cleaning code

EvoMSA/back_prop.py

@@ -18,7 +18,6 @@
 from jax.experimental.sparse import BCSR
 from sklearn.model_selection import StratifiedShuffleSplit
 from sklearn.base import clone
-from IngeoML.utils import Batches
 from IngeoML.optimizer import classifier
 from EvoMSA.text_repr import BoW, DenseBoW
 
@@ -57,34 +56,23 @@ def __init__(self, voc_size_exponent: int=15,
                  estimator_kwargs=dict(dual=True, class_weight='balanced'),
                  deviation=None,
                  validation_set=None,
-                 batches=None,
                  optimizer_kwargs: dict=None,
                  **kwargs):
         super(BoWBP, self).__init__(voc_size_exponent=voc_size_exponent,
                                     estimator_kwargs=estimator_kwargs, **kwargs)
         self.deviation = deviation
         self.validation_set = validation_set
         self.optimizer_kwargs = optimizer_kwargs
-        self.batches = batches
 
     @property
     def evolution(self):
         """Evolution of the objective-function value"""
         return self._evolution
-    
+
     @evolution.setter
     def evolution(self, value):
         self._evolution = value
 
-    @property
-    def batches(self):
-        """Instance to create the batches"""
-        return self._batches
-
-    @batches.setter
-    def batches(self, value):
-        self._batches = value
-
     @property
     def optimizer_kwargs(self):
         """Arguments for the optimizer"""
@@ -100,7 +88,7 @@ def optimizer_kwargs(self, value):
     def validation_set(self):
         """Validation set"""
         return self._validation_set
-        
+
     @validation_set.setter
     def validation_set(self, value):
         if value is None or value == 0:
@@ -122,11 +110,11 @@ def validation_set(self, value):
     def deviation(self):
         """Function to measure the deviation between the true observations and the predictions."""
         return self._deviation
-    
+
     @deviation.setter
     def deviation(self, value):
         self._deviation = value
-        
+
     @property
     def parameters(self):
         """Parameter to optimize"""
@@ -170,19 +158,13 @@ def set_validation_set(self, D: List[Union[dict, list]],
     def _combine_optimizer_kwargs(self):
         decoder = self.estimator_instance.classes_
         n_outputs = 1 if decoder.shape[0] == 2 else decoder.shape[0]
-        optimizer_defaults = dict(array=BCSR.from_scipy_sparse,
-                                  every_k_schedule=4, n_outputs=n_outputs,
-                                  epochs=100, learning_rate=1e-4,
-                                  return_evolution=True,
-                                  n_iter_no_change=5)
+        optimizer_defaults = dict(array=BCSR.from_scipy_sparse, n_outputs=n_outputs,
+                                  return_evolution=True)
         optimizer_defaults.update(self.optimizer_kwargs)        
         return optimizer_defaults
 
     def fit(self, D: List[Union[dict, list]], 
             y: Union[np.ndarray, None]=None) -> 'BoWBP':
-        if self.batches is None:
-            self.batches = Batches(size=512 if len(D) >= 2048 else 256,
-                                   random_state=0)
         D, y = self.set_validation_set(D, y=y)
         super(BoWBP, self).fit(D, y=y)
         optimizer_kwargs = self._combine_optimizer_kwargs()
@@ -192,7 +174,7 @@ def fit(self, D: List[Union[dict, list]],
                        texts, labels,
                        deviation=self.deviation,
                        validation=self.validation_set,
-                       batches=self.batches, **optimizer_kwargs)
+                       **optimizer_kwargs)
         if optimizer_kwargs['return_evolution']:
             self.evolution = p[1]
             p = p[0]
@@ -225,14 +207,14 @@ def __init__(self, emoji: bool=True,
     @property
     def model(self):
         return dense_model
-    
+
     def _transform(self, X):
         return self.bow.transform(X)
 
     @property
     def weights(self):
         return np.array([x.coef for x in self.text_representations])
-    
+
     @property
     def bias(self):
         return np.array([x.intercept for x in self.text_representations])