experimental changes

geo_deep_learning/tasks_with_models/segmentation_dofa.py

@@ -37,9 +37,13 @@ def __init__(self,
         self.num_classes = num_classes
         self.model = DOFASeg(encoder, pretrained, image_size, self.num_classes)
         self.loss = loss
-        self.metric= MulticlassJaccardIndex(num_classes=num_classes, average=None, zero_division=np.nan)
+        self.iou_metric = MulticlassJaccardIndex(num_classes=num_classes,
+                                                  average=None, 
+                                                  zero_division=np.nan
+                                                  )
         self.labels = [str(i) for i in range(num_classes)] if class_labels is None else class_labels
-        self.classwise_metric = ClasswiseWrapper(self.metric, labels=self.labels)
+        self.iou_classwise_metric = ClasswiseWrapper(self.iou_metric, 
+                                                      labels=self.labels)
         self._total_samples_visualized = 0
 
     def forward(self, image: Tensor) -> Tensor:
@@ -77,11 +81,9 @@ def test_step(self, batch, batch_idx):
         y_hat = self(x)
         loss = self.loss(y_hat, y)
         y_hat = y_hat.softmax(dim=1).argmax(dim=1)
-        test_metrics = self.classwise_metric(y_hat, y)
-        test_metrics["loss"] = loss
-        self.log_dict(test_metrics, 
-                      prog_bar=True, logger=True, 
-                      on_step=False, on_epoch=True, sync_dist=True, rank_zero_only=True)
+        self.test_metrics = self.iou_classwise_metric(y_hat, y)
+        self.test_metrics["loss"] = loss
+
         if self._total_samples_visualized < self.max_samples:
             remaining_samples = self.max_samples - self._total_samples_visualized
             num_samples = min(remaining_samples, len(x))
@@ -102,6 +104,9 @@ def test_step(self, batch, batch_idx):
                 self._total_samples_visualized += 1
                 if self._total_samples_visualized >= self.max_samples:
                     break
+    def on_test_epoch_end(self):
+        print(f"test_metrics: {self.test_metrics}")
+        self.log_dict(self.test_metrics)
 
     def on_train_end(self):
         if self.trainer.is_global_zero and self.trainer.checkpoint_callback is not None: