compared double and single channel models. further work on prediction…

… plotting.

decoder.py

@@ -116,12 +116,14 @@ def predict_step(self, batch, batch_idx, test_mode=True, dataloader_idx=0):
         if test_mode:
             # x_tot, y_tot: [batch_size, 1, buffer_size], [batch_size, 1, num_groups]
             print(x_tot.shape, y_tot.shape)
-            num_groups = y_tot.shape[2]
-            group_size = x_tot.shape[2] - (num_groups - 1) * self.step
+            num_groups = y_tot.shape[1]
+            group_size = x_tot.shape[1] - (num_groups - 1) * self.step
             y_hat = []
             for i in range(num_groups):
                 start_idx = i * self.step
-                x = x_tot[:, :, start_idx:start_idx + group_size]  # [batch_size, 1, group_size]
+                x = x_tot[:, start_idx:start_idx + group_size, :]
+                x = x.reshape(x.shape[0], x.shape[2], x.shape[1])
+                # [batch_size, group_size, num_channels]
                 y_hat.append(self.model(x).flatten())
             y_hat = torch.unsqueeze(torch.transpose(torch.stack(y_hat), dim0=0, dim1=1),
                                     dim=1)  # [batch_size, 1, num_groups]

main.py

@@ -17,9 +17,9 @@
         # train_file = 'C:\\Users\\aj14\\Desktop\\SMI\\data\\training_max10kHz_30to1kHz_10kshots_dec=256_randampl.h5py'
         # test_file = 'C:\\Users\\aj14\\Desktop\\SMI\\data\\test_max10kHz_30to1kHz_2kshots_dec=256_randampl.h5py'
 
-        train_file = "/Users/nolanpeard/Desktop/SMI_sim/train_single.h5"
-        valid_file = "/Users/nolanpeard/Desktop/SMI_sim/valid_single.h5"
-        test_file = "/Users/nolanpeard/Desktop/SMI_sim/test_single.h5"
+        train_file = "/Users/nolanpeard/Desktop/SMI_sim/train_double.h5"
+        valid_file = "/Users/nolanpeard/Desktop/SMI_sim/valid_double.h5"
+        test_file = "/Users/nolanpeard/Desktop/SMI_sim/test_double.h5"
 
         print('begin main', datetime.datetime.now())
         # step_list = [256]#, 128, 64, 32] # step sizes for rolling input
@@ -30,7 +30,8 @@
 
         runner = train.TrainingRunner(train_file, valid_file, test_file,
                                       step=256)
-        runner.plot_predictions(model_name="CNN", model_id="tdwhpu2l")
+        #runner.plot_predictions(model_name="CNN", model_id="tdwhpu2l")
+        runner.plot_predictions(model_name="CNN", model_id="e8vpuie1")
 
     else:
         print("Error: Unsupported number of command-line arguments")

train.py

@@ -109,15 +109,16 @@ def open_hdf5(self, rolling=True, step=256, group_size=256):
                                                           dim=1), dim=1)
                 # [num_shots * num_groups, 1]
 
-        if num_channels == 1:
-            # self.inputs = torch.unsqueeze(self.inputs, dim=1)
-            # self.targets = torch.unsqueeze(self.targets, dim=1)
-            print(self.inputs.shape, self.targets.shape)
-            self.inputs = torch.reshape(self.inputs, (-1, 1, group_size))
-            self.targets = torch.reshape(self.targets, (-1, 1, 1))
-        else:
-            self.inputs = torch.reshape(self.inputs, (-1, num_channels, group_size))
-            self.targets = torch.reshape(self.targets, (-1, 1, 1))
+        if not self.test_mode:
+            if num_channels == 1:
+                # self.inputs = torch.unsqueeze(self.inputs, dim=1)
+                # self.targets = torch.unsqueeze(self.targets, dim=1)
+                print(self.inputs.shape, self.targets.shape)
+                self.inputs = torch.reshape(self.inputs, (-1, 1, group_size))
+                self.targets = torch.reshape(self.targets, (-1, 1, 1))
+            else:
+                self.inputs = torch.reshape(self.inputs, (-1, num_channels, group_size))
+                self.targets = torch.reshape(self.targets, (-1, 1, 1))
 
         # total number of group_size length sequences = num_shots * num_groups
         # print("open_hdf5 input size", self.inputs.size())  # [self.length, 256]
@@ -296,6 +297,7 @@ def plot_predictions(self, model_name="CNN", model_id="i52c3rlz"):
         y = trainer.predict(model, dataloaders=self.test_loader)
 
         print(y[0][0].numpy().shape)
+        print(y[0][1].numpy().shape)
         print(y[0][2].numpy().shape)
         # y[batch_idx][return_idx], return_idx 0...3: 0: Predictions, 1:
         # Targets, 2: inputs, 3: encoded
@@ -304,12 +306,13 @@ def plot_predictions(self, model_name="CNN", model_id="i52c3rlz"):
         for i in range(len(y[0][0].numpy()[:, 0])):
             fig = plt.figure(figsize=(5, 10))
             ax1, ax2 = fig.subplots(2, 1)
-            for channel in range(y[0][2].numpy().shape[1]):
-                ax1.plot(y[0][2].numpy()[i, channel, :], label="Input"+str(channel))
+            for channel in range(y[0][2].numpy().shape[2]):
+                ax1.plot(y[0][2].numpy()[i, :, channel], label="Input"+str(
+                    channel))
             ax1.set_title("Inputs")
             ax2.legend()
 
-            ax2.plot(y[0][1].numpy()[i, 0, :], label="Targets")
+            ax2.plot(y[0][1].numpy()[i, :], label="Targets")
             ax2.plot(y[0][0].numpy()[i, 0, :], label="Predictions")
             ax2.set_title("MSE Loss: " + str(nn.MSELoss(reduction='sum')
                           (y[0][0][i, :], y[0][1][i, :]).item()))