Debugging Validation and testing

lib/models/generators/cnn.py

@@ -1,3 +1,5 @@
+from statistics import mean
+
 from random import choice
 
 import torch
@@ -65,13 +67,12 @@ class CNNRouteGeneratorModel(LightningBaseModule):
 
     def validation_epoch_end(self, outputs):
         evaluation = ROCEvaluation(plot_roc=True)
-        predictions = torch.cat([x['prediction'] for x in outputs])
+        pred_label = torch.cat([x['pred_label'] for x in outputs])
         labels = torch.cat([x['label'] for x in outputs]).unsqueeze(1)
-        losses = torch.cat([x['discriminated_bce_loss'] for x in outputs]).unsqueeze(1)
-        mean_losses = losses.mean()
+        mean_losses = torch.stack([x['discriminated_bce_loss'] for x in outputs]).mean()
 
         # Sci-py call ROC eval call is eval(true_label, prediction)
-        roc_auc, tpr, fpr = evaluation(labels.cpu().numpy(), predictions.cpu().numpy(), )
+        roc_auc, tpr, fpr = evaluation(labels.cpu().numpy(), pred_label.cpu().numpy(), )
         # self.logger.log_metrics(score_dict)
         self.logger.log_image(f'{self.name}_ROC-Curve_E{self.current_epoch}', plt.gcf())
         plt.clf()
@@ -103,7 +104,7 @@ class CNNRouteGeneratorModel(LightningBaseModule):
 
         # Dataset
         self.dataset = TrajData(self.hparams.data_param.map_root, mode='just_route',
-                                length=self.hparams.train_param.batch_size * 1000)
+                                length=self.hparams.data_param.dataset_length)
 
         # Additional Attributes
         self.in_shape = self.dataset.map_shapes_max
@@ -159,6 +160,10 @@ class CNNRouteGeneratorModel(LightningBaseModule):
 
         self.traj_lin = nn.Linear(reduce(mul, self.traj_conv_3.shape), self.feature_dim)
 
+        #
+        # Mixed Encoder
+        self.mixed_lin = nn.Linear(self.lat_dim, self.lat_dim)
+
         #
         # Variational Bottleneck
         self.mu = nn.Linear(self.lat_dim, self.hparams.model_param.lat_dim)
@@ -242,7 +247,7 @@ class CNNRouteGeneratorModel(LightningBaseModule):
         return z, mu, logvar
 
     def generate_random(self, n=6):
-        maps = [self.map_storage[choice(self.map_storage.keys())] for _ in range(n)]
+        maps = [self.map_storage[choice(self.map_storage.keys)] for _ in range(n)]
         trajectories = torch.stack([x.get_random_trajectory() for x in maps] * 2)
         maps = torch.stack([x.as_2d_array for x in maps] * 2)
         labels = torch.as_tensor([0] * n + [1] * n)