Fix ImbalancedSampler on sliced InMemoryDataset (#6374)

CHANGELOG.md

@@ -45,6 +45,7 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Added support for dropping nodes in `utils.to_dense_batch` in case `max_num_nodes` is smaller than the number of nodes  ([#6124](https://github.com/pyg-team/pytorch_geometric/pull/6124))
 - Added the RandLA-Net architecture as an example ([#5117](https://github.com/pyg-team/pytorch_geometric/pull/5117))
 ### Changed
+- Fix `ImbalancedSampler` on sliced `InMemoryDataset` ([#6374](https://github.com/pyg-team/pytorch_geometric/pull/6374))
 - Breaking Change: Changed the interface and implementation of `GraphMultisetTransformer` ([#6343](https://github.com/pyg-team/pytorch_geometric/pull/6343))
 - Fixed the approximate PPR variant in `transforms.GDC` to not crash on graphs with isolated nodes ([#6242](https://github.com/pyg-team/pytorch_geometric/pull/6242))
 - Added a warning when accesing `InMemoryDataset.data` ([#6318](https://github.com/pyg-team/pytorch_geometric/pull/6318))

torch_geometric/loader/imbalanced_sampler.py

@@ -78,7 +78,7 @@ def __init__(
             y = y[input_nodes] if input_nodes is not None else y
 
         elif isinstance(dataset, InMemoryDataset):
-            y = dataset.data.y.view(-1)
+            y = dataset.y.view(-1)
             assert len(dataset) == y.numel()
 
         else: