'AnchorTabularExplainer' object has no attribute 'validation'
ArpitSisodia opened this issue · comments
exp = explainer.explain_instance(numpy_test, catBoostModel.predict, threshold=0.95).
/opt/anaconda3/lib/python3.7/site-packages/anchor/anchor_tabular.py in sample_from_train(self, conditions_eq, conditions_neq, conditions_geq, conditions_leq, num_samples, validation)
98 bla
99 """
--> 100 train = self.train if not validation else self.validation
101 d_train = self.d_train if not validation else self.d_validation
102 idx = np.random.choice(range(train.shape[0]), num_samples,
AttributeError: 'AnchorTabularExplainer' object has no attribute 'validation'
Experiencing the same issue.
Can you check your package version? Those lines of code should not be in there in 0.0.1.0, which is what is on pypi and in the master of this repo.
When I run pip install anchor_exp, v0.0.0.6 is installed. After running pip install --upgrade anchor_exp I did get v0.0.1.0.
However with this new version I am now getting a new error which I was not getting before for the below code:
explainer = anchor_tabular.AnchorTabularExplainer(['Bad', 'Good'], dfTrain.columns, dfTrain)
The above issue was happening since my data was a DataFrame. After changing it to a numpy array, all worked correctly. Thanks.