ReasonGNN/lrgb-pascalvoc-neuralwalker

import torch
from datasets import load_dataset
from torch_geometric.data import Data, Batch
from tqdm import tqdm

# Load the HuggingFace dataset from hub
hf_dataset = load_dataset("samm393/lrgb-pascalvoc-neuralwalker")

# Load multiple graphs
num_graphs = 10

# Convert each to PyG Data object
pyg_graphs = []
for i in tqdm(range(num_graphs), desc="Loading graphs"):
    hf_row = hf_dataset['train'][i]
    pyg_dict = {k: torch.tensor(v) if isinstance(v, list) else v for k, v in hf_row.items()}
    pyg_data = Data.from_dict(pyg_dict)
    pyg_graphs.append(pyg_data)

# Create a batch object
batch = Batch.from_data_list(pyg_graphs)

print(f"Batch object: {batch}")
print(f"Total nodes in batch: {batch.num_nodes}")
print(f"Total edges in batch: {batch.num_edges}")
print(f"Number of graphs in batch: {batch.num_graphs}")
print(f"Batch tensor shape: {batch.batch.shape}")  # Maps each node to its graph
print(f"Batched features shape: {batch.x.shape}")
print(f"Batched NeuralWalker embeddings shape: {batch.neuralwalker_embeddings.shape}")

Batch object: DataBatch(x=[4827, 14], edge_index=[2, 27318], edge_attr=[27318, 2], y=[4827], neuralwalker_embeddings=[4827, 56], batch=[4827], ptr=[11])
Total nodes in batch: 4827
Total edges in batch: 27318
Number of graphs in batch: 10
Batch tensor shape: torch.Size([4827])
Batched features shape: torch.Size([4827, 14])
Batched NeuralWalker embeddings shape: torch.Size([4827, 56])