3fa25407f34d5c5afa91a36212001eeb520d5838,mindsdb/libs/ml_models/probabilistic_validator.py,ProbabilisticValidator,evaluate_prediction_accuracy,#ProbabilisticValidator#Any#Any#,55

Before Change

        predicted_value_b = self._get_value_bucket(predicted_value, None)

        if self._use_features:
            features_b = list(map(lambda x: self._get_value_bucket(x, None), features))
            features_b.append(predicted_value_b)
            
            // Note: probabilities are rturned for the sorted vector of classes
            So, in thise case, since True(1) > False(0) the second value in the vecotr

After Change

    def evaluate_prediction_accuracy(self, features, predicted_value):
        predicted_value_b = self._get_value_bucket(predicted_value, None)

        feature_existence = list(map(self._features_missing,features))

        X = feature_existence
        X.append(predicted_value_b)
        print(X)

Instances