cleanup

.github/workflows/test_codebase.yml

@@ -1,6 +1,6 @@
 name: "Testing Codebase"
 
-on: 
+on:
   workflow_dispatch:
   push:
     branches: [main,github-actions-test]
@@ -54,7 +54,7 @@ jobs:
           pip install -e .[all]
       - name: Run tests for codebase [pytest]
         run: |
-          pytest -n 2 --cov --cov-report=xml:coverage.xml test/transforms/feature_liftings test/transforms/liftings 
+          pytest -n 2 --cov --cov-report=xml:coverage.xml test/transforms/feature_liftings test/transforms/liftings
       - name: Upload coverage
         uses: codecov/codecov-action@v3
         with:

.github/workflows/test_tutorials.yml

@@ -1,6 +1,6 @@
 name: "Testing Tutorials"
 
-on: 
+on:
   workflow_dispatch:
   push:
     branches: [main,github-actions-test]
@@ -54,6 +54,6 @@ jobs:
       - name: Install main package
         run: |
           pip install -e .[all]
-      - name: Run tests for tutorials on all domains [pytest] 
+      - name: Run tests for tutorials on all domains [pytest]
         run: |
           pytest test/tutorials/test_tutorials.py

README.md

@@ -6,7 +6,7 @@ Official repository for the ICML 2024 Topological Deep Learning Challenge, hoste
 - Please, check out the [main webpage of the challenge](https://pyt-team.github.io/packs/challenge.html) for the full description of the competition (motivation, submission requirements, evaluation, etc.)
 
 ## Brief Description
-The main purpose of the challenge is to further expand the current scope and impact of Topological Deep Learning (TDL), enabling the exploration of its applicability in new contexts and scenarios. To do so, we propose participants to design and implement lifting mappings between different data structures and topological domains (point-clouds, graphs, hypergraphs, simplicial/cell/combinatorial complexes), potentially bridging the gap between TDL and all kinds of existing datasets. 
+The main purpose of the challenge is to further expand the current scope and impact of Topological Deep Learning (TDL), enabling the exploration of its applicability in new contexts and scenarios. To do so, we propose participants to design and implement lifting mappings between different data structures and topological domains (point-clouds, graphs, hypergraphs, simplicial/cell/combinatorial complexes), potentially bridging the gap between TDL and all kinds of existing datasets.
 
 
 ## General Guidelines

modules/data/load/loaders.py

@@ -6,8 +6,8 @@
 from omegaconf import DictConfig
 
 from modules.data.load.base import AbstractLoader
-from modules.data.utils.custom_dataset import CustomDataset
 from modules.data.utils.concat2geometric_dataset import ConcatToGeometricDataset
+from modules.data.utils.custom_dataset import CustomDataset
 from modules.data.utils.utils import (
     load_cell_complex_dataset,
     load_hypergraph_pickle_dataset,

modules/data/utils/concat2geometric_dataset.py

@@ -1,6 +1,7 @@
 import torch
 from torch_geometric.data import Data, Dataset
 
+
 class ConcatToGeometricDataset(Dataset):
     def __init__(self, concat_dataset):
         super().__init__()
@@ -11,18 +12,18 @@ def len(self):
 
     def get(self, idx):
         data = self.concat_dataset[idx]
-        
+
         x = data.x.float()
         edge_index = data.edge_index
         edge_attr = data.edge_attr
         y = data.y
-        if len(x.shape)==1:
+        if len(x.shape) == 1:
             x = x.unsqueeze(dim=1)
-        if len(edge_attr.shape)==1:
+        if len(edge_attr.shape) == 1:
             edge_attr = edge_attr.unsqueeze(dim=1)
-        if len(y.shape)==1:
+        if len(y.shape) == 1:
             y = y.unsqueeze(dim=1)
 
         # Construct PyTorch Geometric Data object
         data = Data(x=x, edge_index=edge_index, edge_attr=edge_attr, y=y)
-        return data
+        return data

modules/transforms/liftings/graph2hypergraph/knn_lifting.py

@@ -44,13 +44,28 @@ def lift_topology(self, data: torch_geometric.data.Data) -> dict:
         # check for loops, since KNNGraph is inconsistent with nodes with equal features
         if self.loop:
             for i in range(num_nodes):
-                if not torch.any(torch.all(data_lifted.edge_index == torch.tensor([[i,i]]).T, dim=0)):
-                    connected_nodes = data_lifted.edge_index[0, data_lifted.edge_index[1]==i]
-                    dists = torch.sqrt(torch.sum((data.pos[connected_nodes]-data.pos[i].unsqueeze(0)**2),dim=1))
+                if not torch.any(
+                    torch.all(data_lifted.edge_index == torch.tensor([[i, i]]).T, dim=0)
+                ):
+                    connected_nodes = data_lifted.edge_index[
+                        0, data_lifted.edge_index[1] == i
+                    ]
+                    dists = torch.sqrt(
+                        torch.sum(
+                            (data.pos[connected_nodes] - data.pos[i].unsqueeze(0) ** 2),
+                            dim=1,
+                        )
+                    )
                     furthest = torch.argmax(dists)
-                    idx = torch.where(torch.all(data_lifted.edge_index==torch.tensor([[connected_nodes[furthest],i]]).T, dim=0))[0]
-                    data_lifted.edge_index[:,idx] = torch.tensor([[i,i]]).T
-
+                    idx = torch.where(
+                        torch.all(
+                            data_lifted.edge_index
+                            == torch.tensor([[connected_nodes[furthest], i]]).T,
+                            dim=0,
+                        )
+                    )[0]
+                    data_lifted.edge_index[:, idx] = torch.tensor([[i, i]]).T
+
         incidence_1[data_lifted.edge_index[1], data_lifted.edge_index[0]] = 1
         incidence_1 = torch.Tensor(incidence_1).to_sparse_coo()
         return {

modules/utils/utils.py

@@ -1,5 +1,6 @@
 import pprint
 import random
+import shutil
 
 import matplotlib.pyplot as plt
 import networkx as nx
@@ -9,9 +10,8 @@
 import torch
 import torch_geometric
 from matplotlib.patches import Polygon
-import shutil
 
-plt.rcParams['text.usetex']= True if shutil.which('latex') else False
+plt.rcParams["text.usetex"] = True if shutil.which("latex") else False
 rootutils.setup_root("./", indicator=".project-root", pythonpath=True)