RMalkiv/torch-audit
The Linter for PyTorch: Detects silent training bugs
This tool helps machine learning engineers and researchers building PyTorch models catch hidden issues during training. It acts like a "check engine light" for your model's training loop, inspecting real data, gradients, and model behavior as it runs. You provide your PyTorch model and optimizer, and it flags potential problems like incorrect data formats, unstable gradients, or misconfigured optimizers that might silently degrade performance or waste computing resources.
Available on PyPI.
Use this if you are training PyTorch models and want to automatically detect silent bugs and inefficiencies that don't cause crashes but lead to poor model performance or wasted compute.
Not ideal if you are looking for a static code linter for general Python code or if you are not using PyTorch for your deep learning projects.
Stars
12
Forks
—
Language
Python
License
MIT
Category
Last pushed
Jan 24, 2026
Commits (30d)
0
Dependencies
3
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/RMalkiv/torch-audit"
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, ...)