Skip to content
Get packs

PyTorch Geometric Graph Models

Build and train graph neural networks with PyTorch Geometric's layers, datasets and transforms

Source Download skill
Repository
K-Dense-AI/scientific-agent-skills
Stars
55,780
Licence
MIT
Files
8 files

What it does

Once installed, the agent knows how to work with PyTorch Geometric, a library for graph neural networks, which are models that learn from data made of nodes and connections (such as social networks, molecules or citation graphs). It guides the agent to structure graph data correctly, pick built-in layers like GCN, GAT or GraphSAGE, load standard graph datasets, and handle heterogeneous graphs, neighbor sampling and custom datasets. It is the same kind of guidance an experienced engineer would give, applied while writing code.

Great for

  • Machine learning engineers building node, link or graph classification models in PyTorch
  • Researchers prototyping on standard benchmarks such as Cora, TUDataset or OGB
  • Students learning how graph neural networks represent and pass messages between nodes

Skip it if

  • General network analysis or graph statistics with NetworkX, which the skill explicitly excludes
  • Ordinary PyTorch models that do not use graphs, since the guidance is specific to graph data
  • Production training on large datasets without first verifying installed versions and extension wheels, because the file notes that some backend combinations were not tested

Try asking

  • “Build a two-layer GCN in PyTorch Geometric to classify nodes in the Cora citation dataset”
  • “Convert my list of edge pairs into a correct edge_index tensor and validate the graph object”
  • “Set up a link prediction model on an OGB dataset while keeping the official split”
  • “Write a custom Dataset class for my own graphs and check isolated-node handling”
Step-by-step guide: how to use PyTorch Geometric Graph ModelsInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 8 original files, the MIT licence and an attribution note.

    torch-geometric.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip torch-geometric.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip torch-geometric.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Pin the exact torch-geometric and PyTorch versions the skill names, and check the installed version before using newer APIs
  • Install optional extensions only from the wheel index matching your exact Python, OS, PyTorch and CUDA combination
  • Keep official dataset splits and preprocessing when benchmarking so your results are comparable

What's inside

  • references/custom_datasets.md
  • references/explainability.md
  • references/heterogeneous.md
  • references/link_prediction.md
  • references/message_passing.md
  • references/review.md
  • references/scaling.md
  • SKILL.md

Redistributed unmodified from K-Dense-AI/scientific-agent-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.