KNN lifting (Pointcloud to Graph) with New dataset MNIST

.github/workflows/test_codebase.yml

@@ -3,23 +3,22 @@ name: "Testing Codebase"
 on:
   workflow_dispatch:
   push:
-    branches: [main,github-actions-test]
+    branches: [main, github-actions-test]
     paths-ignore:
-    - 'docs/**'
-    - 'README.md'
-    - 'LICENSE.txt'
-    - '.gitignore'
+      - "docs/**"
+      - "README.md"
+      - "LICENSE.txt"
+      - ".gitignore"
 
   pull_request:
     branches: [main]
     paths-ignore:
-    - 'docs/**'
-    - 'README.md'
-    - 'LICENSE.txt'
-    - '.gitignore'
+      - "docs/**"
+      - "README.md"
+      - "LICENSE.txt"
+      - ".gitignore"
 
 jobs:
-
   pytest:
     runs-on: ${{ matrix.os }}
 
@@ -39,12 +38,12 @@ jobs:
         with:
           python-version: ${{ matrix.python-version }}
           cache: "pip"
-          cache-dependency-path: '**/pyproject.toml'
+          cache-dependency-path: "**/pyproject.toml"
 
       - name: Install PyTorch ${{ matrix.torch-version }}+cpu
         run: |
           pip install --upgrade pip setuptools wheel
-          pip install torch==${{ matrix.torch-version}} --extra-index-url https://download.pytorch.org/whl/cpu
+          pip install torch==${{ matrix.torch-version}} torchvision --extra-index-url https://download.pytorch.org/whl/cpu
           pip install torch-scatter -f https://data.pyg.org/whl/torch-${{ matrix.torch-version }}+cpu.html
           pip install torch-sparse -f https://data.pyg.org/whl/torch-${{ matrix.torch-version }}+cpu.html
           pip install torch-cluster -f https://data.pyg.org/whl/torch-${{ matrix.torch-version }}+cpu.html
@@ -60,4 +59,4 @@ jobs:
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
           file: coverage.xml
-          fail_ci_if_error: false
+          fail_ci_if_error: false

.github/workflows/test_tutorials.yml

@@ -3,20 +3,20 @@ name: "Testing Tutorials"
 on:
   workflow_dispatch:
   push:
-    branches: [main,github-actions-test]
+    branches: [main, github-actions-test]
     paths-ignore:
-    - 'docs/**'
-    - 'README.md'
-    - 'LICENSE.txt'
-    - '.gitignore'
+      - "docs/**"
+      - "README.md"
+      - "LICENSE.txt"
+      - ".gitignore"
 
   pull_request:
     branches: [main]
     paths-ignore:
-    - 'docs/**'
-    - 'README.md'
-    - 'LICENSE.txt'
-    - '.gitignore'
+      - "docs/**"
+      - "README.md"
+      - "LICENSE.txt"
+      - ".gitignore"
 
 # Disable debugger's warnings from nbconvert in test_tutorials.py
 env:
@@ -47,7 +47,7 @@ jobs:
       - name: Install PyTorch ${{ matrix.torch-version }}+cpu
         run: |
           pip install --upgrade pip setuptools wheel
-          pip install torch==${{ matrix.torch-version}} --extra-index-url https://download.pytorch.org/whl/cpu
+          pip install torch==${{ matrix.torch-version}} torchvision --extra-index-url https://download.pytorch.org/whl/cpu
           pip install torch-scatter -f https://data.pyg.org/whl/torch-${{ matrix.torch-version }}+cpu.html
           pip install torch-sparse -f https://data.pyg.org/whl/torch-${{ matrix.torch-version }}+cpu.html
           pip install torch-cluster -f https://data.pyg.org/whl/torch-${{ matrix.torch-version }}+cpu.html

README.md

@@ -52,7 +52,7 @@ Then:
 4. Install torch, torch-scatter, torch-sparse with or without CUDA depending on your needs.
 
       ```bash
-      pip install torch==2.0.1 --extra-index-url https://download.pytorch.org/whl/${CUDA}
+      pip install torch==2.0.1 torchvision --extra-index-url https://download.pytorch.org/whl/${CUDA}
       pip install torch-scatter torch-sparse -f https://data.pyg.org/whl/torch-2.0.1+${CUDA}.html
       pip install torch-cluster -f https://data.pyg.org/whl/torch-2.0.0+${CUDA}.html
       ```

configs/datasets/MNIST.yaml

@@ -0,0 +1,13 @@
+data_domain: point_cloud
+data_type: MNIST
+data_name: MNIST
+data_dir: datasets/${data_domain}/${data_type}
+#data_split_dir: ${oc.env:PROJECT_ROOT}/datasets/data_splits/${data_name}
+
+# Dataset parameters
+num_features: 1
+num_classes: 10
+task: classification
+loss_type: cross_entropy
+monitor_metric: accuracy
+task_level: graph

configs/models/graph/gcn.yaml

@@ -0,0 +1,4 @@
+in_channels: null # This will be set by the dataset
+hidden_channels: 32
+out_channels: null # This will be set by the dataset
+n_layers: 2

configs/transforms/liftings/pointcloud2graph/knn_lifting.yaml

@@ -0,0 +1,5 @@
+transform_type: "lifting"
+transform_name: "GraphKNNLifting"
+k_value: 5
+loop: True
+feature_lifting: ProjectionSum

modules/data/load/loaders.py

@@ -13,6 +13,7 @@
     load_hypergraph_pickle_dataset,
     load_manual_graph,
     load_simplicial_dataset,
+    mnist_to_pointcloud,
 )
 
 
@@ -108,6 +109,12 @@ def load(self) -> torch_geometric.data.Dataset:
             data = load_manual_graph()
             dataset = CustomDataset([data], self.data_dir)
 
+        elif self.parameters.data_name in ["MNIST"]:
+            data = mnist_to_pointcloud(self.parameters)
+            dataset = CustomDataset(
+                data, self.data_dir, data_name=self.parameters.data_name
+            )
+
         else:
             raise NotImplementedError(
                 f"Dataset {self.parameters.data_name} not implemented"

modules/data/utils/concat2geometric_dataset.py

@@ -2,14 +2,31 @@
 
 
 class ConcatToGeometricDataset(Dataset):
+    r"""Concatenates list of PyTorch Geometric Data objects to form a dataset."""
+
     def __init__(self, concat_dataset):
+        r"""Initializes the dataset.
+
+        Parameters
+        ----------
+        concat_dataset : list
+            List of PyTorch Geometric Data objects to be concatenated.
+        """
         super().__init__()
         self.concat_dataset = concat_dataset
 
     def len(self):
+        r"""Returns the length of the dataset."""
         return len(self.concat_dataset)
 
     def get(self, idx):
+        r"""Returns the PyTorch Geometric Data object at the specified index.
+
+        Parameters
+        ----------
+        idx : int
+            Index of the data object to be returned.
+        """
         data = self.concat_dataset[idx]
 
         x = data.x.float()

modules/data/utils/custom_dataset.py

@@ -2,8 +2,9 @@
 
 
 class CustomDataset(torch_geometric.data.InMemoryDataset):
-    def __init__(self, data_list, data_dir, transform=None):
+    def __init__(self, data_list, data_dir, data_name="Custom", transform=None):
         self.data_list = data_list
+        self.data_name = data_name
         super().__init__(data_dir, transform)
         self.load(self.processed_paths[0])
 
@@ -13,3 +14,6 @@ def processed_file_names(self):
 
     def process(self):
         self.save(self.data_list, self.processed_paths[0])
+
+    def __repr__(self):
+        return f"{self.data_name}({len(self)})"

modules/data/utils/utils.py

@@ -50,16 +50,16 @@ def get_complex_connectivity(complex, max_rank, signed=False):
                 )
             except ValueError:  # noqa: PERF203
                 if connectivity_info == "incidence":
-                    connectivity[f"{connectivity_info}_{rank_idx}"] = (
-                        generate_zero_sparse_connectivity(
-                            m=practical_shape[rank_idx - 1], n=practical_shape[rank_idx]
-                        )
+                    connectivity[
+                        f"{connectivity_info}_{rank_idx}"
+                    ] = generate_zero_sparse_connectivity(
+                        m=practical_shape[rank_idx - 1], n=practical_shape[rank_idx]
                     )
                 else:
-                    connectivity[f"{connectivity_info}_{rank_idx}"] = (
-                        generate_zero_sparse_connectivity(
-                            m=practical_shape[rank_idx], n=practical_shape[rank_idx]
-                        )
+                    connectivity[
+                        f"{connectivity_info}_{rank_idx}"
+                    ] = generate_zero_sparse_connectivity(
+                        m=practical_shape[rank_idx], n=practical_shape[rank_idx]
                     )
     connectivity["shape"] = practical_shape
     return connectivity
@@ -421,3 +421,38 @@ def make_hash(o):
     hash_as_hex = sha1.hexdigest()
     # Convert the hex back to int and restrict it to the relevant int range
     return int(hash_as_hex, 16) % 4294967295
+
+
+def mnist_to_pointcloud(cfg, threshold=0.0):
+    r"""Convert MNIST dataset into point cloud.
+
+    Parameters
+    ----------
+    dataset : torchvision.datasets.MNIST
+        MNIST dataset.
+    threshold : float
+        Threshold value to filter out pixel values.
+
+    Returns
+    -------
+    list
+        List containing the point cloud.
+    """
+    from torchvision.datasets import MNIST
+
+    dataset = MNIST(root=cfg["data_dir"], download=True)
+    Xs = dataset.data
+    ys = dataset.targets
+
+    point_clouds = []
+    for i, sample in enumerate(Xs):
+        sample = sample.where(sample > threshold, 0)
+        values = sample[sample > threshold].float()
+        values = values.unsqueeze(1)
+        x, y = (sample > 0).nonzero(as_tuple=True)
+        pos = torch.stack((x, y), dim=1).float()
+
+        data = Data(x=values, pos=pos, y=ys[i])
+        point_clouds.append(data)
+
+    return point_clouds

