I downloaded the project and opened in my visual studio where I am unable to find which python file to be run first in order to avoid the unnecessary import errors. In readme file I can see the code which I couldn't find in the project which I downloaded.

boston housing prices in ReadMe has replaced by California housing Price. But in readme document, with explaining tree model, the sample also uses boston housing price. I think it may be confused for some greeners when use it.

I am using something like this on a catboost model, trying to plot a single decision plot: explainer = shap.TreeExplainer(catboost_model) df = pandas.DataFrame(columns=feature_names, data=[feature_values]) shap_values = explainer.shap_values(df[model_input_feature_names]) shap.decision_plot(...

Hi, I was looking at SHAP interaction values, and this paper: Consistent Individualized Feature Attribution for Tree Ensembles I was hoping someone could help me understand section 4 on interaction values better. It says these can be interpreted as "the difference between the SHAP values for featur...

shap.utils._exceptions.ExplainerError: Additivity check failed in TreeExplainer! Please ensure the data matrix you passed to the explainer is the same shape that the model was trained on. If your data shape is correct then please report this on GitHub. Consider retrying with the feature_perturbation...

Hi, since numpy 1.24 some dtype aliases have been removed. Therefore, in the code np.bool and np.int have to be replaced by np.bool_ and np.int_, resp. Regards, Jörg.

I have all coalitions values already calculated. How can I use SHAP to calculate Shapley values for all variables (players in coalition game)? shap_values() is deprecated.

plt.bar() raises IndexError when number of cohorts is larger than 10. This is due to number of patterns available to hatch is only ten. below is a simple example to reproduce the error. cohorts = list(map(str, shap_values[:,"Sex"].data)) # two cohorts shap.plots.bar(shap_values.cohorts(cohorts)) #...

I noticed this issue originally when trying to install mlflow on python 3.11 which depends on shap (see mlflow/mlflow#7681). The reason that shap fails to install is that it depends on numba which in turn requires llvmlite which does not work with python 3.11 yet. Namba 0.56.4 also requires numpy &l...

Hello, I have a multi-class category model made using MLJAR ("perform"). When I try to understand, i get the following: --------------------------------------------------------------------------- TypeError Traceback (most recent call last) Cell In[8], line 2 1 #...