a 100644 --- a/math/R-cran-conf.design/pkg-descr +++ b/math/R-cran-conf.design/pkg-descr @@ -1,24 +1,2 @@ -The R-cran-conf.design package provides a specialized set of tools -within the R environment for the construction and manipulation of -confounded and fractional factorial designs. These experimental designs -are fundamental in statistics and engineering for efficiently studying -the effects of multiple factors on an outcome, especially when resources -are limited. - -Confounded designs allow for the study of a large number of factors -with a smaller number of experimental runs by strategically sacrificing -information about higher-order interactions. Fractional factorial designs -are a type of confounded design that uses a fraction of the full factorial -experiment, making them highly efficient for screening important factors. - -This library simplifies the process of setting up and analyzing such -designs, making it invaluable for: - -- Experiment design in industrial and scientific research. -- Quality improvement and process optimization. -- Situations where a full factorial experiment is impractical due to - cost or time constraints. - -By offering these simple yet powerful tools, R-cran-conf.design enables -researchers and practitioners to conduct more efficient and insightful -experiments. +This small library contains a series of simple tools for constructing and +manipulating confounded and fractional factorial designs. diff --git a/math/R-cran-cvar/pkg-descr b/math/R-cran-cvar/pkg-descr index efad457a0534..ca1bec8d7a68 100644 --- a/math/R-cran-cvar/pkg-descr +++ b/math/R-cran-cvar/pkg-descr @@ -1,23 +1,7 @@ -The R-cran-cvar package provides essential tools for risk management, -enabling the computation of Expected Shortfall (ES) and Value at Risk (VaR). -ES, also known as Conditional Value at Risk (CVaR), and VaR are key metrics -used to quantify potential financial losses in portfolios or investments. - -This package offers high flexibility, allowing users to compute these -risk measures from various input types, including: - -- Quantile functions -- Distribution functions -- Random number generators -- Probability density functions - -It supports virtually any continuous distribution, making it adaptable -to diverse financial models. The functions are vectorized for efficient -computation across multiple arguments. The calculations are performed -directly from their definitions, as detailed by Acerbi and Tasche (2002). -Additionally, the package includes some support for GARCH (Generalized -Autoregressive Conditional Heteroskedasticity) models, further enhancing -its utility for analyzing financial time series volatility. - -R-cran-cvar is an invaluable resource for financial analysts, risk managers, -and quantitative researchers working with R to assess and manage financial risk. +Compute expected shortfall (ES) and Value at Risk (VaR) from a quantile +function, distribution function, random number generator or probability density +function. ES is also known as Conditional Value at Risk (CVaR). Virtually any +continuous distribution can be specified. The functions are vectorized over the +arguments. The computations are done directly from the definitions, see e.g. +Acerbi and Tasche (2002) . Some support for GARCH +models is provided, as well. diff --git a/math/R-cran-fracdiff/pkg-descr b/math/R-cran-fracdiff/pkg-descr index 229027fb8b72..2058c9e95b8e 100644 --- a/math/R-cran-fracdiff/pkg-descr +++ b/math/R-cran-fracdiff/pkg-descr @@ -1,20 +1,3 @@ -The R-cran-fracdiff package provides robust functionality for the -maximum likelihood estimation of parameters in fractionally differenced -ARIMA(p,d,q) models. These models are a powerful extension of traditional -ARIMA models, designed to capture long-range dependence in time series data, -where the 'd' parameter (differencing order) can be a non-integer value. - -Fractionally differenced ARIMA models are particularly useful for -analyzing phenomena that exhibit persistent memory effects, such as: - -- Financial time series (e.g., stock prices, volatility) -- Hydrological data (e.g., river flows, rainfall) -- Environmental data (e.g., temperature anomalies) -- Long-memory processes in various scientific and engineering fields - -Based on the methodology by Haslett and Raftery (Applied Statistics, 1989), -this package offers a reliable and statistically sound approach to -modeling time series with fractional integration. It enables researchers -and practitioners in R to accurately estimate the parameters of these -complex models, leading to more precise forecasts and a deeper understanding -of long-memory processes. +Maximum likelihood estimation of the parameters of a fractionally +differenced ARIMA(p,d,q) model (Haslett and Raftery, Appl.Statistics, +1989). diff --git a/math/R-cran-gbutils/pkg-descr b/math/R-cran-gbutils/pkg-descr index 6d016a6ddf25..8182290d8372 100644 --- a/math/R-cran-gbutils/pkg-descr +++ b/math/R-cran-gbutils/pkg-descr @@ -1,17 +1,8 @@ -The R-cran-gbutils package offers general-purpose utilities for numerical -and statistical computations in R, enhancing flexibility and ease of use. - -Key functionalities include: - -- **Distribution Analysis**: Plotting density/distribution functions, - numerically inverting distributions for quantiles, and simulating - real/complex numbers from magnitude/argument distributions. -- **Polynomial Manipulation**: Creating polynomials from roots - (Cartesian or polar form). -- **Programming Utilities**: Checking for NA identity, counting - positional arguments, computing set intersections for multiple sets, - identifying unnamed arguments, and graphing S4 classes. - -This invaluable toolkit streamlines common tasks in data analysis, -statistical modeling, and numerical programming, boosting productivity -and analytical capabilities for R users. +Plot density and distribution functions with automatic selection of suitable +regions. Numerically invert (compute quantiles) distribution functions. +Simulate real and complex numbers from distributions of their magnitude and +arguments. Optionally, the magnitudes and/or arguments may be fixed in almost +arbitrary ways. Create polynomials from roots given in Cartesian or polar form. +Small programming utilities: check if an object is identical to NA, count +positional arguments in a call, set intersection of more than two sets, check +if an argument is unnamed, compute the graph of S4 classes in packages. diff --git a/math/R-cran-magic/pkg-descr b/math/R-cran-magic/pkg-descr index 973187f4cb25..500c3c8314bf 100644 --- a/math/R-cran-magic/pkg-descr +++ b/math/R-cran-magic/pkg-descr @@ -1,19 +1,11 @@ -The R-cran-magic package provides efficient, vectorized algorithms for -creating and investigating magic squares and hypercubes. It includes -functions for manipulating and analyzing multi-dimensional arrays. - -Key features: - -- **Magic Square Creation**: Methods for generating normal magic - squares of any order greater than 2. -- **Analysis Tools**: Functions for the manipulation and analysis of - arbitrarily dimensioned arrays, including numerical verification - of magic square properties (e.g., determinant of odd-ordered - semimagic squares). -- **Antimagic Functionality**: Support for antimagic squares and - related concepts. - -The package aims to be a comprehensive computerized embodiment of magic -square knowledge, offering direct numerical verification of their -properties. It is a valuable resource for mathematicians, statisticians, -and R users interested in combinatorial designs and recreational mathematics. +A collection of efficient, vectorized algorithms for the creation +and investigation of magic squares and hypercubes, including a +variety of functions for the manipulation and analysis of arbitrarily +dimensioned arrays. The package includes methods for creating normal +magic squares of any order greater than 2. The ultimate intention +is for the package to be a computerized embodiment all magic square +knowledge, including direct numerical verification of properties +of magic squares (such as recent results on the determinant of +odd-ordered semimagic squares). Some antimagic functionality is +included. The package also serves as a rebuttal to the often-heard +comment "I thought R was just for statistics". diff --git a/math/R-cran-nortest/pkg-descr b/math/R-cran-nortest/pkg-descr index 9931aa8c41fa..c6ea3f75138a 100644 --- a/math/R-cran-nortest/pkg-descr +++ b/math/R-cran-nortest/pkg-descr @@ -1,17 +1 @@ -The R-cran-nortest package provides a suite of five omnibus tests -for assessing the composite hypothesis of normality in statistical data. -Normality tests are crucial in statistics to determine if a data set -is well-modeled by a normal distribution, which is a common assumption -for many parametric statistical methods. - -This package includes implementations of the following widely used tests: -- Anderson-Darling test -- Cramer-von Mises test -- Shapiro-Francia test -- Lilliefors test (Kolmogorov-Smirnov test with estimated parameters) -- Pearson chi-square test - -These tests are valuable tools for statisticians, researchers, and data -analysts working with R, enabling them to rigorously evaluate the -distributional assumptions of their data before applying further -statistical procedures. +Five omnibus tests for testing the composite hypothesis of normality. diff --git a/math/R-cran-quadprog/pkg-descr b/math/R-cran-quadprog/pkg-descr index c4a3622d3ad8..95c41d16dc00 100644 --- a/math/R-cran-quadprog/pkg-descr +++ b/math/R-cran-quadprog/pkg-descr @@ -1,18 +1,3 @@ -The R-cran-quadprog package provides an efficient and reliable implementation -of the dual method by Goldfarb and Idnani (1982, 1983) for solving -quadratic programming problems. - -Quadratic programming is a type of mathematical optimization problem that -involves minimizing a quadratic objective function subject to linear -constraints. This package is particularly useful for tasks such as -portfolio optimization, support vector machines, and other statistical -modeling applications where such optimization is required. - -Specifically, it solves problems of the form: -minimize -d'b + 1/2 b'Db -subject to A'b >= b0 - -where 'b' is the vector of variables to be optimized, 'd' is a vector, -'D' is a symmetric positive-definite matrix, 'A' is a matrix, and 'b0' -is a vector. The routine ensures accurate and robust solutions for -these types of constrained optimization problems within the R environment. +This routine implements the dual method of Goldfarb and Idnani +(1982, 1983) for solving quadratic programming problems of the form +min(?dT b + 1/2bT Db) with the constraints AT b >= b0. diff --git a/math/R-cran-qualityTools/pkg-descr b/math/R-cran-qualityTools/pkg-descr index 22f262d64ea9..a5da45e35c2e 100644 --- a/math/R-cran-qualityTools/pkg-descr +++ b/math/R-cran-qualityTools/pkg-descr @@ -1,21 +1,10 @@ -The R-cran-qualityTools package provides a comprehensive suite of -statistical methods essential for Quality Science and Six Sigma -Quality Management, particularly supporting the Define, Measure, -Analyze, Improve, and Control (DMAIC) cycle. +qualityTools: Statistical Methods for Quality Science -Key functionalities include: - -- **Distribution Fitting**: Tools for fitting various statistical - distributions to data. -- **Process Capability Analysis**: Calculation of normal and non-normal - process capability indices. -- **Measurement Systems Analysis (MSA)**: Techniques such as gauge - capability indices and Gauge Repeatability and Reproducibility (GR&R) - studies. -- **Experimental Design**: Support for factorial and fractional - factorial designs. -- **Response Surface Methods**: Including the use of desirability functions. - -This package is an invaluable resource for quality engineers, statisticians, -and practitioners implementing Six Sigma methodologies, enabling robust -analysis and improvement of processes. +Contains methods associated with the Define, Measure, Analyze, Improve and +Control (i.e. DMAIC) cycle of the Six Sigma Quality Management +methodology.It covers distribution fitting, normal and non-normal process +capability indices, techniques for Measurement Systems Analysis especially +gage capability indices and Gage Repeatability (i.e Gage RR) and +Reproducibility studies, factorial and fractional factorial designs as +well as response surface methods including the use of desirability +functions. diff --git a/math/algae/pkg-descr b/math/algae/pkg-descr index 153feada2264..ddbd3871533e 100644 --- a/math/algae/pkg-descr +++ b/math/algae/pkg-descr @@ -1,19 +1,4 @@ -Algae is a specialized programming language meticulously designed for -numerical analysis, particularly adept at tackling complex and large-scale -computational problems. Developed by the Boeing Company, Algae was -created to meet the demanding requirements of a fast, versatile, and -robust tool for advanced engineering and scientific applications. - -Its core strengths lie in efficiently handling numerical computations -involving large systems, making it suitable for: - -- Solving differential equations -- Performing matrix operations -- Implementing optimization algorithms -- Simulating complex physical phenomena - -With a proven track record of over a decade in aerospace and related -fields, Algae continues to be a valuable asset for researchers and -engineers who require a powerful and reliable language for high-performance -numerical analysis. Its design emphasizes both speed and the ability -to manage extensive datasets and intricate models. +Algae is a programming language for numerical analysis. It was written in +the Boeing Company to fulfill their need for a fast and versatile tool, +capable of handling large systems. Algae has been applied to interesting +problems in aerospace and related fields for more than a decade. diff --git a/math/apc/pkg-descr b/math/apc/pkg-descr index f5785f99e19e..7d46f00f722e 100644 --- a/math/apc/pkg-descr +++ b/math/apc/pkg-descr @@ -1,20 +1,17 @@ -APC (Auto Payment Calculator) is a simple, Xforms-based graphical -application designed for the X Window System. It provides a user-friendly -interface for calculating auto loan payments. + the Auto Payment Calculator V1.0 Release + Copyright (C) 1997 Eric A. Griff -Users can easily input the principal amount, loan term (in months), -and interest rate. Upon calculation, it displays the monthly payment, -as well as the number of weeks and the corresponding weekly payment. +Auto Payment Calculator is a simple, xforms based, application for +use under the X-windows system, that calculates auto loan payments. -Key features include: +It is pretty straight forward. You enter the Principal (Amount), +Term (in months), and Rate, and then with either [RETURN] +(or [enter] or whatever your keyboard equivelent is), (ALT-C), or +clicking the calculate button; you will have the payment in months, +as well as number of weeks, and weekly payment. -- **Intuitive Interface**: Built with Xforms for a straightforward - graphical user experience. -- **Loan Calculation**: Quickly determines monthly and weekly payments - based on user-provided loan details. -- **Interactive Input**: Supports keyboard navigation (e.g., Tab, Enter) - and mouse interaction for efficient data entry. - -APC is a practical utility for individuals needing to quickly estimate -car loan payments, offering a clear and concise solution within the -X Window environment. +You may also [TAB] through the Amount, Term, and Rate, as well as +hold down ALT and press the character in its Name that is underlined +to go do that function. As long as all three are filled in, you may +hit [ENTER] to Calculate right there. This makes it easy to cycle +quickly through numerous terms, amounts, and rates. diff --git a/math/aribas/pkg-descr b/math/aribas/pkg-descr index 31ea2becfaf4..165f8177d398 100644 --- a/math/aribas/pkg-descr +++ b/math/aribas/pkg-descr @@ -1,22 +1,6 @@ -ARIBAS is an interactive interpreter designed for advanced arithmetic, -offering robust support for both big integer and multi-precision -floating-point calculations. Its Pascal/Modula-like syntax provides -a familiar and structured environment for users to perform complex -mathematical operations. - -This powerful tool comes equipped with a rich set of built-in functions -specifically tailored for algorithmic number theory, including: - -- **Number Theoretic Functions**: Greatest Common Divisor (GCD), - Jacobi symbol, and continued fraction expansions. -- **Primality Testing**: Rabin probabilistic prime test for efficient - identification of prime numbers. -- **Integer Factorization Algorithms**: - - Quadratic sieve factorization for general integers. - - Pollard's rho factorization for finding smaller prime factors. - -ARIBAS is an invaluable resource for mathematicians, computer scientists, -and cryptographers who require precise and efficient tools for number -theoretic research, cryptographic analysis, and other applications -involving large numbers and complex arithmetic. Its interactive nature -makes it ideal for experimentation and exploration of numerical properties. +ARIBAS is an interactive interpreter for big integer arithmetic and +multi-precision floating point arithmetic with a Pascal/Modula like +syntax. It has several builtin functions for algorithmic number +theory like gcd, Jacobi symbol, Rabin probabilistic prime test, +continued fraction and quadratic sieve factorization, Pollard rho +factorization, etc. diff --git a/math/arpack++/pkg-descr b/math/arpack++/pkg-descr index 2e84fa9b3d96..be857c4234e9 100644 --- a/math/arpack++/pkg-descr +++ b/math/arpack++/pkg-descr @@ -1,18 +1,4 @@ -ARPACK++ provides an object-oriented C++ interface to ARPACK (ARnoldi -PACKage), a widely used Fortran library for solving large-scale -eigenvalue problems. This wrapper allows C++ developers to leverage -ARPACK's power within a modern programming paradigm. - -ARPACK is known for efficiently computing a few eigenvalues and -eigenvectors of large, sparse matrices, making it vital in quantum -mechanics, structural engineering, and data analysis. ARPACK++ retains -the original Fortran package's strengths: - -- **Full Capability**: Access to all ARPACK functionalities for - various eigenvalue problems. -- **High Performance**: Maintains computational speed and efficiency. -- **Exceptional Accuracy**: Delivers precise numerical results. -- **Low Memory Requirements**: Optimized for large matrices. - -By integrating ARPACK's robust numerical algorithms with C++ flexibility, -ARPACK++ offers a powerful solution for complex eigenvalue computations. +ARPACK++ is a collection of classes that offers c++ programmers an interface +to ARPACK. It preserves the full capability, performance, accuracy and low +memory requirements of the FORTRAN package, but takes advantage of the C++ +object-oriented programming environment. diff --git a/math/atlas/pkg-descr b/math/atlas/pkg-descr index c07eeac58026..6e7e6aed22a0 100644 --- a/math/atlas/pkg-descr +++ b/math/atlas/pkg-descr @@ -1,21 +1,18 @@ -ATLAS (Automatically Tuned Linear Algebra Software) is a high-performance -software library for numerical linear algebra. It focuses on applying -empirical optimization techniques to deliver portable and efficient -performance across diverse hardware architectures. +The ATLAS (Automatically Tuned Linear Algebra Software) project is an ongoing +research effort focusing on applying empirical techniques in order to provide +portable performance. At present, it provides C and Fortran77 interfaces to +a portable, efficient BLAS implementation, as well as enhanced versions of a +few routines from LAPACK. To link with ATLAS shared libraries: -ATLAS provides optimized implementations of: - -- **BLAS (Basic Linear Algebra Subprograms)**: Offers C and Fortran77 - interfaces for Level 1, 2, and 3 BLAS routines, crucial for vector, - matrix-vector, and matrix-matrix operations. Both serial (thread-safe) - and multi-threaded versions are available. -- **LAPACK (Linear Algebra Package)**: Includes enhanced versions of - key LAPACK routines, providing efficient solutions for problems - like solving systems of linear equations, eigenvalue problems, and - singular value decomposition. - -The project's core strength lies in its ability to automatically tune -itself to the specific characteristics of the underlying hardware during -installation, ensuring optimal performance. ATLAS is an invaluable -resource for scientific computing, engineering simulations, and any -application requiring fast and reliable linear algebra computations. +Serial (thread-safe) Fortran77 BLAS: + -lf77blas +Multi-threaded Fortran77 BLAS: + -lptf77blas +Serial (thread-safe) C BLAS: + -lcblas +Multi-threaded C BLAS: + -lptcblas +ATLAS-enhanced LAPACK, serial (thread-safe) interface: + -lalapack -lf77blas -lcblas +ATLAS-enhanced LAPACK, multi-threaded interface: + -lalapack -lptf77blas -lptcblas diff --git a/math/blacs/pkg-descr b/math/blacs/pkg-descr index ee27799d8874..23b4f5cf0aa5 100644 --- a/math/blacs/pkg-descr +++ b/math/blacs/pkg-descr @@ -1,23 +1,5 @@ -The BLACS (Basic Linear Algebra Communication Subprograms) library is a -fundamental component for high-performance parallel computing, specifically -designed to facilitate linear algebra operations on distributed memory -platforms. It provides a standardized and efficient message passing -interface tailored for numerical linear algebra algorithms. - -BLACS enables the communication and synchronization of data between -processors in a parallel computing environment, which is crucial for -implementing scalable versions of dense linear algebra routines. This -makes it an essential building block for: - -- **Distributed Linear Algebra Libraries**: Such as ScaLAPACK, which - relies on BLACS for inter-processor communication. -- **Scientific Simulations**: Large-scale computations in physics, - engineering, and other fields that require solving complex linear - systems or eigenvalue problems across multiple nodes. -- **High-Performance Computing (HPC)**: Optimizing numerical workloads - on clusters and supercomputers. - -By offering a uniform and efficient communication layer, BLACS allows -developers to write portable and high-performing parallel linear algebra -code, ensuring that numerical applications can effectively utilize the -power of distributed memory architectures. +The BLACS (Basic Linear Algebra Communication Subprograms) +project is an ongoing investigation whose purpose is to create +a linear algebra oriented message passing interface +that may be implemented efficiently and uniformly across +a large range of distributed memory platforms. diff --git a/math/blocksolve95/pkg-descr b/math/blocksolve95/pkg-descr index 6b4bed580e5d..522f5f3c5f41 100644 --- a/math/blocksolve95/pkg-descr +++ b/math/blocksolve95/pkg-descr @@ -1,15 +1,13 @@ -BlockSolve95 is a scalable parallel software library designed for the -efficient solution of large, sparse linear systems. It is particularly -optimized for problems arising from physical models, especially those -with multiple degrees of freedom at each node (e.g., finite element -methods in structural engineering). +BlockSolve95 is a scalable parallel software library primarily intended for the +solution of sparse linear systems that arise from physical models, especially +problems involving multiple degrees of freedom at each node. For example, when +the finite element method is used to solve practical problems in structural +engineering, each node typically has two to five degrees of freedom; +BlockSolve95 is designed to take advantage of problems with this type of local +structure. BlockSolve95 is also reasonably efficient for problems that have +only one degree of freedom associated with each node, such as the three- +dimensional Poisson problem. -The library effectively handles problems with this local structure, -while also remaining reasonably efficient for systems with a single -degree of freedom per node (e.g., three-dimensional Poisson problems). - -BlockSolve95 is a general-purpose solver, requiring only that matrices -are sparse and symmetric in structure (though not necessarily in value). -It provides a robust solution for complex scientific and engineering -simulations that demand high-performance parallel computation for -large sparse linear systems. +BlockSolve95 is general purpose; we do not require that the matrices have any +particular structure other than being sparse and being symmetric in structure +(but not necessarily in value). diff --git a/math/brial/pkg-descr b/math/brial/pkg-descr index 4b7d2deb9b3a..bc7e9d80a98b 100644 --- a/math/brial/pkg-descr +++ b/math/brial/pkg-descr @@ -1,22 +1,11 @@ -BRiAl (Boolean Rings and Algebra) is a powerful C++ library for -computations with polynomials over Boolean rings, serving as the -successor to PolyBoRi. It provides high-level data types and efficient -algorithms for symbolic computation in this specialized algebraic domain. +BRiAl is the successor to PolyBoRi. -Key features include: - -- **High-level Data Types**: For Boolean polynomials, monomials, - exponent vectors, and related algebraic structures. -- **Binary Decision Diagrams (BDDs)**: Utilizes BDDs as the internal - storage type for polynomial structures, enabling efficient - representation and manipulation. -- **Python Interface**: Offers a convenient Python binding, allowing - for parsing complex polynomial systems and implementing sophisticated - strategies for Grobner basis computation. -- **Grobner Basis Computation**: Provides a robust and powerful - reference implementation for Grobner basis algorithms, essential - for solving systems of polynomial equations. - -BRiAl is an invaluable tool for researchers and developers in areas -such as cryptography, coding theory, formal verification, and computer -algebra, where efficient manipulation of Boolean polynomials is critical. +The core of PolyBoRi is a C++ library, which provides high-level data +types for Boolean polynomials and monomials, exponent vectors, as well +as for the underlying polynomial rings and subsets of the powerset of +the Boolean variables. As a unique approach, binary decision diagrams +are used as internal storage type for polynomial structures. On top of +this C++-library we provide a Python interface. This allows parsing of +complex polynomial systems, as well as sophisticated and extendable +strategies for Groebner base computation. PolyBoRi features a powerful +reference implementation for Groebner basis computation. diff --git a/math/clblas/pkg-descr b/math/clblas/pkg-descr index 6b9073ad0249..a63d390014d0 100644 --- a/math/clblas/pkg-descr +++ b/math/clblas/pkg-descr @@ -1,21 +1,11 @@ -clBLAS is a high-performance software library that provides optimized -BLAS (Basic Linear Algebra Subprograms) functions implemented in OpenCL. -BLAS routines are fundamental building blocks for numerical linear algebra, -widely used in scientific computing, engineering, and data analysis. +clBLAS -The primary goal of clBLAS is to empower developers to leverage the -performance and power efficiency of heterogeneous computing environments. -It achieves this by: +a software library containing BLAS functions written in OpenCL -- **OpenCL Integration**: Directly utilizes OpenCL interfaces, allowing - users full control over OpenCL state management for maximum - performance and flexibility. -- **Optimized Kernel Generation**: Automatically generates and enqueues - optimized OpenCL kernels, freeing users from the complex task of - writing, optimizing, and maintaining kernel code. - -clBLAS is an invaluable resource for developers and researchers who need -to accelerate their linear algebra workloads by harnessing the parallel -processing capabilities of GPUs and other OpenCL-compatible devices. -It streamlines the development of high-performance computing applications -by providing a robust and efficient foundation for numerical operations. +The primary goal of clBLAS is to make it easier for developers to utilize the +inherent performance and power efficiency benefits of heterogeneous computing. +clBLAS interfaces do not hide nor wrap OpenCL interfaces, but rather leaves +OpenCL state management to the control of the user to allow for maximum +performance and flexibility. The clBLAS library does generate and enqueue +optimized OpenCL kernels, relieving the user from the task of writing, +optimizing and maintaining kernel code themselves. diff --git a/math/clblast/pkg-descr b/math/clblast/pkg-descr index c3c31015723b..cf3cfb06b914 100644 --- a/math/clblast/pkg-descr +++ b/math/clblast/pkg-descr @@ -1,22 +1,2 @@ -CLBlast is a cutting-edge, lightweight, and highly performant OpenCL -BLAS (Basic Linear Algebra Subprograms) library. It provides efficient -and accelerated linear algebra computations on OpenCL-compatible devices. - -BLAS routines are fundamental building blocks for numerical algorithms -in scientific computing, machine learning, and data analysis. CLBlast -leverages OpenCL to offload these tasks to GPUs and other accelerators, -significantly speeding up applications. - -Key features and benefits: - -- **Modern Design**: Built with contemporary OpenCL practices for - optimal performance. -- **Lightweight Footprint**: Minimizes overhead for diverse systems. -- **High Performance**: Achieves superior execution speeds through - careful optimization. -- **Tunable**: Allows fine-grained control to extract maximum - performance from specific hardware (Intel, AMD, NVIDIA accelerators). - -CLBlast is an invaluable resource for developers and researchers seeking -to accelerate numerical workloads by harnessing parallel processing -capabilities of modern hardware through OpenCL. +Modern, lightweight, performant and tunable OpenCL BLAS library. Tuned for +Intel, AMD, and NVIDIA accelerators. diff --git a/math/clfft/pkg-descr b/math/clfft/pkg-descr index 4c745e85430d..321354154a5f 100644 --- a/math/clfft/pkg-descr +++ b/math/clfft/pkg-descr @@ -1,24 +1,7 @@ -clFFT is a high-performance software library providing optimized Fast -Fourier Transform (FFT) functions implemented in OpenCL. The FFT is a -fundamental algorithm in digital signal processing and numerical analysis, -used for tasks such as spectral analysis, image processing, and solving -partial differential equations. +clFFT -Leveraging the OpenCL framework, clFFT enables efficient computation -of FFTs on a wide range of parallel processing devices. Its key features -include: +a software library containing FFT functions written in OpenCL -- **GPU Acceleration**: Primarily designed to harness the power of - Graphics Processing Units (GPUs) for significant speedups in FFT - computations. -- **CPU Support**: Also supports execution on Central Processing Units - (CPUs), which is beneficial for debugging, development, and - heterogeneous computing environments where a mix of device types - is utilized. -- **OpenCL Standard**: Adheres to the OpenCL standard, ensuring - portability across different hardware vendors and platforms. - -clFFT is an invaluable resource for developers and researchers who need -to perform fast and efficient Fourier transforms on large datasets, -particularly in applications that can benefit from the parallel -processing capabilities of modern GPUs and multi-core CPUs. +clFFT is a software library containing FFT functions written in OpenCL. In +addition to GPU devices, the libraries also support running on CPU devices to +facilitate debugging and heterogeneous programming. diff --git a/math/cliquer/pkg-descr b/math/cliquer/pkg-descr index e23a63616285..421b623dcc63 100644 --- a/math/cliquer/pkg-descr +++ b/math/cliquer/pkg-descr @@ -1,23 +1,9 @@ -Cliquer is a highly efficient C library designed for finding cliques -in arbitrary weighted graphs. In graph theory, a clique is a subset -of vertices where every pair of vertices is connected by an edge. -Finding cliques is a fundamental problem with applications in social -network analysis, bioinformatics, and computer vision. +Cliquer is a set of C routines for finding cliques in an arbitrary weighted +graph. It uses an exact branch-and-bound algorithm developed by Patric +Ostergard. It is designed with the aim of being efficient while still being +flexible and easy to use. -This library implements an exact branch-and-bound algorithm developed -by Patric Ostergard, ensuring optimal solutions. Cliquer is meticulously -designed to be: +Note: this port do not use the upstream version, but the version autotoolized +by Dima Pasechnik. -- **Efficient**: Optimized for performance, even on complex graphs. -- **Flexible**: Adaptable to various graph structures and problem - specifications. -- **Easy to Use**: Provides a straightforward API for integration - into other applications. - -Note that this port utilizes a version of Cliquer that has been -autotoolized by Dima Pasechnik, enhancing its build system and -portability. This ensures a robust and well-maintained package. - -Cliquer is an invaluable resource for researchers and developers working -with graph algorithms, offering a powerful and reliable tool for -identifying dense subgraphs and solving related combinatorial problems. +See also: https://github.com/dimpase/autocliquer diff --git a/math/clrng/pkg-descr b/math/clrng/pkg-descr index 4d94cc188962..93c0bf766561 100644 --- a/math/clrng/pkg-descr +++ b/math/clrng/pkg-descr @@ -1,24 +1,11 @@ -clRNG is a specialized library designed for high-quality uniform random -number generation within OpenCL environments. It provides a robust and -efficient solution for parallel applications requiring statistically -sound random numbers on GPUs and other OpenCL-compatible devices. +clRNG -The library introduces the concept of "streams of random numbers," which -act as virtual random number generators. These streams can be created -in unlimited quantities on the host system and then utilized by work -items on computing devices to generate random numbers. Each stream also -features equally-spaced substreams, offering additional flexibility for -complex simulations. +a library for uniform random number generation in OpenCL. -clRNG currently implements a selection of well-regarded pseudorandom -number generators, including: - -- MRG31k3p -- MRG32k3a -- LFSR113 -- Philox-4x32-10 - -This library is an invaluable resource for researchers and developers -in fields such as Monte Carlo simulations, scientific computing, and -machine learning, where efficient and reliable parallel random number -generation is crucial. +Streams of random numbers act as virtual random number generators. +They can be created on the host computer in unlimited numbers, and +then used either on the host or on computing devices by work items +to generate random numbers. Each stream also has equally-spaced +substreams, which are occasionally useful. The API is currently +implemented for four different RNGs, namely the MRG31k3p, MRG32k3a, +LFSR113 and Philox-4x32-10 generators. diff --git a/math/cocoalib/pkg-descr b/math/cocoalib/pkg-descr index 7d3c9bd0133c..6b610cd891f4 100644 --- a/math/cocoalib/pkg-descr +++ b/math/cocoalib/pkg-descr @@ -1,20 +1,6 @@ -CoCoALib is a powerful C++ library dedicated to Computations in -Commutative Algebra. This field of mathematics is fundamental to -algebraic geometry, number theory, and computer algebra systems, -focusing on algebraic structures like rings and ideals. +CoCoALib is a C++ library for Computations in Commutative Algebra, +focused mainly on polynomial rings, ideals, Groebner basis and +similar topics. -The library provides a robust set of tools for working with: - -- **Polynomial Rings**: Operations on multivariate polynomials. -- **Ideals**: Computations with ideals in polynomial rings. -- **Grobner Bases**: A cornerstone algorithm for solving systems of - polynomial equations and performing other algebraic manipulations. -- **Related Topics**: Other advanced concepts in commutative algebra. - -For users who prefer an interactive environment without direct C++ -programming, the companion CoCoA-5 shell (available separately) offers -a user-friendly interface to most of CoCoALib's functionalities. - -CoCoALib is an invaluable resource for mathematicians, computer scientists, -and researchers engaged in algebraic computations, providing a high-performance -and flexible framework for exploring complex algebraic structures. +You might like to install CoCoA-5 too, a shell that lets you interact +with most of CoCoALib without the need to learn C++. diff --git a/math/concorde/pkg-descr b/math/concorde/pkg-descr index 84039a11969c..2fe47df09e48 100644 --- a/math/concorde/pkg-descr +++ b/math/concorde/pkg-descr @@ -1,22 +1,14 @@ -Concorde is a highly optimized computer code designed for solving the -Traveling Salesman Problem (TSP) and various related network optimization -problems. Implemented in ANSI C, it is renowned for its ability to find -optimal solutions to extremely large and complex instances of the TSP. +Concorde is a computer code for the traveling salesman problem (TSP) +and some related network optimization problems. The code is written +in the ANSI C programming language and it is available for academic +research use; for other uses, contact bico@isye.gatech.edu for +licensing options. -Key features and capabilities include: +Concorde's TSP solver has been used to obtain the optimal solutions to +106 of the 110 TSPLIB instances; the largest having 15,112 cities. -- **Optimal TSP Solutions**: Concorde's TSP solver has successfully - found optimal solutions for 106 of the 110 TSPLIB instances, - including problems with up to 15,112 cities. -- **Extensive Callable Library**: Provides over 700 functions, allowing - users to develop specialized codes for TSP-like problems and integrate - Concorde's powerful algorithms into their own applications. -- **Parallel Computing Support**: All functions are thread-safe for - shared-memory parallel environments. The main TSP solver also - supports execution across networks of Unix workstations, enabling - distributed computation for even larger problems. - -Concorde is an invaluable resource for researchers and practitioners -in combinatorial optimization, operations research, and computer science, -offering a robust and efficient solution for one of the most famous -problems in theoretical computer science. +The Concorde callable library includes over 700 functions permitting +users to create specialized codes for TSP-like problems. All Concorde +functions are thread-safe for programming in shared-memory parallel +environments; the main TSP solver includes code for running over +networks of Unix workstations. diff --git a/math/crlibm/pkg-descr b/math/crlibm/pkg-descr index 04d020e0cfb8..7a3c0fe62f25 100644 --- a/math/crlibm/pkg-descr +++ b/math/crlibm/pkg-descr @@ -1,24 +1,21 @@ -CRlibm is an efficient and rigorously proven mathematical library -providing correctly rounded implementations of double-precision C99 -standard elementary functions. It supports all four IEEE-754 rounding -modes, offering high accuracy and reliability for numerical computations. +CRlibm is an efficient and proven mathematical library, which +provides implementations of the double-precision C99 standard +elementary functions, correctly rounded in the four IEEE-754 rounding +modes, and sufficiently efficient in average time, worst-case time, +and memory consumption to replace existing libms transparently. -Designed for transparent replacement of existing `libm` implementations, -CRlibm maintains efficiency in average and worst-case time, along with -optimized memory consumption. Its development includes extensive -documentation with formal proofs for each function, making it an -excellent resource for understanding software elementary function +The distribution includes extensive documentation with the proof +of each function (currently more than 100 pages), as well as all +the Maple scripts used to develop the functions. This makes this +library an excellent tutorial on software elementary function development. -CRlibm also integrates scslib (Software Carry Save Library), a lightweight -multiple-precision library. scslib is tailored for CRlibm's needs, -offering precision up to a few hundred bits, portability, IEEE -floating-point compatibility, and performance comparable to or better -than GMP, all within a small footprint. It efficiently handles -multiple-precision additions, subtractions, multiplications, and conversions -by avoiding carry propagations during multiplication. - -CRlibm is an invaluable tool for applications demanding high-precision, -correctly rounded mathematical functions, particularly in scientific -computing, financial modeling, and other fields where numerical accuracy -is paramount. +CRlibm also includes a lightweight library for multiple precision, +scslib (Software Carry Save Library). This library has been developed +specifically to answer the needs of the CRlibm project: precision +up to a few hundred bits, portability, compatibility with IEEE +floating-point standards, performance comparable to or better than +GMP, and a small footprint. It uses a data-structure which allows +carry propagations to be avoided during multiple-precision +multiplications, and supports addition, subtraction, multiplication, +and conversions. diff --git a/math/dieharder/pkg-descr b/math/dieharder/pkg-descr index 4efd18b2e129..327aac1c5bda 100644 --- a/math/dieharder/pkg-descr +++ b/math/dieharder/pkg-descr @@ -1,21 +1,22 @@ -Dieharder is a comprehensive, GPL-licensed test suite for evaluating -the quality of random number generators (RNGs). It builds upon the -legacy of George Marsaglia's "Diehard battery of tests" and expands -upon it with modern statistical methodologies. +At the suggestion of Linas Vepstas on the Gnu Scientific Library (GSL) list, +this GPL'd suite of random number tests will be named "Dieharder". Using a +movie sequel pun for the name is a double tribute to George Marsaglia, whose +"Diehard battery of tests" of random number generators has enjoyed years of +enduring usefulness as a test suite. -This suite incorporates a diverse collection of tests, including: +The dieharder suite is more than just the diehard tests cleaned up and given a +pretty GPL'd source face in native C: tests from the Statistical Test Suite +(STS) developed by the National Institute for Standards and Technology (NIST) +are being incorporated, as are new tests developed by rgb. Where possible, +tests are parametrized and controllable so that failure, at least, is +unambiguous. -- **Diehard Tests**: Classic tests for assessing RNG randomness. -- **NIST Statistical Test Suite (STS)**: Tests developed by the - National Institute for Standards and Technology. -- **New Tests**: Additional tests developed by the project's author. +A further design goal is to provide some indication of *why* a generator fails +a test, where such information can be extracted during the test process and +placed in usable form. For example, the bit-distribution tests should +(eventually) be able to display the actual histogram for the different bit +n-tuplets. -Dieharder is designed with extensibility in mind, allowing for the -incorporation of new tests and analysis methods. A key design goal is -to provide not just pass/fail results, but also insights into *why* an -RNG might fail a particular test, offering diagnostic information -(e.g., displaying histograms for bit distributions). - -This makes Dieharder an invaluable tool for researchers, cryptographers, -and developers who require rigorous validation of RNGs for applications -in simulations, security, and statistical analysis. +Dieharder is by design extensible. It is intended to be the "Swiss army knife +of random number test suites", or if you prefer, "the last suite you'll ever +ware" for testing random numbers. diff --git a/math/edenmath/pkg-descr b/math/edenmath/pkg-descr index 2a4a94c42681..addee44d96db 100644 --- a/math/edenmath/pkg-descr +++ b/math/edenmath/pkg-descr @@ -1,20 +1,4 @@ -EdenMath is a user-friendly scientific calculator designed to perform -a wide array of mathematical computations. It offers a comprehensive -set of functionalities, making it a versatile tool for students, -educators, and professionals alike. +EdenMath is a scientific calculator. It does standard arithmetic, +probability, and trigonometric functions. -Key features of EdenMath include: - -- **Standard Arithmetic**: Basic operations such as addition, subtraction, - multiplication, and division. -- **Advanced Mathematical Functions**: Support for powers, roots, logarithms, - and exponential functions. -- **Trigonometric Functions**: Calculations involving sine, cosine, tangent, - and their inverses. -- **Probability and Statistics**: Functions for permutations, combinations, - and basic statistical calculations. - -With its intuitive interface, EdenMath simplifies complex calculations, -providing accurate and quick results. It is distributed under the terms -of the GPL2 or later license, ensuring it remains free and open-source -for everyone to use and modify. +LICENSE: GPL2 or later diff --git a/math/eispack/pkg-descr b/math/eispack/pkg-descr index 82fbd65a543e..02e4215b6d3c 100644 --- a/math/eispack/pkg-descr +++ b/math/eispack/pkg-descr @@ -1,19 +1,15 @@ -EISPACK is a classic software library for solving eigenvalue problems. -It provides a comprehensive collection of Fortran subroutines for -computing eigenvalues and eigenvectors of various types of matrices. +This is eispack from research.att.com. I've cleaned up the Makefile, but +it is otherwise the same. The package is described in: -Originally developed by a team including B.T. Smith, J.M. Boyle, J.J. -Dongarra, B.S. Garbow, Y. Ikebe, V.C. Klema, and C.B. Moler, EISPACK -has been a foundational resource in numerical linear algebra. The -library is described in detail in: +1. Smith, B.T, Boyle, J.M, Dongerra, J.J., Garbow, B.S., Ikebe, Y., + Klema, V.C., and Moler, C.B., Matrix Eigensystem Routines -- EISPACK + Guide, Lecture Notes in Computer Science, Vol. 6, Second Edition, + Springer-Verlag, New York, Heidelberg, Berlin, 1976 -1. Smith et al., "Matrix Eigensystem Routines -- EISPACK Guide," - Lecture Notes in Computer Science, Vol. 6, Springer-Verlag, 1976. -2. Garbow et al., "Matrix Eigensystem Routines -- EISPACK Guide - Extension," Lecture Notes in Computer Science, Vol. 51, - Springer-Verlag, 1977. +2. Garbow, B.S., Boyle J.M., Dongerra, J.J, and Moler C.B., Matrix + Eigensystem Routines -- EISPACK Guide Extension, Lecture Notes in + Computer Science, Vol. 51, Springer-Verlag, New York, Heidelberg, + Berlin, 1977 -As a Fortran-based library, it primarily installs the compiled library -itself. EISPACK remains a valuable reference and tool for researchers -and developers working with eigenvalue computations, particularly in -scientific and engineering applications. +As the package is in FORTRAN there are no include files for the library, +and the only thing to install is the library itself. diff --git a/math/emc2/pkg-descr b/math/emc2/pkg-descr index 3652c87a1473..4d9ac4b08048 100644 --- a/math/emc2/pkg-descr +++ b/math/emc2/pkg-descr @@ -1,20 +1,8 @@ -Emc2 is a portable, interactive, and graphical editor for two-dimensional -mesh geometries. It provides a comprehensive environment for creating, -modifying, and analyzing mesh structures, making it an invaluable tool -for numerical simulations and engineering applications. +Emc2 is a portable, interactive, graphical editor of two-dimensional mesh +geometries. It can create and modify geometries (as in CAD), and define line +discretizations, subdomains, and reference numbers (to take into account +boundary conditions and material properties). Grid and Delaunay-Voronoi +meshes composed of triangles or quadrilaterals can be regularized, rotated, +and modified via the addition, removal, or moving of vertices. -Key functionalities include: - -- **Geometry Creation and Modification**: Functions similar to CAD - software for designing and altering 2D shapes. -- **Mesh Definition**: Define line discretizations, subdomains, and - assign reference numbers for boundary conditions and material properties. -- **Mesh Generation**: Create grid and Delaunay-Voronoi meshes composed - of triangles or quadrilaterals. -- **Mesh Manipulation**: Regularize, rotate, and modify meshes by - adding, removing, or moving vertices, ensuring optimal mesh quality. - -Emc2 streamlines the preprocessing stage for finite element analysis, -computational fluid dynamics, and other simulation methods. For enhanced -mesh generation capabilities, it is highly suggested that users also -install the `math/bamg` port. +It is suggested that users also install the math/bamg port. diff --git a/math/ent/pkg-descr b/math/ent/pkg-descr index c13730e54306..d457f4f297d5 100644 --- a/math/ent/pkg-descr +++ b/math/ent/pkg-descr @@ -1,20 +1,5 @@ -Ent is a powerful command-line utility for applying statistical tests -to byte sequences within files. It provides insights into the randomness -and information density of data, crucial for evaluating various digital -processes. - -Key analyses performed include: - -- **Entropy Calculation**: Measures information content per byte. -- **Chi-Square Test**: Assesses uniformity of byte distribution. -- **Arithmetic Mean and Monte Carlo Pi Estimation**: Statistical indicators. -- **Serial Correlation Coefficient**: Detects patterns between bytes. - -Ent is particularly useful for: - -- **Evaluating Pseudorandom Number Generators (PRNGs)**: Essential for - cryptography and statistical sampling. -- **Assessing Compression Algorithms**: Analyzing information density. -- **Analyzing Data Streams**: Identifying non-random patterns. - -It helps users ensure the integrity and quality of their data and algorithms. +ent applies various tests to sequences of bytes stored in files and reports the +results of those tests. The program is useful for evaluating pseudorandom +number generators for encryption and statistical sampling applications, +compression algorithms, and other applications where the information density of +a file is of interest. diff --git a/math/fftw/pkg-descr b/math/fftw/pkg-descr index 8f75c980a493..8595fe3ffd99 100644 --- a/math/fftw/pkg-descr +++ b/math/fftw/pkg-descr @@ -1,23 +1,14 @@ -FFTW (Fastest Fourier Transform in the West) is a highly optimized C -subroutine library for computing the Discrete Fourier Transform (DFT). -It supports one or more dimensions, both real and complex data, and -arbitrary input size, making it a versatile tool for various applications. - -FFTW is renowned for its exceptional performance, consistently outperforming -other publicly available FFT software across a wide range of platforms. -Its performance is also highly portable, ensuring optimal execution on -most architectures without requiring modifications. - -Key features and benefits include: - -- **Versatile DFT Computation**: Handles 1D, 2D, and multi-dimensional - DFTs for both real and complex datasets. -- **Arbitrary Input Size**: Efficiently processes data of any length, - without power-of-two restrictions. -- **Self-Optimizing**: Employs adaptive algorithms that tune themselves - to the underlying hardware for peak performance. - -Developed at MIT by Matteo Frigo and Steven G. Johnson, FFTW has become -the de facto standard for FFT computations in scientific computing, -signal processing, image analysis, and many other fields where fast -and accurate Fourier transforms are critical. +FFTW is a C subroutine library for computing the Discrete Fourier Transform +(DFT) in one or more dimensions, of both real and complex data, and of +arbitrary input size. We believe that FFTW, which is free software, should +become the FFT library of choice for most applications. Our benchmarks, +performed on a variety of platforms, show that FFTW's performance is +typically superior to that of other publicly available FFT software. +Moreover, FFTW's performance is portable: the program will perform well on +most architectures without modification. + +The FFTW package was developed at MIT by Matteo Frigo and Steven G. +Johnson. Please send email to fftw@theory.lcs.mit.edu so that we can keep +track of users and send you information about new releases. The latest +version of FFTW, benchmarks, links, and other information can be found at +the FFTW home page. diff --git a/math/frobby/pkg-descr b/math/frobby/pkg-descr index f202f5d90ba4..00d91886bcc3 100644 --- a/math/frobby/pkg-descr +++ b/math/frobby/pkg-descr @@ -1,21 +1,13 @@ -Frobby is a powerful software system and project dedicated to computations -with monomial ideals. As free software, it serves as a vital tool for -computational and mathematical research in this specialized area of algebra. +Frobby is a software system and project for computations with monomial +ideals. Frobby is free software and it is intended as a vehicle for +computational and mathematical research on monomial ideals. -Its extensive functionality includes: - -- **Ideal Properties**: Euler characteristic, Hilbert series, maximal - standard monomials, primary decomposition, irreducible decomposition, - Alexander dual, and associated primes. -- **Ideal Operations**: Minimization and intersection of monomial ideals. -- **Frobenius Problems**: Computation of Frobenius problems, even with - very large numbers, leveraging the capabilities of 4ti2. -- **Interoperability**: Supports translation between formats compatible - with various computer algebra systems, including Macaulay2, Monos, - 4ti2, CoCoA4, and Singular. This allows Frobby to seamlessly integrate - with and extend the capabilities of these systems. - -Frobby is an invaluable resource for researchers and students in -commutative algebra, algebraic geometry, and computational mathematics, -providing a flexible and robust platform for exploring the intricate -properties of monomial ideals. +The current functionality includes Euler characteristic, Hilbert +series, maximal standard monomials, combinatorial optimization on +monomial ideals, primary decomposition, irreducible decomposition, +Alexander dual, associated primes, minimization and intersection of +monomial ideals as well as the computation of Frobenius problems +(using 4ti2) with very large numbers. Frobby is also able to translate +between formats that can be used with several different computer +systems, such as Macaulay 2, Monos, 4ti2, CoCoA4 and Singular. Thus +Frobby can be used with any of those systems. diff --git a/math/gexpr/pkg-descr b/math/gexpr/pkg-descr index 20485cdc3642..373f9b9e11e5 100644 --- a/math/gexpr/pkg-descr +++ b/math/gexpr/pkg-descr @@ -1,12 +1,3 @@ -Gexpr is a powerful and lightweight command-line calculator designed -for shell environments. It supports floating-point arithmetic, a wide -range of standard C mathematical functions, and relational operators -for conditional expressions. Gexpr offers flexible output options, -allowing results to be displayed in binary, octal, decimal, or -hexadecimal bases. - -It serves as an excellent alternative to traditional command-line -calculators like bc(1), especially when a more streamlined and -feature-rich solution is desired. Furthermore, gexpr can be seamlessly -integrated into shell scripts, providing robust floating-point math -capabilities for automation and complex calculations. +Gexpr is a shell calculator with floating point, standard C functions, +relational operators, and output in base 2/8/10/16. It is a light alternative +to bc(1). It can also be used to add floating point math to shell scripts. diff --git a/math/glgraph/pkg-descr b/math/glgraph/pkg-descr index 4360cd67124f..9573516c24cf 100644 --- a/math/glgraph/pkg-descr +++ b/math/glgraph/pkg-descr @@ -1,15 +1,2 @@ -GLgraph is a powerful and interactive visualization tool designed to -render complex mathematical functions in a graphical environment. -Leveraging OpenGL, it provides dynamic 3D and 4D representations, -making abstract mathematical concepts tangible and explorable. - -This utility excels at visualizing functions with up to three independent -variables (typically denoted as x, z, and t). Notably, it can generate -stunning 4D function plots, where three dimensions represent space and -the fourth dimension represents time, allowing for the visualization of -evolving systems or dynamic surfaces. - -GLgraph is an invaluable resource for students, educators, and researchers -in mathematics, physics, engineering, and other scientific disciplines -who need to understand and analyze multi-dimensional functions through -intuitive and high-quality graphical representations. +GLgraph visualize mathematical functions. It can handle 3 unknowns (x,z,t) and +can produce a 4D function with 3 space and 1 time dimension. diff --git a/math/gmp-ecm/pkg-descr b/math/gmp-ecm/pkg-descr index db04a907fa7e..6420ee302738 100644 --- a/math/gmp-ecm/pkg-descr +++ b/math/gmp-ecm/pkg-descr @@ -1,20 +1,2 @@ -GMP-ECM is a specialized program for efficient integer factorization -using the Elliptic Curve Method (ECM). This method excels at finding -relatively small prime factors of very large numbers, making it a -crucial tool in computational number theory and cryptography. - -It leverages the high-performance GNU MP (Multiple Precision) library -for arbitrary-precision arithmetic, allowing it to handle integers of -virtually any size with accuracy and speed. - -Key applications include: - -- **Cryptanalysis**: Analyzing cryptographic systems (e.g., RSA). -- **Number Theory Research**: Exploring properties of integers and - primes. -- **Computational Mathematics**: Decomposing numbers into prime - components. - -GMP-ECM offers a robust and optimized solution for researchers, -cryptographers, and mathematicians needing efficient large integer -factorization. +GMP-ECM is a program to factor integers using the Elliptic Curve Method +(ECM), based on the GNU MP multiprecision library. diff --git a/math/grace/pkg-descr b/math/grace/pkg-descr index 7d92f6141c1e..353625118e9a 100644 --- a/math/grace/pkg-descr +++ b/math/grace/pkg-descr @@ -1,24 +1,16 @@ -Grace is a powerful WYSIWYG (What You See Is What You Get) 2D plotting -tool for the X Window System, built with Motif. It is the successor to -ACE/gr (Xmgr) and provides an extensive set of features for creating -high-quality scientific plots. +Grace is a WYSIWYG 2D plotting tool for the X Window System and M*tif, +successor of ACE/gr (Xmgr). A few of its features are: -Key features include: + * User defined scaling, tick marks, labels, symbols, line styles, + colors. + * Batch mode for unattended plotting. + * Read and write parameters used during a session. + * Polynomial regression, splines, running averages, DFT/FFT, + cross/auto-correlation. + * Exports high-resolution graphics to (E)PS, PDF, MIF, and SVG + formats + * Supports cross-platform PNM, JPEG and PNG formats -- **Customizable Plot Elements**: User-defined scaling, tick marks, - labels, symbols, line styles, and colors. -- **Batch Processing**: Supports batch mode for unattended plotting, - ideal for automated data visualization. -- **Session Management**: Ability to read and write parameters used - during a session for reproducibility. -- **Data Analysis**: Built-in functions for polynomial regression, - splines, running averages, DFT/FFT, and cross/auto-correlation. -- **High-Resolution Export**: Exports graphics to various formats, - including (E)PS, PDF, MIF, and SVG. -- **Image Format Support**: Supports cross-platform PNM, JPEG, and PNG - image formats. -- **Dual Interface**: Offers both a convenient point-and-click graphical - interface and a comprehensive command-line interface for advanced control. - -Grace is an invaluable tool for scientists, engineers, and researchers -who require precise and visually appealing data representation. +While grace has a convenient point-and-click interface, most parameter +settings and operations are available through a command line interface +(found in Data/Commands). diff --git a/math/grpn/pkg-descr b/math/grpn/pkg-descr index 0bd93d9e6f01..bab25a81dcde 100644 --- a/math/grpn/pkg-descr +++ b/math/grpn/pkg-descr @@ -1,21 +1,7 @@ -GRPN is a powerful and intuitive Reverse Polish Notation (RPN) calculator -designed for the X Window System, built with the GIMP Toolkit (GTK). -RPN calculators streamline complex calculations by eliminating the need -for parentheses and prioritizing operational flow. +GRPN is a RPN calculator for the X Window system built using +the GIMP Toolkit (GTK). -This graphical calculator offers extensive support for various number -types and display formats, making it a versatile tool for mathematicians, -engineers, and students. Key features include: - -- **Diverse Data Types**: Handles calculations with real numbers, - complex numbers, matrices, and complex matrices. -- **Flexible Radix Modes**: Numbers can be displayed in four different - radix modes (e.g., binary, octal, decimal, hexadecimal) for enhanced - versatility in various computing contexts. -- **Complex Number Representation**: Complex numbers can be viewed in - either Cartesian (a + bi) or polar (r * e^(i*theta)) form, catering - to different analytical needs. - -GRPN provides a robust and user-friendly environment for performing -advanced mathematical computations, combining the efficiency of RPN -with the visual clarity of a modern graphical interface. +GRPN works with real numbers, complex numbers, matrices, and +complex matrices. Numbers can be displayed in 4 different +radix modes, and complex numbers can be displayed in either +Cartesian or polar form. diff --git a/math/ised/pkg-descr b/math/ised/pkg-descr index 8b5b60298904..09a59104fa4e 100644 --- a/math/ised/pkg-descr +++ b/math/ised/pkg-descr @@ -1,21 +1,10 @@ -ised is a versatile command-line tool designed for generating number -sequences and performing arithmetic evaluations. Unlike large GUI-based -mathematical software, ised is specifically tailored for efficient use -within shell scripting environments, complementing GNU core utilities. +ised is a command-line tool for generating number sequences and +arithmetic evaluation. Unlike big gui-based software (e.g. Mathematica, +Derive, Matlab, Octave,...) it is intended for use in shell scripting, +together with gnu core utilities. -Its primary advantage lies in its ability to generalize all functions -to operate on one-dimensional arrays, making it highly flexible for -various tasks: - -- **Sequence Generation**: Functions similarly to `seq` for creating - numerical sequences, useful for loop indexing. -- **File Processing**: Enables line-by-line arithmetic processing of - data files. -- **Floating-Point Math**: Provides robust floating-point arithmetic - capabilities for shells that lack native support. -- **Interactive Calculator**: Can be used interactively as an extended - calculator for quick computations. - -ised is an invaluable utility for system administrators, developers, -and researchers who require a lightweight yet powerful tool for numerical -manipulation and scripting in command-line environments. +Its main advantage is that all functions are generalized to operate +on one-dimensional arrays. It can be used for loop indexing (much +like seq), line-by-line arithmetic processing of files, floating +point math for shells that don't support it natively, or interactively, +as extended calculator. diff --git a/math/jags/pkg-descr b/math/jags/pkg-descr index a49c272e75fe..6b90268b9334 100644 --- a/math/jags/pkg-descr +++ b/math/jags/pkg-descr @@ -1,22 +1,10 @@ -JAGS (Just Another Gibbs Sampler) is a program for analyzing Bayesian -hierarchical models using Markov Chain Monte Carlo (MCMC) simulation. -It provides a flexible framework for statistical modeling, especially -for complex models lacking analytical solutions. +JAGS is Just Another Gibbs Sampler -- a program for analysis of +Bayesian hierarchical models using Markov Chain Monte Carlo (MCMC) +simulation. -JAGS builds upon the foundational concepts of the BUGS program, with -specific differences detailed in its documentation. +The functionality of JAGS is based on the BUGS program created by the +MRC Biostatistics Unit (http://www.mrc-bsu.cam.ac.uk/). There is a short +manual that describes the differences between JAGS and BUGS. -Key features: - -- **Bayesian Modeling**: Facilitates specification and analysis of - Bayesian hierarchical models. -- **MCMC Simulation**: Employs Gibbs sampling for efficient exploration - of posterior distributions. -- **Extensibility**: Allows users to define custom distributions and - functions. -- **Compatibility**: Many BUGS examples are adapted to run with JAGS, - forming a robust test suite. - -JAGS is an invaluable tool for statisticians and researchers in fields -like epidemiology, ecology, and social sciences, where Bayesian inference -and MCMC methods are widely applied. +Some of the BUGS examples have been modified to run with JAGS, and have +been turned into a test suite. diff --git a/math/jeuclid/pkg-descr b/math/jeuclid/pkg-descr index d965e7f9e3ae..f3796b20eb5c 100644 --- a/math/jeuclid/pkg-descr +++ b/math/jeuclid/pkg-descr @@ -1,21 +1,4 @@ -JEuclid offers a comprehensive MathML rendering solution, providing -tools to display and convert mathematical expressions. MathML (Mathematical -Markup Language) is an XML-based standard for describing mathematical -notation, and JEuclid ensures its accurate and versatile presentation. - -This package includes several key components: - -- **MathViewer application**: A standalone viewer for MathML content. -- **Command-line converters**: Transform MathML into various other - formats for broader compatibility. -- **Ant task**: Facilitates automated conversion workflows, ideal for - build processes. -- **Display components for AWT and Swing**: Enables seamless integration - of MathML rendering into Java desktop applications. -- **Apache Cocoon component**: Supports dynamic generation and delivery - of MathML content within web applications. - -JEuclid is an invaluable resource for developers and content creators -working with mathematical content, providing flexible options for -rendering, conversion, and integration across different platforms and -applications. +JEuclid is a complete MathML rendering solution, consisting of: a MathViewer +application, command line converters from MathML to other formats, an ant +task for autmated conversion, display components for AWT and Swing and a +component for Apache Cocoon. diff --git a/math/jlatexmath/pkg-descr b/math/jlatexmath/pkg-descr index baff68705d0f..a87b7ca1ed18 100644 --- a/math/jlatexmath/pkg-descr +++ b/math/jlatexmath/pkg-descr @@ -1,24 +1,19 @@ -JLaTeXMath is a Java API designed to render mathematical formulas -written in LaTeX. It provides a robust solution for displaying complex -mathematical expressions within Java applications, with UTF-8 as the -default encoding. +The goal of this Java API is to display mathematical formulas written in +LaTeX. The default encoding is UTF-8. -The library supports a wide range of LaTeX commands and features, including: +The most of LaTeX commands are available and : -- **Comprehensive Command Support**: Most standard LaTeX commands, - macros from `amsmath`, and symbols from `amssymb` and `stmaryrd`. -- **Graphics Integration**: Supports `\includegraphics` (without options) - and `\over` macro. -- **Extended Accents**: Includes accents from the `amsxtra` package. -- **Color/Graphics Manipulation**: Provides `\definecolor`, `\textcolor`, - `\colorbox`, `\fcolorbox` (from `color` package), and `\rotatebox`, - `\reflectbox`, `\scalebox` (from `graphicx`). -- **Unicode/Font Support**: Detects and loads fonts for Latin, Cyrillic, - and Greek Unicode characters. -- **Custom Commands/Environments**: Supports `\newcommand`, `\newenvironment`, - `array`, `matrix`, `pmatrix`, `eqnarray`, and `cases` environments. -- **Embedded Fonts**: Fonts embedded in JAR for PDF, PS, EPS generation - via FOP 0.95; SVG export also supported. - -JLaTeXMath is an invaluable tool for Java developers needing to integrate -high-quality mathematical typesetting into their applications. + 1) macros from amsmath and symbols from amssymb and stmaryrd; + 2) \includegraphics (without options); + 3) the TeX macro \over; + 4) accents from amsxtra package; + 5) the macros \definecolor, \textcolor, \colorbox and \fcolorbox from the + package color; + 6) the macros \rotatebox, \reflectbox and \scalebox from the package graphicx; + 7) the most of latin unicode characters are available and cyrillic or + greek characters are detected for the loading of the different fonts; + 8) the commands \newcommand and \newenvironment; + 9) the environments array, matrix, pmatrix,..., eqnarray, cases; +10) the fonts are embedded in the jar file to be used by fop 0.95 to generate + PDF, PS or EPS (SVG export with shaped fonts works fine too); +11) and probably other things I forgot... diff --git a/math/lcalc/pkg-descr b/math/lcalc/pkg-descr index 52cd08ff623e..0ec4998fb03e 100644 --- a/math/lcalc/pkg-descr +++ b/math/lcalc/pkg-descr @@ -1,19 +1,4 @@ -Lcalc is a specialized program designed for computations involving -L-functions, which are fundamental objects in analytic number theory -with deep connections to prime numbers and other arithmetic properties. -This utility provides powerful capabilities for both calculating values -and finding zeros of these complex functions. +This program computes zeros and values of L-function. -The package offers a dual approach for users: - -- **lcalc command-line program**: A direct and efficient tool for - interactive computation and analysis of L-functions. -- **L-function C++ class library**: Provides a robust and flexible - interface for C++ programmers to integrate L-function computations - into their own applications, leveraging the library's optimized - algorithms. - -Lcalc is an invaluable resource for mathematicians, number theorists, -and researchers working on problems related to the Riemann Hypothesis, -elliptic curves, and other areas of advanced number theory, offering -precise numerical tools for exploring the intricate behavior of L-functions. +It installs the L-function c++ class library and, the command line program +lcalc. diff --git a/math/ldouble/pkg-descr b/math/ldouble/pkg-descr index aef898ccdb99..c53aea0789d0 100644 --- a/math/ldouble/pkg-descr +++ b/math/ldouble/pkg-descr @@ -1,20 +1,2 @@ -The ldouble library provides a comprehensive collection of elementary -mathematical functions, probability functions, and elliptic integrals, -all implemented with extended precision. This library is specifically -designed for computations requiring higher accuracy than standard -double-precision floating-point numbers. - -It supports 80-bit long double precision on x86 architectures and -128-bit long double precision where available, offering enhanced -numerical stability and reduced round-off errors for demanding -scientific and engineering applications. This extended precision is -crucial for fields such as: - -- Computational physics and chemistry -- High-precision financial modeling -- Numerical analysis and algorithm development -- Any application where accumulated floating-point errors can lead - to significant inaccuracies. - -By leveraging ldouble, developers can achieve greater confidence in the -accuracy of their complex mathematical computations. +Library of elementary mathematical functions, probability and elliptic +integrals in 80-bit (x86) or 128-bit long double precision. diff --git a/math/libbraiding/pkg-descr b/math/libbraiding/pkg-descr index db445dce1c0c..9ebe3e74d4cd 100644 --- a/math/libbraiding/pkg-descr +++ b/math/libbraiding/pkg-descr @@ -1,22 +1,7 @@ -libbraiding is a C++ shared library that exposes the functionalities -of the Braiding program, a tool for computations within braid groups. -Braid groups are fundamental mathematical objects in topology and algebra, -used to describe the intertwining of strands. +This is a project to expose the functionalitis of the Braiding program as a +shared library. The original goal is to include it as a component of SageMath, +but it can be used in any other c++ program. -This library provides a robust and efficient framework for performing -various computations on braid groups, including: +It allows various computations on braid groups, such as normal forms. -- **Normal Forms**: Calculating canonical representations of braids, - essential for comparison and analysis. -- **Other Computations**: Facilitating a range of algebraic operations - and manipulations within braid groups. - -While initially conceived for integration as a component of SageMath, -a comprehensive open-source mathematics software system, libbraiding -is designed to be a standalone library. This allows any C++ program -to leverage its powerful capabilities for braid group computations. - -libbraiding is an invaluable resource for mathematicians, computer -scientists, and researchers working in topology, group theory, and -related fields, offering a flexible and high-performance solution -for exploring the intricate world of braids. +See also: https://github.com/miguelmarco/libbraiding diff --git a/math/libhomfly/pkg-descr b/math/libhomfly/pkg-descr index 968289414460..346e677d7a2d 100644 --- a/math/libhomfly/pkg-descr +++ b/math/libhomfly/pkg-descr @@ -1,21 +1,6 @@ -libhomfly is a specialized C library designed to compute the HOMFLY -polynomial of a link. The HOMFLY polynomial is a powerful invariant -in knot theory, used to distinguish different knots and links, providing -crucial information about their topological properties. +Library to compute the homfly polynomial of a link -This library is a robust conversion of the original program by -Robert J. Jenkins Jr., refactored into a shared library for broader -applicability and ease of integration. It offers a straightforward -interface: - -- **Input**: Accepts a character string formatted identically to the - input files used by the original standalone program, representing - the link's structure. -- **Output**: Returns a string containing the computed HOMFLY polynomial, - mirroring the output format of the original utility. - -libhomfly is an invaluable resource for mathematicians, researchers, -and developers working in knot theory, topology, and related fields. -It provides a reliable and efficient computational engine for analyzing -the complex structures of knots and links, enabling further research -and application development. +This is basically a conversion of the program written by Robert J Jenkins Jr +into a shared library. It accepts as entry a character string, formatted in the +same way as the input files that the original code used. The returned value is +the string that the original program would print on screen. diff --git a/math/libocas/pkg-descr b/math/libocas/pkg-descr index f5a6f5cfb723..3647dd5e18da 100644 --- a/math/libocas/pkg-descr +++ b/math/libocas/pkg-descr @@ -1,21 +1,8 @@ -libocas implements the Optimized Cutting Plane Algorithm (OCAS), a -highly efficient method for training linear Support Vector Machine (SVM) -classifiers on large-scale datasets. SVMs are powerful supervised -learning models used for classification and regression analysis. +libocas implements an Optimized Cutting Plane Algorithm (OCAS) for training +linear SVM classifiers from large-scale data. The computational effort of +OCAS scales with O(m log m) where m is the sample size. In an extensive +empirical evaluation, OCAS significantly outperforms current state-of-the-art +SVM solvers. -OCAS is specifically designed to handle big data, offering exceptional -computational efficiency that scales with O(m log m), where 'm' is the -sample size. Empirical evaluations demonstrate that OCAS significantly -outperforms many current state-of-the-art SVM solvers, making it ideal -for applications requiring fast and accurate classification. - -Beyond core SVM training, libocas also includes the COFFIN framework, -which enables efficient training of translation-invariant image classifiers -from virtual examples. This feature is particularly useful for computer -vision tasks where robust image classification is needed, even with -limited real-world training data. - -libocas is an invaluable library for machine learning practitioners, -data scientists, and researchers who need high-performance tools for -large-scale classification problems, especially in areas like image -recognition and pattern detection. +libocas also implements the COFFIN framework for efficient training of +translation invariant image classifiers from virtual examples. diff --git a/math/libranlip/pkg-descr b/math/libranlip/pkg-descr index 2f475a42f002..3e104d52d64a 100644 --- a/math/libranlip/pkg-descr +++ b/math/libranlip/pkg-descr @@ -1,17 +1,5 @@ -libranlip is a C++ library by G. Beliakov for generating random variates -from distributions with arbitrary Lipschitz-continuous densities. It -produces random numbers following smooth probability distributions, -essential for advanced simulations and statistical modeling. - -The library uses the efficient acceptance/rejection method for sampling, -effective for densities up to approximately five dimensions. Users provide -their desired density function via a simple syntax. - -libranlip is invaluable for researchers and developers in: - -- **Monte Carlo simulations**: For complex systems. -- **Statistical inference**: Generating samples for Bayesian methods. -- **Numerical analysis**: Exploring probability distributions. - -It offers a flexible and powerful solution for high-fidelity random -number generation in C++ applications. +libranlip is a C++ library created by G. Beliakov, which generates random +variates with arbitrary Lipschitz-continuous densities via the acceptance / +rejection method. The density should have a dimension of no more than about +five. The user needs to supply the density function using a simple syntax, and +then call the methods of construction and generation provided in libranlip. diff --git a/math/linpack/pkg-descr b/math/linpack/pkg-descr index da6d1ea6ad84..29253078121f 100644 --- a/math/linpack/pkg-descr +++ b/math/linpack/pkg-descr @@ -1,20 +1,4 @@ -LINPACK is a foundational software library for performing numerical -linear algebra. Developed by a distinguished team including Jack Dongarra, -Jim Bunch, Cleve Moler, and Pete Stewart, it was released on February 1, 1984. - -Despite its age, LINPACK remains highly significant in the field of -high-performance computing. It provides a robust collection of Fortran -subroutines for solving common problems in linear algebra, such as: - -- Solving systems of linear equations -- Least squares solutions -- Eigenvalue problems -- Matrix factorization (e.g., LU, Cholesky, QR) - -Historically, LINPACK has been a cornerstone for benchmarking the -performance of supercomputers (via the LINPACK Benchmark). It also -formed a critical component of early numerical software environments, -including MATLAB. While more modern libraries exist, LINPACK's -algorithms are still widely respected for their accuracy and efficiency, -making it a valuable resource for understanding and implementing core -linear algebra operations. +Developed by Jack Dongarra, Jim Bunch, Cleve Moler and Pete Stewart. + 1 Feb 84 +Used as part of Matlab, and often used to benchmark machines. +Otherwise it is a very good linear algebra package. diff --git a/math/lll_spect/pkg-descr b/math/lll_spect/pkg-descr index cc8a319fb3d0..288be09eac13 100644 --- a/math/lll_spect/pkg-descr +++ b/math/lll_spect/pkg-descr @@ -1,21 +1,6 @@ -lll_spect evaluates Linear Congruential Pseudorandom Number Generators -(LCGs) using a normalized spectral test. This test measures an LCG's -statistical randomness. - -It employs the Lenstra-Lenstra-Lovasz (LLL) basis reduction algorithm -to approximate the shortest vector, enabling efficient calculation in -polynomial time. This makes the spectral test practical for higher -dimensions. - -Key features: - -- **LCG Quality Assessment**: Robust method for LCG statistical - properties. -- **High-Dimensional Testing**: Tests up to 24 dimensions for modern - simulations. -- **Parameter Optimization**: Includes examples for finding optimal - LCG parameters. - -lll_spect is invaluable for researchers and developers working with PRNGs, -helping select and validate LCGs for applications requiring strong -statistical properties. +This software calculates a normalized version of the classical spectral test for +linear congruential pseudorandom number generators (LCGs), where the shortest +vector is replaced by an approximation obtained with the Lenstra-Lenstra-Lovasz +basis reduction algorithm, which can be calculated in polynomial time. The code +is able to test in up to 24 dimensions, and includes an example of how to use +the test to search for good LCG parameters. diff --git a/math/lrng/pkg-descr b/math/lrng/pkg-descr index 154a3176a74d..1cef25662d44 100644 --- a/math/lrng/pkg-descr +++ b/math/lrng/pkg-descr @@ -1,9 +1,3 @@ -LRNG (Lightweight Random Number Generator) is a comprehensive collection of -high-quality, uniform pseudorandom number generators (PRNGs). Implemented in C, -LRNG is built upon robust algorithms developed by prominent researchers -Francois Panneton, Pierre L'Ecuyer, and Makoto Matsumoto. - -This library provides a reliable and efficient solution for applications -requiring statistically sound random numbers, such as simulations, -cryptography, and scientific computing. Its lightweight nature makes it -suitable for integration into various projects without significant overhead. +lrng is a collection of uniform pseudorandom number +generators, written in C, and based on algorithms by Francois +Panneton, Pierre L'Ecuyer, and Makoto Matsumoto. diff --git a/math/m4ri/pkg-descr b/math/m4ri/pkg-descr index 523688e9107f..066f20310840 100644 --- a/math/m4ri/pkg-descr +++ b/math/m4ri/pkg-descr @@ -1,24 +1,5 @@ -M4RI is a high-performance library designed for rapid arithmetic operations -with dense matrices over the finite field F2 (GF(2)). F2, the field with -two elements (0 and 1), is fundamental in areas like coding theory, -cryptography, and computational algebra. - -The library's name, M4RI, originates from its core implementation of the -"Method of the Four Russians" inversion algorithm, a technique published -by Gregory Bard. This algorithm, inspired by Kronrod's multiplication -method, provides significant speedups for matrix operations over F2. - -M4RI offers optimized routines for various matrix manipulations, including: - -- **Multiplication**: Efficiently computes products of dense matrices. -- **Inversion**: Fast calculation of matrix inverses. -- **Gaussian Elimination**: Solves systems of linear equations. - -This makes M4RI an invaluable tool for: - -- **Error-correcting codes**: Implementing and analyzing binary codes. -- **Cryptography**: Developing and testing algorithms based on binary fields. -- **Computational Algebra**: Solving problems in linear algebra over F2. - -M4RI provides researchers and developers with a powerful and efficient -solution for numerical computations in specialized algebraic structures. +M4RI is a library for fast arithmetic with dense matrices over F2. The name M4RI +comes from the first implemented algorithm: The "Method of the Four Russians" +inversion algorithm published by Gregory Bard. This algorithm in turn is named +after the "Method of the Four Russians" multiplication algorithm which is +probably better referred to as Kronrod's method. diff --git a/math/m4rie/pkg-descr b/math/m4rie/pkg-descr index b20784e6debb..1ad7f9e56440 100644 --- a/math/m4rie/pkg-descr +++ b/math/m4rie/pkg-descr @@ -1,21 +1,4 @@ -M4RIE is a high-performance library dedicated to fast arithmetic operations -with dense matrices over finite fields of the form GF(2^e), where 'e' -ranges from 2 to 16. Finite fields, also known as Galois fields, are -fundamental in areas like coding theory, cryptography, and computational -algebra. +M4RIE is a library for fast arithmetic with dense matrices over GF(2^e) for +2<=e<=16. The name stems from the fact that is relies heavily on M4RI. -Building upon the capabilities of the M4RI library, M4RIE extends its -functionality to support these larger finite fields, providing optimized -routines for matrix manipulations such as multiplication, inversion, -and Gaussian elimination. This makes it an invaluable tool for: - -- **Error-correcting codes**: Designing and implementing robust codes - for data transmission and storage. -- **Cryptography**: Developing and analyzing cryptographic algorithms - that rely on finite field arithmetic. -- **Computational Algebra**: Solving systems of linear equations and - performing other algebraic computations over specific finite fields. - -M4RIE offers researchers and developers a powerful and efficient solution -for numerical computations in specialized algebraic structures, crucial -for various advanced scientific and engineering applications. +See also: https://github.com/malb/m4rie diff --git a/math/math77/pkg-descr b/math/math77/pkg-descr index fed52e2276fe..0713b02ab1ac 100644 --- a/math/math77/pkg-descr +++ b/math/math77/pkg-descr @@ -1,17 +1,2 @@ -MATH77 provides a foundational collection of mathematical subprogram -libraries specifically designed for use with Fortran 77. Developed by -CalTech, these libraries offer a robust and historically significant -suite of routines for various numerical computations. - -Fortran 77 has been a cornerstone of scientific and engineering computing -for decades, and MATH77 complements this by offering optimized and -reliable implementations of common mathematical operations. These -subprograms typically cover areas such as: - -- Linear algebra (e.g., matrix operations, solving linear systems) -- Numerical analysis (e.g., root finding, integration, differentiation) -- Special functions (e.g., Bessel functions, error functions) - -This package is particularly valuable for maintaining and developing -applications that rely on established Fortran 77 codebases, ensuring -accuracy and performance in numerical tasks. +Mathematical subprogram libraries for Fortran 77. +Created by CalTech. diff --git a/math/mbasecalc/pkg-descr b/math/mbasecalc/pkg-descr index 0fbae66b4b05..d8d60adacc17 100644 --- a/math/mbasecalc/pkg-descr +++ b/math/mbasecalc/pkg-descr @@ -1,15 +1,3 @@ -mbasecalc is a versatile base calculator, inspired by the original -'basecalc' example from the Xlib Programming Manual by O'Reilly. -It provides a convenient tool for performing arithmetic operations -and conversions across different numerical bases. - -Unlike its predecessor, mbasecalc is designed to be cross-platform, -making it accessible and functional on various operating systems -and environments. Users can effortlessly switch between common bases -such as binary, octal, decimal, and hexadecimal, facilitating tasks -involving low-level programming, network protocols, or digital logic. - -This utility is particularly useful for developers, students, and -anyone who frequently needs to work with numbers in different bases, -offering a straightforward and efficient solution for base conversions -and calculations. +basecalc came with Xlib Programming Manual from O'Reilly as an +example of X lib programming. mbasecalc is an immitation of basecalc +which is available on different platforms. diff --git a/math/miracl/pkg-descr b/math/miracl/pkg-descr index 3e94ad7657de..390b1d04931a 100644 --- a/math/miracl/pkg-descr +++ b/math/miracl/pkg-descr @@ -1,21 +1,16 @@ -MIRACL (Multiprecision Integer and Rational Arithmetic C/C++ Library) -is a powerful Big Number Library for implementing cryptographic systems. -It provides primitives for integrating advanced number-theoretic -cryptography into real-world applications. +[ excerpt from developer's web site ] -Primarily a tool for cryptographic system implementers, MIRACL offers -robust support for: - -- **Public Key Cryptography**: RSA, Diffie-Hellman Key Exchange, DSA. -- **Elliptic Curve Cryptography (ECC)**: Full support over GF(p) - and GF(2m), including esoteric ECC schemes. -- **Lucas Function Based Schemes**: Support for less common techniques. - -MIRACL enables efficient work with large numbers foundational to modern -cryptography. It's a C library with a C++ wrapper that simplifies -program development, with examples in both languages. - -This library is invaluable for security researchers, cryptographers, -and developers building secure communication protocols, digital -signature systems, and other applications requiring high-assurance -cryptographic primitives. +MIRACL is a Big Number Library which implements all of the primitives +necessary to design Big Number Cryptography into your real-world +application. It is primarily a tool for cryptographic system +implementors. RSA public key cryptography, Diffie-Hellman Key +exchange, DSA digital signature, they are all just a few procedure +calls away. Support is also included for even more esoteric Elliptic +Curves and Lucas function based schemes. The latest version offers +full support for Elliptic Curve Cryptography over GF(p) and GF(2m). +Less well-known techniques can also be implemented as MIRACL allows +you to work directly and efficiently with the big numbers that are +the building blocks of number-theoretic cryptography. Although +implemented as a C library, a well-thought out C++ wrapper is +provided, which greatly simplifies program development. Most example +programs (25+ of them) are provided in both C and C++ versions. diff --git a/math/mtrxmath/pkg-descr b/math/mtrxmath/pkg-descr index 938dc6919807..3880b306adb9 100644 --- a/math/mtrxmath/pkg-descr +++ b/math/mtrxmath/pkg-descr @@ -1,20 +1,4 @@ -Matrix Math (mtrxmath) is a dedicated software tool designed for the -rapid and straightforward computation of various matrix functions and -operations. It provides a user-friendly environment for manipulating -matrices of arbitrary size, making complex linear algebra tasks -accessible and efficient. - -Key functionalities include: - -- **Fundamental Operations**: Addition, subtraction, and multiplication - of matrices. -- **Advanced Operations**: Matrix inversion and division, crucial for - solving systems of linear equations and other analytical tasks. -- **Extensibility**: Designed to support a comprehensive and growing - suite of matrix operations, adapting to diverse mathematical needs. - -Mtrxmath is an invaluable resource for students, engineers, scientists, -and anyone requiring quick and accurate matrix calculations. It -streamlines the process of working with linear systems, data transformations, -and other matrix-dependent computations, offering a reliable and efficient -solution for numerical analysis. +Matrix Math is software to quickly and easily compute functions of +matrices of any size. It supports addition, subtraction, +multiplication, inversion, division, and will support whatever else is +necessary. diff --git a/math/mumps4/pkg-descr b/math/mumps4/pkg-descr index d1f92de0a118..3e06b7e0a5c3 100644 --- a/math/mumps4/pkg-descr +++ b/math/mumps4/pkg-descr @@ -1,22 +1,16 @@ -MUMPS (MUltifrontal Massively Parallel sparse direct Solver) is a -Fortran 90 and MPI-based software package for efficiently solving -large sparse linear systems. It uses dynamic distributed scheduling -to handle numerical fill-in and multi-user environments. +MUMPS is a Distributed Multifrontal Solver (F90, MPI based) with Dynamic +Distributed Scheduling to accomodate both numerical fill-in and multi-user +environment. -Key capabilities: +- Solution of large linear systems with symmetric positive definite +matrices; general symmetric matrices; general unsymmetric matrices. +- Version for complex