suriyadeepan/torchtest
Unit Testing for pytorch, based on mltest
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.
Stars
312
Forks
17
Language
Python
License
GPL-3.0
Category
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.
Higher-rated alternatives
pytorch/pytorch
Tensors and Dynamic neural networks in Python with strong GPU acceleration
keras-team/keras
Deep Learning for humans
Lightning-AI/torchmetrics
Machine learning metrics for distributed, scalable PyTorch applications.
Lightning-AI/pytorch-lightning
Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
lanpa/tensorboardX
tensorboard for pytorch (and chainer, mxnet, numpy, ...)