minor updates

configs/downstream/dev.yaml

@@ -28,9 +28,9 @@ model:
   - name: gifflar
     hidden_dim: 1024
     batch_size: 33
-    num_layers: 10
+    num_layers: 5
     epochs: 100
     learning_rate: 0.001
     optimizer: Adam
-    suffix: _1024_12
+    suffix: _1024_5
 

gifflar/data.py

@@ -433,7 +433,8 @@ def process_(self, data, path_idx: int = 0) -> None:
         if self.pre_filter is not None:
             data = [d for d in data if self.pre_filter(d)]
         if self.pre_transform is not None:
-            data = [self.pre_transform(d) for d in data]
+            # data = [self.pre_transform(d) for d in data]
+            data = self.pre_transform(data)
 
         torch.save((data, self.dataset_args), self.processed_paths[path_idx])
 

gifflar/pretransforms.py

@@ -251,13 +251,22 @@ def __call__(self, data):
 
 class TQDMCompose(Compose):
     def forward(self, data: Union[Data, HeteroData]):
-        with tqdm(total=len(self.transforms), desc="Transforms") as t_bar:
-            for transform in self.transforms:
-                if isinstance(data, (list, tuple)):
-                    data = transform(data)
-                else:
-                    data = [transform(d) for d in tqdm(data, total=len(data), desc="Samples", leave=False)]
-                t_bar.update(1)
+        # print(self.transforms)
+        # print(len(self.transforms))
+        # print(len(data))
+        # with tqdm(total=len(self.transforms), desc="Transforms") as t_bar:
+        for transform in tqdm(self.transforms, desc=f"Transform"):
+            if not isinstance(data, (list, tuple)):
+                data = transform(data)
+            else:
+                # data = [transform(d) for d in data]
+                t_data = []
+                for d in tqdm(data, leave=False):
+                    t_data.append(transform(d))
+                data = t_data
+                    # s_bar.update(1)
+                # data = [transform(d) for d in tqdm(data, total=len(data), desc="Samples", leave=False)]
+                # t_bar.update(1)
         return data
 
 

gifflar/utils.py

@@ -116,7 +116,7 @@ def get_metrics(
             Accuracy(**metric_args),
             AUROC(**metric_args),
             MatthewsCorrCoef(**metric_args),
-            Sensitivity(**metric_args),
+            # Sensitivity(**metric_args),
         ])
     return {"train": m.clone(prefix="train/"), "val": m.clone(prefix="val/"), "test": m.clone(prefix="test/")}