Package: jti 1.0.0

Mads Lindskou

jti: Junction Tree Inference

Minimal and memory efficient implementation of the junction tree algorithm using the Lauritzen-Spiegelhalter scheme; S. L. Lauritzen and D. J. Spiegelhalter (1988) <https://www.jstor.org/stable/2345762?seq=1>. The jti package is part of the paper <doi:10.18637/jss.v111.i02>.

Authors:Mads Lindskou [aut, cre]

jti_1.0.0.tar.gz
jti_1.0.0.tar.gz(r-4.5-noble)jti_1.0.0.tar.gz(r-4.4-noble)
jti_1.0.0.tgz(r-4.4-emscripten)jti_1.0.0.tgz(r-4.3-emscripten)
jti.pdf |jti.html
jti/json (API)
NEWS

# Install 'jti' in R:
install.packages('jti', repos = 'https://cloud.r-project.org')

Bug tracker:https://github.com/mlindsk/jti/issues1 issues

Uses libs:
  • c++– GNU Standard C++ Library v3
Datasets:

On CRAN:

Conda:

cpp

2.70 score 278 downloads 26 exports 14 dependencies

Last updated 4 months agofrom:403983ea60. Checks:3 OK. Indexed: no.

TargetResultLatest binary
Doc / VignettesOKMar 25 2025
R-4.5-linux-x86_64OKMar 25 2025
R-4.4-linux-x86_64OKMar 25 2025

Exports:bnfit_to_cptscompilecpt_listdim_namesget_clique_rootget_clique_root_idxget_cliquesget_graphget_triang_graphhas_inconsistenciesinitializejtjt_leavesjt_nbinary_opsjt_parentsmpdmpepot_listpropagatequery_beliefquery_evidencesend_messagesset_evidencesim_data_from_bnsim_data_from_dmrftriangulate

Dependencies:clicpp11glueigraphlatticelifecyclemagrittrMatrixpkgconfigRcppRcppArmadillorlangspartavctrs

Using jti

Rendered fromusing_jti.Rmdusingknitr::rmarkdownon Mar 25 2025.

Last update: 2024-11-23
Started: 2022-04-12

Citation

To cite jti/sparta in publications use:

Lindskou M, Tvedebrink T, Eriksen PS, Højsgaard S, Morling N (2024). “jti and sparta: Time and Space Efficient Packages for Model-Based Prediction in Large Bayesian Networks.” Journal of Statistical Software, 111(2), 1–24. doi:10.18637/jss.v111.i02.

Corresponding BibTeX entry:

  @Article{,
    title = {{jti} and {sparta}: Time and Space Efficient Packages for
      Model-Based Prediction in Large {B}ayesian Networks},
    author = {Mads Lindskou and Torben Tvedebrink and Poul Svante
      Eriksen and S{\o}ren H{\o}jsgaard and Niels Morling},
    journal = {Journal of Statistical Software},
    year = {2024},
    volume = {111},
    number = {2},
    pages = {1--24},
    doi = {10.18637/jss.v111.i02},
  }

Readme and manuals

Help Manual

Help pageTopics
jti: Junction Tree Inferencejti-package jti
Asiaasia
Asia2asia2
bnfit to cptsbnfit_to_cpts
Compile informationcompile compile.cpt_list
Conditional probability listcpt_list cpt_list.data.frame cpt_list.list
Various gettersdim_names dim_names.charge dim_names.cpt_list dim_names.jt has_inconsistencies has_inconsistencies.charge has_inconsistencies.jt names.charge names.cpt_list names.jt
Return the cliques of a junction treeget_cliques get_cliques.charge get_cliques.jt get_cliques.pot_list get_clique_root get_clique_root.jt get_clique_root_idx get_clique_root_idx.jt
Get graphget_graph get_graph.charge get_graph.cpt_list
Get triangulated graphget_triang_graph
Initializeinitialize initialize.charge
Junction Treejt jt.charge
Query Parents or Leaves in a Junction Treejt_leaves jt_leaves.jt jt_parents jt_parents.jt
Number of Binary Operationsjt_nbinary_ops jt_nbinary_ops.triangulation
Maximal Prime Decompositionmpd mpd.cpt_list mpd.matrix
Most Probable Explanationmpe mpe.jt
A plot method for junction treesplot.charge
A plot method for junction treesplot.jt
A check and extraction of clique potentials from a Markov random field to be used in the junction tree algorithmpot_list pot_list.data.frame
A print method for compiled objectsprint.charge
A print method for cpt listsprint.cpt_list
A print method for junction treesprint.jt
Propagation of junction treespropagate propagate.jt
Query probabilitiesquery_belief query_belief.jt
Query Evidencequery_evidence query_evidence.jt
Send Messages in a Junction Treesend_messages
Enter Evidenceset_evidence set_evidence.charge set_evidence.jt
Simulate data from a Bayesian networksim_data_from_bn
Simulate data from a decomposable discrete markov random fieldsim_data_from_dmrf
Triangulate a Bayesian networktriangulate triangulate.cpt_list