|
- PyG Documentation — pytorch_geometric documentation
PyG (PyTorch Geometric) is a library built upon PyTorch to easily write and train Graph Neural Networks (GNNs) for a wide range of applications related to structured data It consists of various methods for deep learning on graphs and other irregular structures, also known as geometric deep learning, from a variety of published papers
- Installation — pytorch_geometric documentation
For earlier PyTorch versions (torch<=2 5 0), you can install PyG via Anaconda for all major OS, and CUDA combinations If you have not yet installed PyTorch, install it via conda install as described in its official documentation
- Colab Notebooks and Video Tutorials — pytorch_geometric documentation
PyTorch Geometric Tutorial Project The PyTorch Geometric Tutorial project provides video tutorials and Colab notebooks for a variety of different methods in PyG:
- Design of Graph Neural Networks — pytorch_geometric documentation
torch_geometric torch_geometric nn torch_geometric data torch_geometric loader torch_geometric sampler torch_geometric datasets torch_geometric llm torch_geometric transforms torch_geometric utils torch_geometric explain torch_geometric metrics torch_geometric distributed torch_geometric contrib torch_geometric graphgym torch_geometric profile
- Explaining Graph Neural Networks — pytorch_geometric documentation
PyG (2 3 and beyond) provides the torch_geometric explain package for first-class GNN explainability support that currently includes a flexible interface to generate a variety of explanations via the Explainer class, several underlying explanation algorithms including, e g , GNNExplainer, PGExplainer and CaptumExplainer,
- Introduction by Example — pytorch_geometric documentation
PyG contains its own torch_geometric loader DataLoader, which already takes care of this concatenation process Let’s learn about it in an example:
- External Resources — pytorch_geometric documentation
Sachin Sharma: How to Deploy (almost) any PyTorch Geometric Model on Nvidia’s Triton Inference Server with an Application to Amazon Product Recommendation and ArangoDB [Blog]
- torch_geometric — pytorch_geometric documentation
torch_geometric Tensor Objects Functions seed_everything (seed: int) → None [source] Sets the seed for generating random numbers in PyTorch, numpy and Python Parameters: seed (int) – The desired seed Return type: None get_home_dir () → str [source] Get the cache directory used for storing all PyG -related data
|
|
|