Merge 0e57379372 into d92490b808

lib/utils.py

@@ -186,7 +186,7 @@ def load_dataset(dataset_dir, batch_size, test_batch_size=None, **kwargs):
     for category in ['train', 'val', 'test']:
         data['x_' + category][..., 0] = scaler.transform(data['x_' + category][..., 0])
         data['y_' + category][..., 0] = scaler.transform(data['y_' + category][..., 0])
-    data['train_loader'] = DataLoader(data['x_train'], data['y_train'], batch_size, shuffle=True)
+    data['train_loader'] = DataLoader(data['x_train'], data['y_train'], batch_size, shuffle=False)
     data['val_loader'] = DataLoader(data['x_val'], data['y_val'], test_batch_size, shuffle=False)
     data['test_loader'] = DataLoader(data['x_test'], data['y_test'], test_batch_size, shuffle=False)
     data['scaler'] = scaler

model/pytorch/dcrnn_supervisor.py

@@ -174,12 +174,23 @@ class DCRNNSupervisor:
 
             self.dcrnn_model = self.dcrnn_model.train()
 
+            
+            # shuffle the batches
             train_iterator = self._data['train_loader'].get_iterator()
+            all_train = np.array([(x,y) for _, (x, y) in enumerate(train_iterator)])
+            permutation = np.random.permutation(all_train.shape[0])
+            all_train = all_train[permutation]
+            
+            
             losses = []
-
             start_time = time.time()
 
-            for _, (x, y) in enumerate(train_iterator):
+            for batch in all_train:
+                x = batch[0]
+                y = batch[1]
+                
+
+                
                 optimizer.zero_grad()
 
                 x, y = self._prepare_data(x, y)