Removed option to choose training domain.

ANN_Data_Generator.py

@@ -33,22 +33,13 @@ class TrainingDataGenerator:
         Builds random training data.
 
     """
-    def __init__(self, left_bound=-1, right_bound=1):
-        """Initializes TrainingDataGenerator.
-
-        Parameters
-        ----------
-        left_bound : float, optional
-            Left boundary of interval. Default: -1.
-        right_bound : float, optional
-            Right boundary of interval. Default: 1.
-
-        """
+    def __init__(self):
+        """Initializes TrainingDataGenerator."""
         self._basis_list = [OrthonormalLegendre(pol_deg)
                             for pol_deg in range(7)]
         self._quadrature_list = [Gauss({'num_nodes': pol_deg+1})
                                  for pol_deg in range(7)]
-        self._mesh_list = [Mesh(left_bound=left_bound, right_bound=right_bound,
+        self._mesh_list = [Mesh(left_bound=-1, right_bound=1,
                                 num_ghost_cells=0, num_cells=2**exp)
                            for exp in range(5, 12)]
 

config.yaml

-data_dir: 'Oct03'
+data_dir: 'Oct04'
 random_seed: 1234
 
 # Parameter for Approximation with Troubled Cell Detection
@@ -49,9 +49,6 @@ Approximation:
 ANN_Data:
   sample_number: 100
 
-  left_boundary: -1
-  right_boundary: 1
-
   smooth_troubled_balance: 0.5
 
   stencil_length: 3

workflows/ANN_data.smk

@@ -17,8 +17,6 @@ rule generate_data:
         protected(DIR+'/output_data.npy')
     default_target: True
     params:
-        left_bound = config['left_boundary'],
-        right_bound = config['right_boundary'],
         balance = config['smooth_troubled_balance'],
         stencil_len = config['stencil_length'],
         sample_number = config['sample_number'],
@@ -37,8 +35,7 @@ rule generate_data:
 
         with open(str(log), 'w') as logfile:
             sys.stdout = logfile
-            generator = ANN_Data_Generator.TrainingDataGenerator(
-                left_bound=params.left_bound, right_bound=params.right_bound)
+            generator = ANN_Data_Generator.TrainingDataGenerator()
             data = generator.build_training_data(balance=params.balance,
                 init_cond_list=init_cond_list, directory=DIR,
                 num_samples=params.sample_number,