bnlearn

• New functionalities in creating plots for both interactive and static plots.
  • Change node colors
  • Change node sizes
  • Adjust plot settings such as fonts, font-colors, background color etc.
  • https://erdogant.github.io/bnlearn/pages/html/Plot.html

• Added licence to setup

• Added functionality to save and load models!
• Examples: https://erdogant.github.io/bnlearn/pages/html/saving%20and%20loading.html

update to the latest version with:
pip install -U bnlearn

• added interactive plot with pyvis using the plot functionality with parameter bn.plot(model, interactive=True)

import bnlearn as bn
df = bn.import_example()
# Structure learning
model = bn.structure_learning.fit(df)
bn.plot(model, interactive=True)

• Included fixed_edges when using methodtype='hc'
• Changed the bw_list_method for clarity.
  • bw_list_method='enforce' into bw_list_method='edges'
  • bw_list_method='filter' into bw_list_method='nodes'
• Some docstring updates
• Some code cleaning

Note that bw_list_method='nodes' works for all methodtypes and bw_list_method='edges' only when methodtype='hc'

• Added new tree search based structure learning method: Tree-augmented Naive Bayes (TAN)

Example

import bnlearn as bn
df = bn.import_example()
# Structure learning with TAN
model = bn.structure_learning.fit(df, methodtype='tan', root_node='Cloudy', class_node='Rain')
bn.plot(model)

• Fix for scoring_method that occured in newer versions from pgmpy/python
• Fix for return_type in forward_sampling
• removed examples from library and included functionality to auto-download
• Update of requirements
• Update unit test
• Some code refactoring

• Functionally added to make predictions on dataset with the predict function #17
• Predict examples can be found in the readme and sphinx pages: https://erdogant.github.io/bnlearn/pages/html/Predict.html
• Functionally added that outputs a dataframe with the results for the inference that is made (query2df)
• Included some exceptions to catch errors
• Fixed circular imports
• Unit test updated for the new functions
• Some docstrings updated
• Some code refactoring and cleaning

• New functionality added: topological sorting of the DAG.
• Example of usage: https://erdogant.github.io/bnlearn/pages/html/topological_sort.html

• Fix for the maximum likelihood estimation in parameter learning using the model.fit approach (thanks to ssomanch)

• Fix typo in pip install instructions
  Thanks to the contribution of bencevans!

• tabu_length is configurable in hill climb search (Thanks to contribution of ivylee!)

• the use of black_list and white_list requires setting parameter method type
• update unit test
• update sphinx docs

• Better variable naming in "sampling" function.
• Fix for minimum requirement matplotlib

• Some fixes to install latest packages

• Fixes to stay compatible with latest version of pgmpy (0.1.13)
• Added functionality to perform structure learning using method of Chow-Liu
• Random code cleaning, pep styling

• Fixed output to console using the verbosity. If it is set to 0, nothing should be printed.

• check for CPDs validity.

• fix for plot as it did not work anymore because of update in networkx.
• fix for printing CPDs. This was not always possible.
• verbosity messages capitalized chars are lowered
• Check for networkx version, should be >2.5
• docstring updates
• improvements in code styling

• added to hc: white_list, black_list enforcing or filtering possibility
• added to hc: epsilon parameter
• added to hc: max_iter parameter