suriyadeepan/torchtest

Unit Testing for pytorch, based on mltest

38
/ 100
Emerging

This tool helps machine learning engineers and researchers quickly verify the core functionality of their PyTorch neural network models. It takes your PyTorch model, a loss function, an optimizer, and a batch of data, then checks for common issues like unchanged parameters, values outside expected ranges, or the presence of 'Not a Number' (NaN) or infinity (Inf) values. The output tells you whether your model behaves as expected during a training step.

312 stars. No commits in the last 6 months.

Use this if you are a machine learning engineer or researcher developing PyTorch models and want to ensure basic training mechanics and numerical stability are correct before extensive experimentation.

Not ideal if you need a comprehensive unit testing framework for complex logic, data pipelines, or full system integration tests beyond the core model's training step.

deep-learning neural-networks model-training machine-learning-development pytorch-development
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

312

Forks

17

Language

Python

License

GPL-3.0

Last pushed

Oct 15, 2020

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/suriyadeepan/torchtest"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.