arithmetic. +- Parallel factorization and solve phases (uniprocessor version also +available). +- Iterative refinement and backward error analysis. +- Various matrix input formats: assembled format; distributed assembled +format; elemental format. +- Partial factorization and Schur complement matrix. +- Several orderings interfaced : AMD, AMF, PORD -- **Solution of Large Linear Systems**: Supports symmetric positive - definite, general symmetric, and general unsymmetric matrices. -- **Complex Arithmetic**: Version available for complex computations. -- **Parallel/Uniprocessor Modes**: Offers both parallel (MPI) and - uniprocessor factorization and solve phases. -- **Enhanced Accuracy**: Includes iterative refinement and backward - error analysis. -- **Flexible Matrix Input**: Supports assembled, distributed assembled, - and elemental formats. -- **Advanced Features**: Partial factorization, Schur complement matrix. -- **Integrated Orderings**: Interfaces with AMD, AMF, and PORD. - -This Public Domain version of MUMPS is invaluable for researchers and -engineers in scientific computing and finite element analysis, requiring -high-performance sparse direct solvers. +Note: This is the last version released under Public Domain. diff --git a/math/nfft/pkg-descr b/math/nfft/pkg-descr index 48a1939d323f..5b23f6420926 100644 --- a/math/nfft/pkg-descr +++ b/math/nfft/pkg-descr @@ -1,23 +1,16 @@ -NFFT is a C software library for computing Non-Equispaced Fast Fourier -Transforms (NFFT) and their various generalizations. It provides efficient -algorithms for scenarios where data points are not uniformly spaced, -a common occurrence in many scientific and engineering applications. +NFFT is a software library, written in C, for computing non-equispaced fast +Fourier transforms and related variations. It implements the following +transforms: -The library implements a comprehensive set of transforms, including: +1. Non-equispaced fast Fourier transform (NFFT) + - forward transform (NFFT), i.e. frequency to time/space domain + - adjoint transform (adjoint NFFT), i.e. time/space to frequency domain -- **Non-Equispaced Fast Fourier Transform (NFFT)**: - - Forward transform (frequency to time/space domain). - - Adjoint transform (time/space to frequency domain). -- **Generalizations**: - - NNFFT: For arbitrary nodes in both time and frequency domains. - - NFCT, NFST: Real-valued data, including (co)sine transforms. - - NFSFT: Transforms on the sphere S^2. - - NFSOFT: Transforms on the rotation group. - - NSFFT: Transforms on the hyperbolic cross. -- **Generalized Inverse Transformations**: Based on iterative methods - like CGNR/CGNE. - -NFFT is an invaluable tool for researchers and developers in fields -such as medical imaging, radio astronomy, geophysics, and signal -processing, where non-uniform sampling is prevalent and efficient -Fourier analysis is critical. +2. Generalisations + - to arbitrary nodes in time and frequency domain (NNFFT) + - to real-valued data, i.e. (co)sine transforms, (NFCT, NFST) + - to the sphere S^2 (NFSFT) + - to the rotation group (NFSOFT) + - to the hyperbolic cross (NSFFT) +3. Generalised inverse transformations based on iterative methods, e.g. + CGNR/CGNE diff --git a/math/ngraph/pkg-descr b/math/ngraph/pkg-descr index 37125e44a9d4..72cb3b16d174 100644 --- a/math/ngraph/pkg-descr +++ b/math/ngraph/pkg-descr @@ -1,19 +1,30 @@ -Ngraph is a versatile 2D plotting tool designed for students, scientists, -and engineers. It specializes in visualizing numerical data read from -general ASCII text files, providing a straightforward way to generate -graphs for analysis and presentation. +Ngraph is prepared to plot 2-dimensional graph for students, +scientists and engineers. The program reads numerical data from +general ASCII text files, and plot to graph. -Key features include: +** Tips ** -- **Data Visualization**: Creates 2-dimensional plots from numerical - data stored in ASCII text files. -- **User-Friendly Interface**: Designed for ease of use, catering to - users in scientific and engineering disciplines. -- **Kanji Font Support**: (Optional) Supports Kanji fonts for Japanese - language display, configurable via environment variables and specific - font installations (e.g., `ja-ngraph-fonts`, `ja-kanji18`, `ja-kanji26`). - Users can also customize font settings in `Ngraph.ini`. + - This program support Kanji font. If you want to use it, + please set environment variable LANG to ja_JP.EUC. -Ngraph provides a practical and accessible solution for generating -scientific plots, making it a valuable utility for data analysis and -graphical representation of experimental or simulated results. + (cf, under csh/tcsh) + % setenv LANG ja_JP.EUC + + and you need.... + - kinput2 + - X True Type or X True Type Font server[best], + or kanji18 and kanji26 fonts, these fonts are in below ports[better], + - ja-ngraph-fonts (japanese/ngraph-fonts) + - ja-kanji18 (japanese/kanji18) + - ja-kanji26 (japanese/kanji26) + or to change font name in Ngraph.ini as below[poor]. + +font_map=Mincho,1,-*-fixed-medium-r-normal--*-*-75-75-c-*-jisx0208.1983-0 +font_map=Gothic,1,-*-fixed-medium-r-normal--*-*-75-75-c-*-jisx0208.1983-0 + + - You can get documentation in Japanese from below URL. + +** Acknowledgements to this ports file ** + Special thanks to: + Satoshi Ishizaka + Nobuhiro Yasutomi diff --git a/math/numdiff/pkg-descr b/math/numdiff/pkg-descr index 95dfd74060da..1a99d16b3f93 100644 --- a/math/numdiff/pkg-descr +++ b/math/numdiff/pkg-descr @@ -1,21 +1,8 @@ -Numdiff is a specialized command-line utility designed for intelligent -comparison of files containing numerical data. Unlike standard `diff` -tools, Numdiff excels at identifying semantic differences between files -by intelligently handling numerical variations and formatting discrepancies. +Numdiff is a little program that can be used to compare putatively +similar files line by line and field by field, ignoring small numeric +differences or/and different numeric formats. -Its core functionality allows users to compare putatively similar files -line by line and field by field, with key features including: +Equivalently, Numdiff is a program with the capability to appropriately +compare files containing numerical fields (and not only). -- **Tolerance for Numeric Differences**: Ignores small numerical - variations, focusing on significant changes. -- **Format Agnostic Comparison**: Accommodates different numerical - formats (e.g., scientific notation, varying precision), ensuring - accurate comparisons regardless of presentation. -- **Mixed Content Handling**: Capable of comparing files that contain - both numerical fields and other types of data. - -Numdiff is an invaluable tool for scientists, engineers, and developers -who frequently work with simulation outputs, experimental data, or -configuration files where minor numerical fluctuations or formatting -differences should not be flagged as significant changes. It streamlines -the process of verifying data integrity and tracking meaningful updates. +% numdiff file1 file2 diff --git a/math/ocamlgsl/pkg-descr b/math/ocamlgsl/pkg-descr index 157319abbd30..f75b59966238 100644 --- a/math/ocamlgsl/pkg-descr +++ b/math/ocamlgsl/pkg-descr @@ -1,22 +1,2 @@ -OCamlGSL provides a robust and comprehensive interface to the GNU -Scientific Library (GSL) for the Objective Caml (OCaml) programming -language. The GSL is a vast collection of numerical routines for -scientific computing, offering a wide range of mathematical functions -and algorithms. - -By bridging OCaml with GSL, OCamlGSL empowers OCaml developers to -leverage high-performance, well-tested numerical capabilities directly -within their functional programming environment. This integration is -particularly beneficial for applications requiring: - -- **Numerical Analysis**: Solving differential equations, integration, - differentiation, root finding. -- **Linear Algebra**: Matrix operations, eigenvalue problems, linear systems. -- **Statistics**: Random number generation, probability distributions, - statistical tests. -- **Special Functions**: Bessel functions, Gamma functions, error functions. -- **Optimization**: Minimization and maximization algorithms. - -OCamlGSL allows OCaml programmers to perform complex scientific and -mathematical computations with the efficiency and reliability of GSL, -while retaining the expressiveness and safety features of OCaml. +ocamlgsl is an interface to GSL (GNU scientific library), for the +Objective Caml langage. diff --git a/math/physcalc/pkg-descr b/math/physcalc/pkg-descr index 250a1ceeaa36..de80c9e4e23a 100644 --- a/math/physcalc/pkg-descr +++ b/math/physcalc/pkg-descr @@ -1,21 +1,4 @@ -Physcalc is an exceptionally flexible and powerful mathematical calculator -designed for a wide range of scientific and engineering computations. -Its primary strength lies in its extensive unit conversion capabilities -and highly adaptable problem-solving interface. - -Key features include: - -- **Comprehensive Unit Conversions**: Seamlessly convert between numerous - units across various domains, such as length, mass, time, temperature, - energy, and more. -- **Flexible Math Problem Specification**: Allows users to input and - solve mathematical problems with remarkable versatility, accommodating - complex expressions and scientific notation. -- **User-Extensible Conversions**: Empowering users to define and add - their own custom unit conversions, tailoring the calculator to specific - needs and specialized fields. - -Physcalc is an invaluable tool for physicists, engineers, students, and -anyone who frequently deals with unit conversions and intricate mathematical -problems. It streamlines calculations and enhances productivity by providing -a highly customizable and efficient computational environment. +Physcalc is a neat mathematical calculator that does conversions +from many different units in many forms, and is extremely flexible +as far as specifying math problems go. You can also add your own +types of conversions. diff --git a/math/plplot/pkg-descr b/math/plplot/pkg-descr index e91b127997ae..894b1fad6833 100644 --- a/math/plplot/pkg-descr +++ b/math/plplot/pkg-descr @@ -1,24 +1,13 @@ -PLplot is a powerful and versatile library of C functions designed for -generating high-quality scientific plots from programs written in C, -C++, or Fortran. It provides a comprehensive set of tools for visualizing -data across various scientific and engineering disciplines. - -Key plotting capabilities include: - -- **Diverse Plot Types**: Supports standard x-y plots, semilog plots, - log-log plots, contour plots, 3D plots, mesh plots, bar charts, and - pie charts. -- **Multi-Graph Layouts**: Allows multiple graphs (of varying sizes) - on a single page, with multiple lines per graph. -- **Extensive Customization**: Offers different line styles, widths, - and colors. Features a virtually infinite number of distinct area - fill patterns. -- **Rich Character Set**: Includes an extended character set with four - fonts, the Greek alphabet, and a wide array of mathematical, musical, - and other symbols. Fonts are scalable to any desired size. -- **Device Agnostic Output**: Supports a variety of output devices, - with an extensible architecture for easily adding new device drivers. - -PLplot is an invaluable resource for scientists, engineers, and researchers -who need to create precise, customizable, and visually appealing plots -directly from their numerical applications. +PLplot is a library of C functions that are useful for making scientific +plots from a program written in C, C++, or Fortran. The PLplot library +can be used to create standard x-y plots, semilog plots, log-log plots, +contour plots, 3D plots, mesh plots, bar charts and pie charts. Multiple +graphs (of the same or different sizes) may be placed on a single page +with multiple lines in each graph. Different line styles, widths and +colors are supported. A virtually infinite number of distinct area fill +patterns may be used. There are almost 1000 characters in the extended +character set. This includes four different fonts, the Greek alphabet and +a host of mathematical, musical, and other symbols. The fonts can be +scaled to any desired size. A variety of output devices are supported and +new devices can be easily added by writing a small number of device +dependent routines. diff --git a/math/primegen/pkg-descr b/math/primegen/pkg-descr index 20a649df779b..9cc1d855552d 100644 --- a/math/primegen/pkg-descr +++ b/math/primegen/pkg-descr @@ -1,23 +1,7 @@ -primegen is a highly optimized and fast library designed for generating -prime numbers in sequential order. It provides an efficient solution -for applications requiring lists of primes, particularly for numbers -within the 32-bit range. +primegen is a small, fast library to generate prime numbers in order. +It generates the 50847534 primes up to 1000000000 in just 8 seconds +on a Pentium II-350; it prints them in decimal in just 35 seconds. -A key feature of primegen is its use of the Sieve of Atkin, an advanced -algorithm that significantly outperforms the traditional Sieve of -Eratosthenes for generating primes up to a given limit. This optimization -results in impressive performance: - -- Generates 50,847,534 primes up to 1,000,000,000 in approximately - 8 seconds on a Pentium II-350. -- Prints these primes in decimal format in about 35 seconds on the - same hardware. - -While primegen can theoretically generate primes up to 10^15, its -performance is specifically optimized for primes that fit within 32-bit -integers. It is an invaluable tool for: - -- **Number Theory Research**: Exploring properties of prime numbers. -- **Cryptography**: Generating primes for key creation or testing. -- **Computational Mathematics**: Any application requiring efficient - prime number generation. +primegen can generate primes up to 1000000000000000, although it +is not optimized for primes past 32 bits. It uses the Sieve of Atkin +instead of the traditional Sieve of Eratosthenes. diff --git a/math/prng/pkg-descr b/math/prng/pkg-descr index dbe213caafa0..bb6fc45b5ab2 100644 --- a/math/prng/pkg-descr +++ b/math/prng/pkg-descr @@ -1,22 +1,4 @@ -PRNG (Pseudorandom Number Generators) is a robust collection of portable, -high-performance ANSI-C implementations of various pseudorandom number -generators. Developed by Otmar Lendl and Josef Leydold, this library -is designed to provide statistically sound and efficient random number -sequences for a wide range of applications. - -The collection includes implementations of: - -- **Linear Congruential Generators (LCG)**: A classic and widely - understood method for generating pseudorandom numbers. -- **Inversive Congruential Generators (ICG)**: Offers improved - statistical properties compared to LCGs, particularly in terms of - period length and lattice structure. -- **Explicit Inversive Congruential Generators (EICG)**: Further - enhances the quality of pseudorandom sequences, often used in - more demanding simulations. - -PRNG is an invaluable resource for researchers, statisticians, and -developers in fields such as Monte Carlo simulations, cryptography, -and scientific modeling, where the quality and performance of random -number generation are critical. Its ANSI-C implementation ensures -broad compatibility and efficient execution across different platforms. +PRNG is a collection of portable, high-performance ANSI-C implementations of +pseudorandom number generators such as linear congruential, inversive +congruential, and explicit inversive congruential random number generators (LCG, +ICG and EICG, respectively) created by Otmar Lendl and Josef Leydold. diff --git a/math/py-claripy/pkg-descr b/math/py-claripy/pkg-descr index 5cda9d5601d7..e86866b3cfa6 100644 --- a/math/py-claripy/pkg-descr +++ b/math/py-claripy/pkg-descr @@ -1,21 +1,3 @@ -Claripy is a powerful and abstracted constraint-solving wrapper for Python, -designed to simplify the interaction with various underlying constraint -solvers. It provides a unified and high-level interface for defining -and solving complex symbolic constraints, making it an invaluable tool -for program analysis, symbolic execution, and security research. +Claripy is an abstracted constraint-solving wrapper for Python. -Developed as a core component of the Angr project, a well-known binary -analysis platform, Claripy is built to handle intricate symbolic expressions -and manage solver states efficiently. Its abstraction layer allows -developers to focus on defining the problem rather than the specifics -of individual solver APIs. - -Key applications include: -- Symbolic execution of binaries -- Automated exploit generation -- Program verification and bug finding -- Solving SMT (Satisfiability Modulo Theories) problems - -By providing a flexible and robust framework for symbolic reasoning, -Claripy empowers Python developers to tackle challenging problems in -computer science and security with greater ease and effectiveness. +It is being developed by the Angr project. diff --git a/math/py-fvcore/pkg-descr b/math/py-fvcore/pkg-descr index f7526eff106d..edfc1da7d7eb 100644 --- a/math/py-fvcore/pkg-descr +++ b/math/py-fvcore/pkg-descr @@ -1,21 +1,4 @@ -fvcore is a lightweight and highly optimized core library designed to -provide essential functionalities shared across various computer vision -frameworks developed at FAIR (Facebook AI Research). It serves as a -foundational toolkit, streamlining the development of advanced vision -models and applications. - -This library encapsulates common utilities and building blocks, such as: - -- **Configuration Management**: Tools for handling model configurations. -- **Logging and Monitoring**: Utilities for tracking experiment progress. -- **Data Structures**: Efficient data representations for computer - vision tasks. -- **Performance Benchmarking**: Integrated tools for evaluating and - optimizing code performance. - -A key strength of fvcore lies in its commitment to quality: all components -are meticulously type-annotated for clarity, thoroughly tested for -reliability, and rigorously benchmarked for optimal performance. It acts -as a crucial dependency for prominent FAIR projects like Detectron2, -PySlowFast, and ClassyVision, enabling developers to build robust and -efficient computer vision systems