erdogant / bnlearn

Python library for learning the graphical structure of Bayesian networks, parameter learning, inference and sampling methods.
https://erdogant.github.io/bnlearn
Other
463 stars 45 forks source link

Verbose message issue #76

Closed paulaten closed 1 year ago

paulaten commented 1 year ago

Hello Erdogant! It's a pleasure to know your work. I have noticed in my latest executions of bnlearn.structure_learning.fit that although I indicate verbose=0, I still get informational messages on the console, specifically: "[bnlearn] >Compute structure scores ['k2', 'bds', 'bic', 'bdeu'] for model comparison (higher is better)."

I have analyzed the code and in the bnlearn/structure_learning.py module, line 194: out['structure_scores'] = bnlearn.structure_scores(out, df) , the verbose information is lost and by default, the call to structure_scores has verbose=3. Would it be possible to fix it by adding out['structure_scores'] = bnlearn.structure_scores(out, df, verbose=config['verbose']) ?

The same thing happens in the bnlearn/parameter_learning.py module, line 155.

Thank you very much. Best regards.

erdogant commented 1 year ago

Thank you for notifying me. I pass the verbose into the function and it should be solved in the next release!

erdogant commented 1 year ago

The update is here! update with: pip install -U bnlearn

paulaten commented 1 year ago

Thank you!

I hope this message finds you well. I wanted to thank you for the latest version of bnlearn library, it has been very useful for my work. However, I noticed that there is still an issue with a line of code that I previously mentioned.

I just ran the latest version (0.7.15) and the message still appears even though I set verbose=0 in bnlearn/structure_learning.fit (line 192). Is there something else that I need to do in order to resolve this issue?

Thank you for your help and for all the hard work you put into maintaining this library.

Best regards.

erdogant commented 1 year ago

Can you update and try again?

paulaten commented 1 year ago

It's fixed! Thank you so much 😊