aGrUM/pyAgrum 0.16.2 released

Posted on sam. 21 septembre 2019 in News

ANNOUNCE: aGrUM 0.16.2

aGrUM/pyAgrum 0.16.2 is out.

Changelog for 0.16.2

  • pyAgrum
    • remove the use of 'f-strings' in pyAgrum.lib.notebook.py

Changelog for 0.16.1

  • aGrUM

    • improve the syntax for BN specification using gum::fastPrototype
    • improve several CMakeFiles.txt and doxygen documentation
    • add CI for python 2.7
    • refresh gum::BayesNet::toString()
    • API change : gum::MarkovBlanket()::{mb()->gum::MarkovBlanket::dag()}
  • pyAgrum

    • pyAgrum.config object for customization (see notebook 08-configForPyAgrum)
    • improving the syntax for BN specification using pyAgrum.fastBN
    • improving pyAgrum's documentation
    • add pyAgrum.causal.counterfactual and pyAgrum.causal.counterfactualModel (see notebook 55-Causality_Counterfactual)
    • improve plots for histograms (particularly : now svg image by default)
    • add pyAgrum.Potential.fullWithFunction()
    • add pyAgrum.{any graph-like}.connectedComponents()
    • add pyAgrum.BayesNet.parents() and pyAgrum.BayesNet.children() with the name of node.
    • pyAgrum.MarkovBlanket(bn,var,level) build now a level-th order Markov Blanket (by default level=1)
    • add access to constants pyAgrum.VarType_{Discretized|Labelized|Range} when testing pyAgrmu.{any discrete variable}.varType()
    • API change : pyAgrum.MarkovBlanket().{mb()->dag()}
    • add pyAgrum.lib.bn_vs_bn.GraphicalBNComparator.hamming()