Changes in version 0.8.1 - Allow plotting a subset of nodes in static networks - Allow not using hierarchical layout while plotting static networks Changes in version 0.8.0 (2026-01-13) - Package published on the Journal of Statistical Software, added a CITATION file - Added new DOI to the DESCRIPTION Changes in version 0.7.9 (2024-06-19) - Fixed unstable behaviour of regex in C++ on both the psoho and natPsoho algorithms. - Added a whitelist_tr argument to the DMMHC algorithm to allow forcing arcs in the transition network. - Fixed the 'forecast_ts' function so that when forecasting future values not present in the dataset having 'plot_res = T' doesn't throw an error - Removed C++11 specification Changes in version 0.7.8 (2022-10-04) - Bringing the package overall status up to scratch for publication on a journal. - Fixed incompatibility with bnlearn "mle-g" parameter. - Generic methods have been added for the "dbn" and "dbn.fit" S3 classes. - Documentation of the exported functions is now more accurate and follows a more stricter standard. - Non-exported functions are now excluded from the man pages. - Regular 'print' in low-level, internal functions has been swapped with 'cat'. - All of bnlearn methods for the classes "bn" and "bn.fit" have been extended to the "dbn" and "dbn.fit" classes in independent functions to ensure compatibility. All these methods are contained in the "bnlearn_compat.R" file. - Generalized matrix calculation now done with MASS::ginv Changes in version 0.7.7 - New auxiliary function that moves the values of the variables backwards in time. Useful to shift the last row of evidence in a dataset and predict in a real world scenario. Changes in version 0.7.6 - Added an option to reverse the naming convention of the nodes when plotting a DBN. It transforms t_0 into t_n, t_1 into t_n-1 and so on. In the literature, t_0 is the oldest time-slice, and we reversed that in dbnR. This option lets users plot the networks following convention, but changes nothing of the models underneath. - Fixed interaction between 'sapply' and some time series in the 'plot_single_result' function that caused the 'lines' function to take a very long time to execute. - Fixed plots not showing the predictions line when the predictions where outside the range of the original time series values. Changes in version 0.7.5 (2022-03-14) - Deprecated the 'size' argument in 'forecast_ts' and 'smooth_ts'. - A named vector is no longer the only option required for the 'mvn_inference' function. It now also takes a single row data.table and makes the conversion to named vector inside. This avoids the user needing to have an external auxiliary function. - New security check to confirm that a data.table has only one row. This is also needed in the inference, because we can only provide one row of evidence. It gave a non-descriptive error before, now it warns about using more than one row in the evidence data.table. Changes in version 0.7.4 - Added the possibility of plotting only a subset of the nodes with the visualization tool. - Unified the 'plot_network' function for visualizing BNs and the 'plot_dynamic_network' function for visualizing DBNs into a single one. - More informative error messages in security_checks.R when dealing with ellipsis. Changes in version 0.7.3 - Added the possibility of passing a blacklist_tr parameter to the dmmhc algorithm in order to avoid certain inter-slice arcs. Changes in version 0.7.2 - Added authors and doi of the structure learning algorithms to the description field as per request of CRAN maintainers. Changes in version 0.7.1 (2021-09-29) - natPsoho algorithm fixed and working - Fixed check results - Removed unused rand import in C++ Changes in version 0.7.0 - Added new order invariant particle swarm structure learning algorithm: natPSOHO Changes in version 0.6.2 - Fixed the create_blacklist method for sizes greater than 10. A regex "$" was missing. Changes in version 0.6.1 - Added variable inertia, global best and local best parameters over time to the PSOHO algorithm - Now filtered_fold_dt() allows a boolean argument to avoid deleting the id_var column before returning the folded dataset Changes in version 0.6.0 - Added the possibility of doing smoothing over a time series - Fixed a bug that returned NaNs when predicting 1 variable in t_0 and providing all other variables as evidence Changes in version 0.5.7 - Parameter for different bnlearn scores in the PSOHO algorithm - New Jupyter notebook that shows how to integrate dbnR with Python - Fixed the 'intra' argument in the mmhc algorithm. The 'if' statements regarding it were wrong. - Several fixes in compatibility with 'as.data.table' calls Changes in version 0.5.6 - Hotfix for the 'f_dt' parameter in the 'learn_dbn_struc' function. It was not being passed down to each algorithm. - New auxiliary function 'filtered_fold_dt' - Added the possibility of not learning intra-slice arcs in the dmmhc algorithm Changes in version 0.5.5 - Removed obsolete and time consuming security checks. The psoho initial checks were supposed to secure the separate usage of the 'Position' and 'Velocity' classes, but they are not exported, and consequently they cannot be misused in the sense that the checks prevented. They also were O(n) on the size of the ordering, and were run each time some operation was performed over positions or velocities. Changes in version 0.5.4 - Additional security checks for the PSOHO algorithm - Added an auxiliary function to reduce the frequency in a data.table by performing the mean of consecutive rows - Added an auxiliary function to fold a dataset and not allow instances from different labeled series to be in the same row - Added the possibility to introduce a previously folded dataset to the structure learning. This helps when the dataset has several iterations of the same process separated by id. If you fold the dataset directly, data from 2 or more different processes will appear in some rows, giving the impression to the model that the ending and the beginning of the processes merge into each other, which most likely is not true Changes in version 0.5.3 (2020-10-13) - Fixed the lack of imports from the 'R6' package with @importFrom, NOTE in CRAN checks results Changes in version 0.5.2 (2020-10-08) - Fixed the visualization of dynamic networks with no intra-slice arcs. The nodes were plotted in a horizontal line, now they are shown in vertical lines in each time slice Changes in version 0.5.1 - More extensive examples in markdowns/usage_example.Rmd - Fixed the test dataset - New learning algorithm present in the docs and readme Changes in version 0.5.0 - Refractored the structure learning to accommodate multiple algorithms - Added a new particle swarm optimization structure learning algorithm (psoho) Changes in version 0.4.6 - Dropped std:vector in favor of Rcpp::Numeric/StringVector in gauss_transform.cpp - Fixed a security check typo Changes in version 0.4.5 (2020-06-11) - Fixed .Rbuildignore markdown and media folders Changes in version 0.4.4 - Bug fixes in the evidence providing - Typo fix in the "approx" mode of forecasting Changes in version 0.4.0 - Added the possibility to provide evidence in each forecasting step Changes in version 0.3.4 (2020-03-25) - Fixed the dependency on R (>= 3.5.0) rather than R (>= 3.6.0). CRAN checks for old released R versions failed because of this. The only difference is that running the package with R (< 3.6.0) displays the plot_dynamic_network() with a different color palette. Changes in version 0.3.3 (2020-03-20) - Fixed \dontrun{} into \donttest{} and cat() into print(x, quote=FALSE) Changes in version 0.3.2 - Redid and added new examples to exported functions - Fixing spell checking and successful builds - Added NEWS.md file. Should have been added earlier.