Update BART Hawks NBs with new plots #775
Conversation
AlexAndorra
added
the
enhancement
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
|
View / edit / reply to this conversation on ReviewNB aloctavodia commented on 2025-02-17T06:24:23Z Use AlexAndorra commented on 2025-02-17T22:16:44Z Addressed |
|
View / edit / reply to this conversation on ReviewNB aloctavodia commented on 2025-02-17T06:24:24Z definetelly nicer! |
|
View / edit / reply to this conversation on ReviewNB aloctavodia commented on 2025-02-17T06:24:24Z Not for this PR. But this is the kind of plot we should replace with a proper calibration plot (as implemented in ArivZ 1.0!). AlexAndorra commented on 2025-02-17T14:32:19Z Completely agree. I'll edit this NB again once I get my calibration plot into ArviZ (if that's the plot you're talking about) aloctavodia commented on 2025-02-18T07:12:46Z I am talking about calibration plots in general, like yours or others like https://arviz-plots.readthedocs.io/en/latest/gallery/plot_pava_calibration.html AlexAndorra commented on 2025-02-19T01:13:49Z Definitely. Actually, is this one ready for me to add it here? aloctavodia commented on 2025-03-10T15:51:23Z The calibration plot in arviz currently does not support checking the levels. The suggestion is to do it by using a One vs Others approach |
|
Completely agree. I'll edit this NB again once I get my calibration plot into ArviZ (if that's the plot you're talking about) View entire conversation on ReviewNB |
|
Addressed View entire conversation on ReviewNB |
|
Addressed your comments @aloctavodia |
|
I am talking about calibration plots in general, like yours or others like https://arviz-plots.readthedocs.io/en/latest/gallery/plot_pava_calibration.html View entire conversation on ReviewNB |
|
Definitely. Actually, is this one ready for me to add it here? View entire conversation on ReviewNB |
|
Updated with the latest changes on pymc-bart's PR @aloctavodia |
|
The calibration plot in arviz currently does not support checking the levels. The suggestion is to do it by using a One vs Others approach View entire conversation on ReviewNB |
softmax_linktoplot_pdpto visualize the partial probabilitiesplot_scatter_submodelsNB: this needs new features from PyMC BART, introduced here
📚 Documentation preview 📚: https://pymc-examples--775.org.readthedocs.build/en/775/