OWNomogram: Add a new widget

Orange/base.py

@@ -106,6 +106,7 @@ def __call__(self, data):
 
         if isinstance(data, Instance):
             data = Table(data.domain, [data])
+        origdata = data
         data = self.preprocess(data)
 
         if len(data.domain.class_vars) > 1 and not self.supports_multiclass:
@@ -123,6 +124,7 @@ def __call__(self, data):
         model.supports_multiclass = self.supports_multiclass
         model.name = self.name
         model.original_domain = origdomain
+        model.original_data = origdata
         return model
 
     def preprocess(self, data):

Orange/classification/naive_bayes.py

@@ -3,7 +3,7 @@
 from Orange.classification import Learner, Model
 from Orange.data import Instance, Storage
 from Orange.statistics import contingency
-from Orange.preprocess import Discretize
+from Orange.preprocess import Discretize, RemoveNaNColumns
 
 __all__ = ["NaiveBayesLearner"]
 
@@ -19,7 +19,8 @@ class NaiveBayesLearner(Learner):
         An ordered list of preprocessors applied to data before training
         or testing.
     """
-    preprocessors = [Discretize()]
+    preprocessors = [RemoveNaNColumns(), Discretize()]
+    name = 'naive bayes'
 
     def fit_storage(self, table):
         if not isinstance(table, Storage):
@@ -31,33 +32,35 @@ def fit_storage(self, table):
         cont = contingency.get_contingencies(table)
         class_freq = np.array(np.diag(
             contingency.get_contingency(table, table.domain.class_var)))
-        return NaiveBayesModel(cont, class_freq, table.domain)
+        class_prob = (class_freq + 1) / (np.sum(class_freq) + len(class_freq))
+        log_cont_prob = [np.log(
+            (np.array(c) + 1) / (np.sum(np.array(c), axis=0)[None, :] +
+                                 c.shape[0]) / class_prob[:, None])
+                         for c in cont]
+        return NaiveBayesModel(log_cont_prob, class_prob, table.domain)
 
 
 class NaiveBayesModel(Model):
-    def __init__(self, cont, class_freq, domain):
+    def __init__(self, log_cont_prob, class_prob, domain):
         super().__init__(domain)
-        self.cont = cont
-        self.class_freq = class_freq
+        self.log_cont_prob = log_cont_prob
+        self.class_prob = class_prob
 
     def predict_storage(self, data):
         if isinstance(data, Instance):
             data = [data]
-        n_cls = len(self.class_freq)
-        class_prob = (self.class_freq + 1) / (np.sum(self.class_freq) + n_cls)
         if len(data.domain.attributes) == 0:
-            probs = np.tile(class_prob, (len(data), 1))
+            probs = np.tile(self.class_prob, (len(data), 1))
         else:
-            log_cont_prob = [np.log(np.divide(np.array(c) + 1,
-                                              self.class_freq.reshape((n_cls, 1)) +
-                                              c.shape[1])) for c in self.cont]
             probs = np.exp(np.array([np.sum(attr_prob[:, int(attr_val)]
                                             for attr_val, attr_prob
-                                            in zip(ins, log_cont_prob)
+                                            in zip(ins, self.log_cont_prob)
                                             if not np.isnan(attr_val))
-                                     for ins in data]) + np.log(class_prob))
+                                     for ins in data]) + np.log(
+                self.class_prob))
         probs /= probs.sum(axis=1)[:, None]
         values = probs.argmax(axis=1)
         return values, probs
 
+
 NaiveBayesLearner.__returns__ = NaiveBayesModel