Fixing load supervised model

PEP8 over the changes

rosie/chamber_of_deputies/classifiers/reimbursement_generalization.py

@@ -1,6 +1,5 @@
 import os
 import unicodedata
-import tempfile
 import shutil
 from io import BytesIO
 from urllib.request import urlopen
@@ -16,8 +15,7 @@
 from PIL import Image as pil_image
 from sklearn.base import TransformerMixin
 from wand.image import Image
-from PIL import Image as pil_image
-from io import BytesIO
+
 
 class MealGeneralizationClassifier(TransformerMixin):
     """
@@ -44,7 +42,6 @@ class MealGeneralizationClassifier(TransformerMixin):
     # Dimensions of our images.
     img_width, img_height = 300, 300
 
-
     # It defines how many iterations will run to find the best model during traaining
     epochs = 20
     # It influences the speed of your learning (Execution)
@@ -93,6 +90,7 @@ def train(self, train_data_dir, validation_data_dir, save_dir):
         # Inspired by biological visual cortex and tailored for computer vision tasks.
         # Authour: Yann LeCun in early 1990s.
         # See http://deeplearning.net/tutorial/lenet.html for introduction.
+        # Or this simplified version: https://www.youtube.com/watch?v=JiN9p5vWHDY
 
         # This is the augmentation configuration we will use for training
         model.compile(loss='binary_crossentropy',
@@ -124,7 +122,7 @@ def train(self, train_data_dir, validation_data_dir, save_dir):
 
         # It allow us to save only the best model between the iterations
         checkpointer = ModelCheckpoint(
-            filepath=os.path.join(save_dir,"weights.hdf5"),
+            filepath=os.path.join(save_dir, "weights.hdf5"),
             verbose=1, save_best_only=True)
 
         # We set it as a parameter to save only the best model
@@ -138,7 +136,7 @@ def train(self, train_data_dir, validation_data_dir, save_dir):
 
     def fit(self, X):
         # Load an existent Keras model
-        if (not os.path.isfile(X) and  (isinstance(X, str) and ( 'https' in X or 'http' in X))):
+        if (not os.path.isfile(X) and (isinstance(X, str) and ('https' in X or 'http' in X))):
             response = urlopen(X)
             with open('weights.hdf5', 'wb') as fp:
                 shutil.copyfileobj(response, fp)
@@ -174,13 +172,13 @@ def predict(self, X):
                 preds = self.keras_model.predict_classes(x, verbose=0)
                 # Get the probability of prediciton
                 prob = self.keras_model.predict_proba(x, verbose=0)
-                # Keep the predictions with more than 80% of accuracy and the class 1 (suspicious)
+                # Keep the predictions = (suspicious)
                 if(prob >= 0.8 and preds == 1):
                     result.append(True)
                 else:
                     result.append(False)
             else:
-                # Case the reimbursement can not be downloaded or convert it is classified as False
+                # Case the reimbursement can not be convereted to png
                 result.append(False)
 
         self._X['y'] = result
@@ -189,7 +187,8 @@ def predict(self, X):
     def __applicable_rows(self, X):
         return (X['category'] == 'Meal')
 
-    """ Creates a new column 'links' containing an url for the files in the chamber of deputies website
+    """ Creates a new column 'links' containing an url
+        for the files in the chamber of deputies website
             Return updated Dataframe
 
         arguments:

rosie/core/__init__.py

@@ -46,12 +46,12 @@ def __call__(self):
     def load_trained_model(self, classifier):
         filename = '{}.pkl'.format(classifier.__name__.lower())
         path = os.path.join(self.data_path, filename)
-
+        keys = self.settings.SUPERVISED_MODEL
         # palliative: this outputs a model too large for joblib
         if classifier.__name__ == 'MonthlySubquotaLimitClassifier':
             model = classifier()
             model.fit(self.dataset)
-        elif classifier.__name__ in  self.settings.SUPERVISED_MODEL:
+        elif (keys is not None and classifier.__name__ in keys):
             model = classifier()
             model.fit(self.settings.SUPERVISED_MODEL[classifier.__name__])
         else:

rosie/core/tests/test_core_init.py

@@ -117,7 +117,8 @@ def test_load_trained_model_for_meal_generalization(self):
         core = Core(settings, self.adapter)
         core.load_trained_model(ClassifierClass)
 
-        classifier_instance.fit.assert_called_once_with(settings.SUPERVISED_MODEL['MealGeneralizationClassifier'])
+        classifier_instance.fit.assert_called_once_with(
+            settings.SUPERVISED_MODEL['MealGeneralizationClassifier'])
 
     def test_predict(self):
         model = MagicMock()