PolyMath
PolyMath copied to clipboard
PolyMathOrg
Scientific Computing with Pharo
_From @bergel on March 4, 2016 11:40_ It is a nonparametric statistical test that operates over two samples of the same size. https://en.wikipedia.org/wiki/Mann%E2%80%93Whitney_U_test _Copied from original issue: SergeStinckwich/SciSmalltalk#26_
Topological skeleton is very useful image representation. There was commercial children application Plopp which was build on Squeak and probably it used this approach. Links to the topic: https://en.wikipedia.org/wiki/Straight_skeleton, http://www.machinelearning.ru/wiki/images/f/f8/Gr_98_Mest.pdf...
From mailing-list: "I’d enjoy having Polymath data structures polymorphic with the Pharo ones. Because once I was working with Polymath and it was really complicated to always convert between the...
Section 1.6 of the NumAnalysis pharo book needs to be revised for Pharo and optimization opportunities need to be explored. Section 1.6 is on page 17 here: http://files.pharo.org/books-pdfs/numerical-methods/2016-04-NumericalMethods.pdf and for...
it would eventually make more sense to implement these methods in package numbersextensions on a lower level collection, have a look at eg Collection *Collections-arithmetic & *Collections-arithmetic-collectors. (and of course...
from http://forum.world.st/first-impressions-td4939166.html message by Serge at Mar 19, 2017; 3:57pm
This continues the simplification of the QR Decomposition method object. I have made the idea of the Householder Reflection explicit. I will continue to look more deeply into why the...
The Math-Complex package contains both [PMComplexNumber.class.st](https://github.com/PolyMathOrg/PolyMath/blob/master/src/Math-Complex/PMComplexNumber.class.st) and [PMComplexNumber.extension.st](https://github.com/PolyMathOrg/PolyMath/blob/master/src/Math-Complex/PMComplexNumber.extension.st). The methods from the extension file show up in Pharo 9.0 as "as yet unclassified". In other Smalltalk implementations, the extension methods...
PMProbabilityDensity>>#privateInverseDistributionValue: is using PMCDFNewtonZeroFinder and this cause a problem of dependency. But while discussing with Oleks we think that maybe this should not use newton method at all. I don't...
To me it makes more sense for Math-Numeric to depend on Math-Distributions. But for now we have a cyclic dependency. Dependencies to Math-Numeric in Math-Distributions: - PMRombergIntegrator>>#computeInitialValues ==> Usage of...