Description
pytdc is an agent skill for working with Therapeutics Data Commons through the PyTDC package, imported as tdc. It covers discovering therapeutic machine-learning tasks, loading approved datasets, applying task-appropriate data splits, evaluating predictions with the standard metrics, and working with curated benchmark groups. The skill tells you to prefer the package's own metadata over copied dataset lists, and to plan the network and storage effects before you construct any loader, since some datasets download when loaded. Use it for drug and therapeutics machine learning benchmarks where a standard split and evaluation matter for comparing methods. It is part of the k-dense-ai scientific agent skills repository on GitHub, which has about 49,600 stars. The listing shows it as free, and it has about 1,800 installs. Its security audits show passes from Gen Agent Trust Hub, Socket and Snyk. Install it with npx skills add https://github.com/k-dense-ai/scientific-agent-skills --skill pytdc.…
Highlights
- Discover therapeutic ML tasks
- Load approved datasets
- Task-aware splits
- Evaluator metrics
- Curated benchmark groups
Ratings & Reviews
No ratings yet
Be the first to rate pytdc.
People also visit
pytdc FAQ
What is Therapeutics Data Commons?
A collection of therapeutic machine-learning tasks and datasets, accessed in Python through PyTDC.
Why use the package metadata?
The skill advises relying on package metadata rather than copied dataset lists, which can go out of date.
Do loaders download data?
Some do, so the skill asks you to plan the network and storage effects before constructing a loader.
What does it help with?
Standard splits, evaluator metrics and benchmark groups, so methods can be compared fairly.
How do I install it?
Run npx skills add https://github.com/k-dense-ai/scientific-agent-skills --skill pytdc.
Is it checked for security?
Its security audits show passes from Gen Agent Trust Hub, Socket and Snyk.
What are the best pytdc alternatives?
Popular alternatives to pytdc include pytorch-patterns, aeon and datamol. See all pytdc alternatives