{"id":24328,"library":"pyg-nightly","title":"PyG Nightly (PyTorch Geometric Nightly)","description":"Nightly build of PyTorch Geometric (PyG), a library for graph neural networks built on PyTorch. Current version 2.8.0.dev20260501. It tracks the latest development, includes experimental features and unreleased fixes, and is updated daily. Requires Python >=3.10 and PyTorch.","status":"active","version":"2.8.0.dev20260501","language":"python","source_language":"en","source_url":"https://github.com/pyg-team/pytorch_geometric","tags":["graph-neural-networks","deep-learning","pytorch","nightly"],"install":[{"cmd":"pip install pyg-nightly","lang":"bash","label":"Install nightly build"},{"cmd":"pip install pyg-nightly -f https://data.pyg.org/whl/torch-2.5.0+cu121.html","lang":"bash","label":"Install with CUDA 12.1 wheels"}],"dependencies":[{"reason":"Core dependency; PyG requires PyTorch.","package":"torch","optional":false}],"imports":[{"note":"The package installs as pyg-nightly but imports as torch_geometric.","wrong":"import pyg","symbol":"torch_geometric","correct":"import torch_geometric"}],"quickstart":{"code":"import torch\nimport torch_geometric\nfrom torch_geometric.data import Data\n\nedge_index = torch.tensor([[0, 1, 1, 2],\n                           [1, 0, 2, 1]], dtype=torch.long)\nx = torch.tensor([[-1], [0], [1]], dtype=torch.float)\ndata = Data(x=x, edge_index=edge_index)\nprint(data)","lang":"python","description":"Create a simple graph with two edges."},"warnings":[{"fix":"Pin to a specific nightly version if reproducibility is required, or use the stable release.","message":"Nightly builds may introduce API-breaking changes without notice. Code that works with stable PyG may break with pyg-nightly.","severity":"breaking","affected_versions":"all nightly"},{"fix":"Use `import torch_geometric`.","message":"Import as `import torch_geometric`, not `import pyg`. The pip package name is different from the import name.","severity":"gotcha","affected_versions":"all"},{"fix":"Consult the changelog and migrate to the recommended replacements.","message":"Some modules like `torch_geometric.nn.dense` are deprecated in favor of newer modules.","severity":"deprecated","affected_versions":">=2.8.0.dev"}],"env_vars":null,"last_verified":"2026-05-01T00:00:00.000Z","next_check":"2026-07-30T00:00:00.000Z","problems":[{"fix":"Use `import torch_geometric`.","cause":"Incorrect import due to name mismatch between install and import.","error":"ModuleNotFoundError: No module named 'pyg'"},{"fix":"Install it: `pip install torch-sparse-nightly` or use `pip install pyg-nightly[full]`.","cause":"Missing optional dependency for sparse operations.","error":"RuntimeError: torch_sparse is not installed. Please install it."}],"ecosystem":"pypi","meta_description":null,"install_score":null,"install_tag":null,"quickstart_score":null,"quickstart_tag":null}