pyxai
pyxai copied to clipboard
crillab
PyXAI (Python eXplainable AI) is a Python library (version 3.6 or later) allowing to bring formal explanations suited to (regression or classification) tree-based ML models (Decision Trees, Random For...
There is a problem with boolean condition when I want to import a model of XGBOOSTClassifier I trained a XGBOOSTClassifier with some booleans features (with true and false and not...
 Use tree_specific_reason() instead until the problem is solved. Works for binary classification problems.
The problem occurs when running the predict method of the Learner class: ``` prediction = explainer.predict(instance) ``` In this case, labels predicted by the learner are not the labels themselves...
Calling `explainer.minimal_contrastive_reason(time_limit=5)` raises this error: ``` File ~/Logiciels/pyxai/pyxai/sources/core/explainer/explainerBT.py:363, in ExplainerBT.minimal_contrastive_reason(self, n, time_limit) 361 starting_time = -time.process_time() 362 contrastive_bt = ContrastiveBT() --> 363 c = contrastive_bt.create_model_and_solve(self, None if self._theory is False...
Calling `explainer.tree_specific_reason()` and `explainer.minimal_tree_specific_reason()` on boosted trees raise this error ``` --------------------------------------------------------------------------- AttributeError Traceback (most recent call last) Cell In[34], line 6 4 for instance, prediction in instances: 5 explainer...
Would it be possible to add support for ARM architecture on Linux-based systems?
The classes of sklearn are not the same in the leaves of the model. For DT: Please use the model.target_class to have a list where each position is the sklearn...