Changed name to get rid of shadowing

jacobbieker · Jul 5, 2020 · 275c65d · 275c65d
1 parent efca857
commit 275c65d
Showing 1 changed file with 13 additions and 8 deletions.
diff --git a/factnn/generator/pytorch/eventfile_dataloader.py b/factnn/generator/pytorch/eventfile_dataloader.py
@@ -19,6 +19,10 @@ def __init__(self, root, transform=None, pre_transform=None):
         super(EventFileDataset, self).__init__(root, transform, pre_transform)
         self.processed_filenames = []
 
+    @property
+    def raw_file_names(self):
+        return []
+
     @property
     def processed_file_names(self):
         return self.processed_filenames
@@ -33,9 +37,9 @@ def process(self):
             if osp.getsize(raw_path) > 0:
                 # Checks that file is not 0
                 with open(raw_path, "rb") as pickled_event:
-                    data, data_format, features, feature_cluster = pickle.load(pickled_event)
+                    event_data, data_format, features, feature_cluster = pickle.load(pickled_event)
                     # Convert List of List to Point Cloud, then truncation is simply cutting in the z direction
-                    event_photons = data[data_format["Image"]]
+                    event_photons = event_data[data_format["Image"]]
                     event_photons = list_of_lists_to_raw_phs(event_photons)
                     point_cloud = np.asarray(raw_phs_to_point_cloud(event_photons,
                                                                     cx=GEOMETRY.x_angle,
@@ -46,13 +50,13 @@ def process(self):
                         data.event_type = torch.tensor(0, dtype=torch.int8)
                     elif "proton" in raw_path:
                         data.event_type = torch.tensor(1, dtype=torch.int8)
-                    data.energy = torch.tensor(data[data_format["Energy"]], dtype=torch.float)
-                    data.disp = torch.tensor(euclidean_distance(data[data_format['Source_X']], data[data_format['Source_Y']],
-                                                                data[data_format['COG_X']], data[data_format['COG_Y']]),
+                    data.energy = torch.tensor(event_data[data_format["Energy"]], dtype=torch.float)
+                    data.disp = torch.tensor(euclidean_distance(event_data[data_format['Source_X']], event_data[data_format['Source_Y']],
+                                                                event_data[data_format['COG_X']], event_data[data_format['COG_Y']]),
                                              dtype=torch.float16)
-                    data.sign = torch.tensor(true_sign(data[data_format['Source_X']], data[data_format['Source_Y']],
-                                                       data[data_format['COG_X']], data[data_format['COG_Y']],
-                                                       data[data_format['Delta']]), dtype=torch.uint8)
+                    data.sign = torch.tensor(true_sign(event_data[data_format['Source_X']], event_data[data_format['Source_Y']],
+                                                       event_data[data_format['COG_X']], event_data[data_format['COG_Y']],
+                                                       event_data[data_format['Delta']]), dtype=torch.uint8)
 
                     if self.pre_filter is not None and not self.pre_filter(data):
                         continue
@@ -70,3 +74,4 @@ def len(self):
     def get(self, idx):
         data = torch.load(osp.join(self.processed_dir, 'data_{}.pt'.format(idx)))
         return data
+