with confidence. +fvcore is a light-weight core library that provides the most common and +essential functionality shared in various computer vision frameworks +developed in FAIR, such as Detectron2, PySlowFast, and ClassyVision. All +components in this library are type-annotated, tested, and benchmarked. diff --git a/math/py-luminol/pkg-descr b/math/py-luminol/pkg-descr index 7cb63e301904..474f02748a1d 100644 --- a/math/py-luminol/pkg-descr +++ b/math/py-luminol/pkg-descr @@ -1,23 +1,3 @@ -Luminol is an efficient and lightweight Python library specifically -designed for comprehensive time series data analysis. It provides -powerful functionalities for two critical aspects of time series -investigation: anomaly detection and correlation analysis. - -As a lightweight library, Luminol is optimized for performance and -resource efficiency, making it suitable for integration into various -data pipelines and applications without significant overhead. - -Its core capabilities include: -- **Anomaly Detection**: Identifying unusual patterns or outliers in - time series data that deviate significantly from expected behavior. - This is crucial for monitoring systems, detecting fraudulent activities, - or pinpointing performance issues. -- **Correlation Analysis**: Discovering relationships and dependencies - between different time series. This feature is particularly useful - for investigating the root causes of detected anomalies, allowing - users to understand which factors might be contributing to unusual events. - -Luminol empowers data scientists, engineers, and analysts to gain deeper -insights from their time series data, enabling proactive problem-solving -and informed decision-making by effectively pinpointing and diagnosing -anomalous behavior. +Luminol is a light weight python library for time series data analysis. +The two major functionalities it supports are anomaly detection and +correlation. It can be used to investigate possible causes of anomaly. diff --git a/math/py-pytorchvideo/pkg-descr b/math/py-pytorchvideo/pkg-descr index a8dd0b918d86..b92cab0a19bc 100644 --- a/math/py-pytorchvideo/pkg-descr +++ b/math/py-pytorchvideo/pkg-descr @@ -1,23 +1,5 @@ -PyTorchVideo is a comprehensive deep learning library specifically -designed to accelerate research and development in video understanding. -Built upon the popular PyTorch framework, it offers a rich collection -of reusable, modular, and highly efficient components tailored for -video-centric deep learning tasks. - -This library streamlines the process of building and deploying video -understanding models by providing: - -- **Pre-trained Video Models**: A diverse set of state-of-the-art - architectures optimized for video data. -- **Video Datasets**: Tools and utilities for handling and processing - various video datasets. -- **Video-Specific Transforms**: Efficient data augmentation and - preprocessing techniques adapted for temporal and spatial video - characteristics. - -PyTorchVideo empowers researchers and developers to tackle challenging -problems such as action recognition, video classification, and object -tracking in videos with greater ease and efficiency. Its modular design -fosters rapid experimentation and allows for seamless integration into -existing PyTorch workflows, making it an invaluable resource for the -computer vision community. +PyTorchVideo is a deeplearning library with a focus on video understanding +work. PytorchVideo provides reusable, modular and efficient components needed +to accelerate the video understanding research. PyTorchVideo is developed using +PyTorch and supports different deeplearning video components like video models, +video datasets, and video-specific transforms. diff --git a/math/py-svgmath/pkg-descr b/math/py-svgmath/pkg-descr index f5a80bb1f657..d3d4663027eb 100644 --- a/math/py-svgmath/pkg-descr +++ b/math/py-svgmath/pkg-descr @@ -1,20 +1,2 @@ -SVGMath is a specialized command-line utility written entirely in Python -that facilitates the conversion of MathML (Mathematical Markup Language) -expressions into SVG (Scalable Vector Graphics) format. - -MathML is an XML-based markup language for describing mathematical -notation, while SVG is an XML-based vector image format for two-dimensional -graphics. The conversion provided by SVGMath is highly beneficial for: - -- **High-quality rendering**: SVG ensures that mathematical equations - are rendered sharply and clearly at any resolution, making them ideal - for web display, print, and presentations. -- **Web integration**: Easily embed complex mathematical formulas into - web pages and other digital documents without relying on raster images - or specialized fonts. -- **Accessibility**: Vector graphics maintain their quality when scaled, - improving readability for users with varying display needs. - -As a pure Python implementation, SVGMath offers a portable and easily -integrable solution for developers and content creators who need to -present mathematical content in a visually appealing and scalable format. +SVGMath is a command-line utility to convert MathML expressions +to SVG, written entirely in Python. diff --git a/math/qtiplot-doc/pkg-descr b/math/qtiplot-doc/pkg-descr index 59b9caf0649f..15600819239d 100644 --- a/math/qtiplot-doc/pkg-descr +++ b/math/qtiplot-doc/pkg-descr @@ -1,22 +1,3 @@ -This package provides the comprehensive documentation for QtiPlot, -a powerful and platform-independent application for data analysis -and visualization. QtiPlot is a free (GPL) software often compared -to commercial alternatives like Origin, offering extensive capabilities -for scientific plotting, data manipulation, and statistical analysis. - -The documentation included in this package is an essential resource -for both new and experienced QtiPlot users. It covers: - -- **Getting Started**: Installation, interface overview, and basic - operations. -- **Data Handling**: Importing, organizing, and manipulating datasets. -- **Plotting**: Creating 2D and 3D plots, customizing graphs, and - exporting results. -- **Analysis**: Performing statistical tests, curve fitting, and - signal processing. -- **Scripting**: Utilizing QtiPlot's scripting capabilities for - automation. - -By providing detailed guides, tutorials, and reference materials, this -documentation package ensures users can fully leverage QtiPlot's -features to effectively analyze and visualize their scientific data. +QtiPlot is a free (GPL) platform independent data analysis and +visualization application similar to the non-free Windows program +Origin. diff --git a/math/randlib/pkg-descr b/math/randlib/pkg-descr index ba80a16b1a4e..e42d51236e6b 100644 --- a/math/randlib/pkg-descr +++ b/math/randlib/pkg-descr @@ -1,21 +1,20 @@ -RANDLIB is a comprehensive library of routines for generating various -types of random numbers and permutations. It provides a robust set of -functions for statistical simulations and modeling, making it a valuable -tool for researchers and developers. - -The library offers routines to return: - -- **Continuous Random Deviates**: Beta, Chi-square, Exponential, F, - Gamma, Noncentral Chi-square, Noncentral F, and Univariate Normal. -- **Discrete Random Deviates**: Binomial, Negative Binomial, Multinomial, - and Poisson. -- **Uniform Random Deviates**: Real uniform deviates between specified - limits, and integer uniform deviates between specified limits. -- **Multivariate Normal Deviates**: With specified mean and covariance matrix. -- **Random Permutations**: Generates random permutations of an integer array. -- **Seed Generation**: Calculates seeds for the random number generator - from a character string. - -RANDLIB ensures a wide range of random number generation capabilities, -essential for Monte Carlo methods, statistical analysis, and other -applications requiring diverse and reliable random variates. +This library provides routines that return: + (1) Beta random deviates + (2) Chi-square random deviates + (3) Exponential random deviates + (4) F random deviates + (5) Gamma random deviates + (6) Multivariate normal random deviates (mean and covariance + matrix specified) + (7) Noncentral chi-square random deviates + (8) Noncentral F random deviates + (9) Univariate normal random deviates + (10) Random permutations of an integer array + (11) Real uniform random deviates between specif + (12) Binomial random deviates + (13) Negative Binomial random deviates + (14) Multinomial random deviates + (15) Poisson random deviates + (16) Integer uniform deviates between specified limits + (17) Seeds for the random number generator calculated from a + character string diff --git a/math/reduce-psl/pkg-descr b/math/reduce-psl/pkg-descr index 80b2ef76dd7f..a5ed5582beda 100644 --- a/math/reduce-psl/pkg-descr +++ b/math/reduce-psl/pkg-descr @@ -1,23 +1,17 @@ -REDUCE Portable Standard Lisp (PSL) is an interactive system for -general algebraic computations. It is designed to assist mathematicians, -scientists, and engineers with symbolic manipulation and numerical tasks. + REDUCE Portable Standard Lisp (PSL) -REDUCE, powered by PSL, is particularly adept at handling complex -calculations that are not feasible to perform by hand. PSL, originally -the implementation of Standard Lisp, has evolved to include numerous -facilities and is optimized for algebraic computation. +REDUCE is an interactive system for general algebraic computations of +interest to mathematicians, scientists and engineers. -Key features and benefits: +PSL was the original implementation of Standard Lisp, but now contains +many more facilities. It is quite efficient in its use of both space and +time, and has been optimized for algebraic computation. All PSL versions +of REDUCE are distributed with sufficient PSL support to run on the given +computing system. PSL is supported on many architectures and is an ideal +system for those wanting to run REDUCE as a stand-alone system. The +principal developer of PSL before it became Open Source was the Konrad +Zuse Center, Berlin (ZIB). -- **General Algebraic Computations**: Performs symbolic manipulation, - simplification, differentiation, integration, and equation solving. -- **Efficiency**: Optimized for both space and time, making it efficient - for algebraic workloads. -- **Stand-alone System**: PSL provides all necessary support to run - REDUCE as a self-contained system. -- **Broad Architecture Support**: Available on many architectures, - ensuring wide compatibility. - -REDUCE-PSL is an invaluable tool for academic and industrial users -requiring a powerful and efficient computer algebra system for research, -development, and problem-solving in various scientific and engineering domains. +It is often used as an algebraic calculator for problems that are possible +to do by hand. However, REDUCE is designed to support calculations that +are not feasible by hand. diff --git a/math/rngstreams/pkg-descr b/math/rngstreams/pkg-descr index 78af90b3a080..e8b1e5a88668 100644 --- a/math/rngstreams/pkg-descr +++ b/math/rngstreams/pkg-descr @@ -1,23 +1,11 @@ -RngStreams is a C implementation of a high-quality uniform random -number generator (RNG) that provides support for multiple "independent" -streams of random numbers. This feature is crucial for parallel and -distributed simulations, ensuring that different parts of a simulation -can draw from distinct, non-overlapping sequences of random numbers. +RngStreams is a C implementation of a high-quality uniform random number +generator that supports multiple "independent" streams of uniform random +numbers. -Developed by Pierre L'Ecuyer and Richard Simard, RngStreams is designed -to offer statistically sound and reproducible random number sequences. -Its core strength lies in managing multiple streams, which is essential -for: +It was written by Pierre L'Ecuyer and Richard Simard, who have a website +at: -- **Parallel Simulations**: Running independent simulation replicates - concurrently. -- **Distributed Computing**: Ensuring unique random number sequences - across different processors or nodes. -- **Reproducibility**: Allowing for the exact replication of simulation - results by controlling the state of each stream. +http://www.iro.umontreal.ca/~simardr/indexe.html -This GNU-style package is compiled and maintained by Josef Leydold, -and is released under the GNU Public License (GPL). RngStreams is an -invaluable tool for researchers and developers in Monte Carlo methods, -statistical modeling, and any application requiring robust and manageable -random number generation. +This GNU-style package is compiled and maintained by Josef Leydold and +released under the GNU Public License (GPL). diff --git a/math/sc/pkg-descr b/math/sc/pkg-descr index 6dbe2aa24631..d0ed9f7ad3c2 100644 --- a/math/sc/pkg-descr +++ b/math/sc/pkg-descr @@ -1,22 +1,7 @@ -sc (spreadsheet calculator) is a powerful and flexible terminal-based -spreadsheet program. It operates on rectangular tables, providing a -familiar interface of rows and columns for data organization and analysis. -Unlike graphical spreadsheets, sc is designed for efficiency and accessibility -within text-based environments. - -Upon invocation, sc presents an organized table of cells. If no file -argument is provided, it starts with an empty table, ready for user input. -Each cell in the spreadsheet can store and display various types of data: - -- **Numeric Values**: Direct numerical entries. -- **Label Strings**: Textual descriptions or headers. -- **Expressions (Formulas)**: Dynamic calculations that evaluate to - either a numeric value or a label string. These formulas can - reference other cell values, enabling complex inter-cell dependencies - and powerful data manipulation. - -sc is an invaluable tool for users who prefer a command-line interface -for data management, financial calculations, or any task requiring -spreadsheet functionality without the overhead of a graphical environment. -Its robust formula support and cell-based operations make it a versatile -choice for data analysis and reporting. +The spreadsheet calculator sc is based on rectangular tables much like +a financial spreadsheet. When invoked it presents you with a table +organized as rows and columns of cells. If invoked without a file +argument, the table is initially empty. Each cell may have associated +with it a numeric value, a label string, and/or an expression (formula) +which evaluates to a numeric value or label string, often based on other +cell values. diff --git a/math/scilab-toolbox-swt/pkg-descr b/math/scilab-toolbox-swt/pkg-descr index ed8fa452af6a..6bfc6c6303b4 100644 --- a/math/scilab-toolbox-swt/pkg-descr +++ b/math/scilab-toolbox-swt/pkg-descr @@ -1,23 +1,15 @@ -The Scilab Wavelet Toolbox (SWT) is a free software package designed -to provide comprehensive wavelet analysis tools within the Scilab -environment. Wavelet analysis is a powerful signal processing technique -used for analyzing signals and images across various scales. +Scilab Wavelet Toolbox (SWT) -SWT enables users to perform a wide range of wavelet-based operations -on both 1-D signals and 2-D images. Key functionalities include: +Wavelet is a powerful signal processing tool developed and developing +in the last two decades. Scilab Wavelet Toolbox is a free software package +to enable you using wavelet analysis tools freely in Scilab on most OSes +including GNU/Linux, BSD and Windows. Scilab Wavelet Toolbox is designed +to work with any Scilab Image Processing Toolbox like SIP +for displaying 2-D results. -- **Discrete Fast Wavelet Transform (DFWT)**: Efficient computation - of wavelet transforms. -- **Daubechies Wavelets**: Support for a popular family of wavelets. -- **1-D Signal Decomposition and Reconstruction**: - - Single-level decomposition and reconstruction. - - Multi-level decomposition and reconstruction. -- **2-D Image Decomposition and Reconstruction**: - - Single-level decomposition and reconstruction. - - Multi-level decomposition and reconstruction. - -SWT is compatible with most operating systems, including GNU/Linux, BSD, -and Windows. It is designed to integrate seamlessly with Scilab Image -Processing Toolboxes (e.g., SIP) for displaying 2-D results. This toolbox -is an invaluable resource for engineers, researchers, and students working -in signal and image processing, data compression, and feature extraction. +What Scilab Wavelet Toolbox supposed to do: +Discrete Fast Wavelet Transform, daubechies wavelets +1-D single level signal decomposition and reconstruction +1-D multi-level signal decomposition and reconstruction +2-D single level image decomposition and reconstruction +2-D multi-level image decomposition and reconstruction. diff --git a/math/scilab/pkg-descr b/math/scilab/pkg-descr index ad036b47262c..98ef3c7949d2 100644 --- a/math/scilab/pkg-descr +++ b/math/scilab/pkg-descr @@ -1,24 +1,34 @@ -Scilab is a powerful, open-source scientific software package for -numerical computations, offering a user-friendly environment for -engineers, scientists, and students. It provides a high-level -programming language and a rich set of functionalities. +Scilab is a scientific software package for numerical computations in a +user-friendly environment. -Key features: - -- **Mathematical Functions**: Hundreds of built-in functions. -- **High-Level Programming Language**: For scripting and automation. -- **2-D and 3-D Graphics**: Comprehensive visualization, including - lines, charts, histograms, surfaces, and animations. Exports to - GIF, BMP, JPEG, SVG, PDF. -- **Xcos**: Hybrid dynamic systems modeler and simulator. -- **Numerical Computation**: Linear algebra, sparse matrices, - polynomials, rational functions, differential equation solvers, - control, optimization. -- **Data Analysis**: Interpolation, approximation, signal processing, - and statistics. -- **Extensibility**: Interfaces with Fortran, C, C++, Java; numerous - modules via ATOMS. - -Scilab is an invaluable tool for numerical analysis, data processing, -and scientific visualization, a free and powerful alternative to -commercial mathematical software. +Main features + * Hundreds of mathematical functions + * High level programming language + * 2-D and 3-D graphics + * Advanced data structures and user defined data types + * Xcos: hybrid dynamic systems modeler and simulator +2-D and 3-D visualization + * Lines + * Pie charts + * Histograms + * Surfaces + * Animations + * Graphics export in many formats: GIF, BMP, JPEG, SVG, PDF... +Numerical computation + * Linear algebra + * Sparse matrices + * Polynomials and rational functions + * Simulation: explicit and implicit systems of differential + equations solvers + * Classic and robust control + * Differentiable and non-differentiable optimization +Data analysis + * Interpolation, approximation + * Signal Processing + * Statistics +Extended features + * Graphs and Networks + * Interface with Fortran, C, C++, Java + * Functions for calling Scilab from C, C++, Fortran and Java + * LabVIEW Gateway + * A large number of modules available via ATOMS diff --git a/math/sfft/pkg-descr b/math/sfft/pkg-descr index 6c02233b3f3b..5901f6f19f9c 100644 --- a/math/sfft/pkg-descr +++ b/math/sfft/pkg-descr @@ -1,19 +1,6 @@ -sfft is an optimized library for computing Discrete Fourier Transforms -(DFTs) of signals with a sparse frequency domain. It exploits this -sparsity for significantly more efficient computations than traditional -FFT algorithms. - -Developed by a team at MIT CSAIL (Hassanieh, Indyk, Katabi, Price), -sfft represents a breakthrough in sparse Fourier transform techniques. -Further performance optimizations were contributed by J. Schumacher at -ETH Zurich. - -This library is beneficial for applications where: - -- **Signals have few dominant frequencies**: E.g., compressed sensing, - spectral analysis of sparse signals. -- **Computational speed is critical**: Offering faster processing. -- **Resource efficiency is desired**: Reducing computational load. - -sfft provides a powerful tool for researchers and engineers working with -sparse spectral data, enabling faster analysis and processing. +sfft is a library to compute discrete Fourier transforms of signals with +a sparse frequency domain, using an algorithm that is more efficient than +other known FFT algorithms. It was developed by Haitham Hassanieh, Piotr +Indyk, Dina Katabi, and Eric Price at the Computer Science and Artifical +Intelligence Lab at MIT. Performance optimizations were developed by J. +Schumacher at the Computer Science Department of ETH Zurich in 2013. diff --git a/math/slatec/pkg-descr b/math/slatec/pkg-descr index a6f7421565cd..d222e326629c 100644 --- a/math/slatec/pkg-descr +++ b/math/slatec/pkg-descr @@ -1,22 +1,2 @@ -SLATEC (Scientific Library for Advanced Technology Exchange) is a -comprehensive and historically significant software library, offering -over 1400 general-purpose mathematical and statistical routines. -Originally developed by a consortium of U.S. government laboratories, -these routines are primarily written in Fortran 77, reflecting their -proven reliability and efficiency in numerical computation. - -This extensive collection covers a broad spectrum of numerical methods, -including but not limited to: - -- Linear algebra (e.g., solving systems of equations, eigenvalue problems) -- Special functions (e.g., Bessel functions, Gamma functions) -- Numerical integration and differentiation -- Interpolation and approximation -- Statistical analysis and probability distributions -- Optimization techniques - -SLATEC serves as a robust foundation for scientific and engineering -applications requiring high-quality, well-tested numerical algorithms. -It remains a valuable resource for researchers, engineers, and developers -working with Fortran-based projects or those needing a reliable suite -of established numerical methods. +SLATEC is a comprehensive software library containing over 1400 general +purpose mathematical and statistical routines written in Fortran 77. diff --git a/math/snns/pkg-descr b/math/snns/pkg-descr index cec2ff3352a8..70b4f67de61e 100644 --- a/math/snns/pkg-descr +++ b/math/snns/pkg-descr @@ -1,23 +1,24 @@ -SNNS (Stuttgart Neural Network Simulator) is a comprehensive software -simulator for neural networks, developed at the Institute for Parallel -and Distributed High Performance Systems (IPVR) at the University of -Stuttgart. It provides an efficient and flexible environment for -research and application of neural networks. +SNNS (Stuttgart Neural Network Simulator) is a software simulator for neural +networks on Unix workstations developed at the Institute for Parallel and +Distributed High Performance Systems (IPVR) at the University of Stuttgart. +The goal of the SNNS project is to create an efficient and flexible +simulation environment for research on and application of neural nets. -SNNS comprises two main components: +The SNNS simulator consists of two main components: -1. **Simulator Kernel (C-based)**: This core component handles internal - network data structures, learning, and recall operations. It supports - arbitrary network topologies and the concept of sites. The kernel - can be extended with user-defined activation functions, output - functions, site functions, and learning procedures written in C. - It can also be embedded in custom applications. -2. **Graphical User Interface (XGUI)**: Built on top of the kernel, - XGUI provides 2D and 3D graphical representations of neural networks. - It controls the kernel during simulation and includes an integrated - network editor for creating, manipulating, and visualizing neural - nets interactively. +1) simulator kernel written in C +2) graphical user interface under X -SNNS is an invaluable tool for researchers, students, and developers -working with neural networks, offering a powerful platform for -experimentation, simulation, and understanding of complex neural models. +The simulator kernel operates on the internal network data structures of the +neural nets and performs all operations of learning and recall. It can also +be used without the other parts as a C program embedded in custom +applications. It supports arbitrary network topologies and the concept of +sites. SNNS can be extended by the user with user defined activation +functions, output functions, site functions and learning procedures, which +are written as simple C programs and linked to the simulator kernel. + +The graphical user interface XGUI (X Graphical User Interface), built on top +of the kernel, gives a 2D and a 3D graphical representation of the neural +networks and controls the kernel during the simulation run. In addition, the +2D user interface has an integrated network editor which can be used to +directly create, manipulate and visualize neural nets in various ways. diff --git a/math/solitaire/pkg-descr b/math/solitaire/pkg-descr index 7e9341f2e1d6..056ca6c86725 100644 --- a/math/solitaire/pkg-descr +++ b/math/solitaire/pkg-descr @@ -1,22 +1,7 @@ -Solitaire is a unique encryption system designed by renowned cryptographer -Bruce Schneier, based on the manipulation of a deck of playing cards. -While ingeniously crafted for manual execution by a human, this system -can also be implemented computationally. +Solitaire is an encryption system based on a deck of cards by Bruce +Schneier. Although it is designed to be worked out by a human, it can +work on computers. This is the reference implementation programmed in +Perl. The program itself is installed as 'solitaire', and the source +code and test vectors are installed in share/doc/solitaire. -This package provides the reference implementation of the Solitaire -encryption algorithm, programmed in Perl. It serves as a faithful -digital representation of Schneier's original design, allowing for -both study and practical application. - -Key components installed: - -- **solitaire executable**: The primary program for performing - encryption and decryption. -- **Source code and test vectors**: Located in share/doc/solitaire, - these resources are invaluable for understanding the algorithm's - mechanics and verifying its correctness. - -**IMPORTANT SECURITY NOTE**: Before relying on this implementation for -any real-world security applications, it is strongly advised to -thoroughly read the official website and understand the nuances and -limitations of the Solitaire algorithm. +Please read the web site below before relying on this for real security. diff --git a/math/spblas/pkg-descr b/math/spblas/pkg-descr index a0903b17e7e0..8fb8a077acc3 100644 --- a/math/spblas/pkg-descr +++ b/math/spblas/pkg-descr @@ -1,24 +1,7 @@ -SPBLAS is an ANSI C++ implementation of Chapter 3 of the BLAS (Basic -Linear Algebra Subprograms) Technical Forum Standard. BLAS routines are -fundamental building blocks for high-performance numerical linear algebra, -widely used in scientific computing, engineering, and data analysis. - -This library serves as a robust starting point for developing optimized -and architecture-dependent BLAS implementations. Its C++ design offers -several advantages: - -- **Complex Arithmetic Support**: Facilitates computations involving - complex numbers directly. -- **Templates**: Enables the creation of various precision codes, - supporting single, double precision, real, and complex data types. - -SPBLAS provides comprehensive coverage of BLAS operations, including: - -- **Level 1**: Vector-vector operations (e.g., dot product, vector scaling). -- **Level 2**: Matrix-vector operations (e.g., matrix-vector multiplication). -- **Level 3**: Matrix-matrix operations (e.g., matrix-matrix multiplication), - which are crucial for many high-performance computing tasks. - -This library is an invaluable resource for developers and researchers -who need a portable, standard-compliant, and extensible foundation for -implementing efficient linear algebra routines in C++. +This is an ANSI C++ implementation of the complete ANSI C specification of +Chapter 3 of the BLAS Technical Forum Standard. The distribution is quite +small and it is meant as a starting point for developing an optimized and +architecture-dependent version. (C++ was used, rather than C, as it has support +for complex arithmetic and templates to facilitate to creation of various +precision codes.) The library includes support for all four precision types +(single, double precision, real, and complex) and Level 1, 2, and 3 operations. diff --git a/math/tomsfastmath/pkg-descr b/math/tomsfastmath/pkg-descr index 19be5c899bfe..ec56fac59dc4 100644 --- a/math/tomsfastmath/pkg-descr +++ b/math/tomsfastmath/pkg-descr @@ -1,18 +1,2 @@ -TomsFastMath (TFM) is a highly optimized and portable fixed-precision -mathematics library, specifically engineered for exceptionally fast -exponentiation operations. Designed for environments where performance -is critical, TFM provides a robust solution for cryptographic applications, -large number arithmetic, and other computational tasks demanding rapid -modular exponentiation. - -The library's fixed-precision nature means it operates on integers of -a predetermined size, offering predictable performance characteristics -and efficient memory usage. Its primary strength lies in its ability -to perform modular exponentiation (e.g., a^b mod n) with remarkable speed, -a fundamental operation in public-key cryptography algorithms like RSA -and Diffie-Hellman. - -TFM's portability ensures it can be easily integrated into various -projects across different platforms, making it a valuable asset for -developers building secure communication protocols, digital signatures, -or any system requiring high-speed, fixed-precision arithmetic. +TomsFastMath is a portable fixed precision math library designed for +very fast exponentiations. diff --git a/math/trlan/pkg-descr b/math/trlan/pkg-descr index 175406fb5c76..c9299a9f54da 100644 --- a/math/trlan/pkg-descr +++ b/math/trlan/pkg-descr @@ -1,24 +1,7 @@ -TRLAN is a portable and modular Fortran 90 software package that -implements the thick-restart Lanczos method. This advanced numerical -technique is specifically designed for efficiently solving large-scale -eigenvalue problems, particularly when only a small subset of eigenvalues -and their corresponding eigenvectors are required. - -TRLAN excels in scenarios involving real symmetric or complex Hermitian -matrices that are too large to be stored entirely in computer memory. -Its key features include: - -- **Memory Efficiency**: Handles very large matrices by avoiding full - storage, making it suitable for high-performance computing. -- **Targeted Eigenvalue Computation**: Focuses on finding a few desired - eigenvalues and eigenvectors, rather than the entire spectrum. -- **BLAS and LAPACK Integration**: Leverages highly optimized Basic - Linear Algebra Subprograms (BLAS) and Linear Algebra Package (LAPACK) - routines for core arithmetic computations, ensuring high performance. -- **Checkpointing Capability**: Supports writing checkpoint files, - allowing computations to be paused and restarted later, which is - crucial for long-running simulations or in case of system interruptions. - -TRLAN is an invaluable tool for researchers and engineers in fields -such as quantum chemistry, structural analysis, and materials science, -where efficient and robust solutions to large eigenvalue problems are essential. +This portable, modular Fortran 90 software package implements the thick-restart +Lanczos method, for use with real symmetric or complex Hermitian eigenvalue +problems where a small number of eigevalues and eigenvectors are needed, and +the matrices involved may be too large to store in computer memory. Most of +the arithmetic computations in the software are done through calls to BLAS +and LAPACK. The software can be instructed to write checkpoint files so that +it can be restarted is a later time. diff --git a/math/tvmet/pkg-descr b/math/tvmet/pkg-descr index c2a673651c07..3b3a45f3757b 100644 --- a/math/tvmet/pkg-descr +++ b/math/tvmet/pkg-descr @@ -1,24 +1,4 @@ -TVMet (Tiny Vector and Matrix template library) is a high-performance -C++ library designed for efficient linear algebra operations, particularly -suited for resource-constrained or performance-critical applications. -It leverages advanced C++ features like Meta-programming and Expression -Templates to achieve remarkable speed and memory efficiency. - -The core innovation of TVMet lies in its ability to evaluate many -mathematical expressions involving vectors and matrices at compile time. -This approach eliminates the creation of costly temporary objects during -runtime, leading to: - -- **Exceptional Performance**: Significantly faster execution compared - to traditional approaches. -- **Minimal Memory Footprint**: Reduced memory consumption, making it - ideal for embedded systems or applications with strict memory budgets. -- **Static Dimensioning**: Vector and matrix dimensions are fixed at - compile time, allowing for aggressive optimizations and compile-time - error checking. - -TVMet is an excellent choice for developers working on scientific simulations, -game development, embedded systems, or any project where fast and -memory-efficient linear algebra is paramount. It provides a robust and -optimized solution for numerical computations without compromising on -performance or resource usage. +This Tiny Vector and Matrix template library uses Meta and Expression +Templates to evaluate results at compile time, thus making it fast for +low-end systems. Temporaries are avoided because of this. The dimensions +are static and bounded at compile time. diff --git a/math/ump/pkg-descr b/math/ump/pkg-descr index 20c0a7833fd7..b8b22400d0fb 100644 --- a/math/ump/pkg-descr +++ b/math/ump/pkg-descr @@ -1,16 +1,2 @@ -UMP (Universal Math Program) is an intuitive and powerful graphical -mathematics program designed for ease of use. It provides a comprehensive -environment for performing a wide range of mathematical operations, -making complex calculations accessible to students, educators, and -professionals alike. - -Key features include robust support for: -- Complex numbers: Perform arithmetic and functions with complex values. -- Matrices: Handle matrix operations, inversions, and determinants. -- Functions: Define, plot, and analyze mathematical functions. -- Advanced calculations: Solve equations, perform calculus operations, - and work with various mathematical expressions. - -With its user-friendly graphical interface, UMP simplifies the process -of exploring mathematical concepts and solving intricate problems, -offering a visual and interactive approach to mathematics. +ump is a graphical, easy to use math program, which works with complex +numbers, matrices, functions and much more. diff --git a/math/xplot/pkg-descr b/math/xplot/pkg-descr index aa11ebcce82f..cb23a6e43f97 100644 --- a/math/xplot/pkg-descr +++ b/math/xplot/pkg-descr @@ -1,9 +1 @@ -Xplot is a versatile and efficient graphing utility specifically designed -for the X11 windowing system. It provides a robust platform for visualizing -various types of data, making it particularly useful for scientific, -engineering, and network analysis applications. - -Commonly employed to display TCP traces, Xplot can also handle other -time-series data, scatter plots, and more. Its interactive features allow -users to zoom, pan, and inspect data points directly within the X11 -environment, offering a dynamic approach to data exploration and analysis. +An X11 graphing utility. Commonly used to display TCP traces. diff --git a/math/xspread/pkg-descr b/math/xspread/pkg-descr index f335c25c1f3e..4555690ac340 100644 --- a/math/xspread/pkg-descr +++ b/math/xspread/pkg-descr @@ -1,15 +1,3 @@ -Xspread is a versatile and robust public domain spreadsheet program -designed for broad accessibility. It offers a flexible user experience, -operating seamlessly in both graphical X Window System environments and -text-based ASCII terminals. - -When running under the X Window System, xspread leverages its graphical -capabilities to provide an intuitive and interactive interface. For users -in command-line or remote environments, it gracefully falls back to -curses and termcap/terminfo for ASCII displays, ensuring functionality -across a wide range of setups. - -As a public domain tool, xspread provides a free and open solution for -data organization, calculation, and analysis. It's an excellent choice -for users seeking a lightweight yet powerful spreadsheet application -that adapts to various computing environments. +The program xspread is a public domain spreadsheet which runs under +X Window system or ascii terminals. Xspread uses the X Window system +if available or curses and term[cap/info] for ascii displays.