Add interval to compose.py

Following to [PR5733](pyg-team#7533)  I've Added the interval argument. @rusty1s With these modifications, both the Compose and ComposeFilters classes accept an interval argument, and it will be passed to the transformed data when composing transforms or applying filters.

torch_geometric/transforms/compose.py

@@ -1,22 +1,27 @@
-from typing import Callable, List, Union
+from typing import Callable, List, Union, Tuple
 
 from torch_geometric.data import Data, HeteroData
 from torch_geometric.transforms import BaseTransform
 
-
 class Compose(BaseTransform):
     r"""Composes several transforms together.
 
     Args:
         transforms (List[Callable]): List of transforms to compose.
+        interval (Tuple[float, float], optional): A tuple representing the
+            interval for the transformation. Defaults to (0.0, 1.0).
     """
-    def __init__(self, transforms: List[Callable]):
+    def __init__(self, transforms: List[Callable], interval: Tuple[float, float] = (0.0, 1.0)):
         self.transforms = transforms
+        self.interval = interval
 
     def forward(
         self,
         data: Union[Data, HeteroData],
     ) -> Union[Data, HeteroData]:
+        # Pass the interval argument to the transformed data
+        data.interval = self.interval
+
         for transform in self.transforms:
             if isinstance(data, (list, tuple)):
                 data = [transform(d) for d in data]
@@ -34,14 +39,20 @@ class ComposeFilters:
 
     Args:
         filters (List[Callable]): List of filters to compose.
+        interval (Tuple[float, float], optional): A tuple representing the
+            interval for the transformation. Defaults to (0.0, 1.0).
     """
-    def __init__(self, filters: List[Callable]):
+    def __init__(self, filters: List[Callable], interval: Tuple[float, float] = (0.0, 1.0)):
         self.filters = filters
+        self.interval = interval
 
     def __call__(
         self,
         data: Union[Data, HeteroData],
     ) -> bool:
+        # Pass the interval argument to the transformed data
+        data.interval = self.interval
+
         for filter_fn in self.filters:
             if isinstance(data, (list, tuple)):
                 if not all([filter_fn(d) for d in data]):