modules/models/graph/gcn.py

@@ -0,0 +1,80 @@
+import torch
+from torch import Tensor
+from torch_geometric.nn.models import GCN
+from torch_geometric.utils import scatter
+
+
+def global_mean_pool(x, batch=None, size=None) -> Tensor:
+    r"""Returns batch-wise graph-level-outputs by averaging node features
+    across the node dimension.
+
+    For a single graph :math:`\mathcal{G}_i`, its output is computed by
+
+    .. math::
+        \mathbf{r}_i = \frac{1}{N_i} \sum_{n=1}^{N_i} \mathbf{x}_n.
+
+    Functional method of the
+    :class:`~torch_geometric.nn.aggr.MeanAggregation` module.
+
+    Parameters
+    ----------
+    x : torch.Tensor
+        Node feature matrix :math:`\mathbf{X}`.
+    batch : torch.Tensor, optional
+        The batch vector :math:`\mathbf{b} \in {\{ 0, \ldots, B-1\}}^N`,
+            which assigns each node to a specific example.
+    size : int, optional
+        The number of examples :math:`B`. Automatically calculated if not given.
+    """
+    dim = -1 if isinstance(x, Tensor) and x.dim() == 1 else -2
+
+    if batch is None:
+        return x.mean(dim=dim, keepdim=x.dim() <= 2)
+    return scatter(x, batch, dim=dim, dim_size=size, reduce="mean")
+
+
+class GCNModel(torch.nn.Module):
+    r"""A simple GCN model that runs over graph data.
+    Note that some parameters are defined by the considered dataset.
+
+    Parameters
+    ----------
+    model_config : Dict | DictConfig
+        Model configuration.
+    dataset_config : Dict | DictConfig
+        Dataset configuration.
+    """
+
+    def __init__(self, model_config, dataset_config):
+        in_channels = (
+            dataset_config["num_features"]
+            if isinstance(dataset_config["num_features"], int)
+            else dataset_config["num_features"][0]
+        )
+        hidden_channels = model_config["hidden_channels"]
+        out_channels = dataset_config["num_classes"]
+        n_layers = model_config["n_layers"]
+        super().__init__()
+        self.base_model = GCN(
+            in_channels=in_channels,
+            hidden_channels=hidden_channels,
+            out_channels=out_channels,
+            num_layers=n_layers,
+        )
+        self.pool = global_mean_pool
+
+    def forward(self, data):
+        r"""Forward pass of the model.
+
+        Parameters
+        ----------
+        data : torch_geometric.data.Data
+            Input data.
+
+        Returns
+        -------
+        torch.Tensor
+            Output tensor.
+        """
+        z = self.base_model(data.x, data.edge_index)
+        return self.pool(z)

modules/transforms/data_transform.py

@@ -15,6 +15,7 @@
 from modules.transforms.liftings.graph2simplicial.clique_lifting import (
     SimplicialCliqueLifting,
 )
+from modules.transforms.liftings.pointcloud2graph.knn_lifting import GraphKNNLifting
 
 TRANSFORMS = {
     # Graph -> Hypergraph
@@ -31,6 +32,8 @@
     "OneHotDegreeFeatures": OneHotDegreeFeatures,
     "NodeFeaturesToFloat": NodeFeaturesToFloat,
     "KeepOnlyConnectedComponent": KeepOnlyConnectedComponent,
+    # Point Cloud -> Graph
+    "GraphKNNLifting": GraphKNNLifting,
 }