xl0/lovely-tensors

Tensors, for human consumption

61
/ 100
Established

When working with PyTorch, JAX, or NumPy, it's often hard to quickly understand what's inside a tensor while debugging. This tool transforms raw tensor output into a concise summary showing its shape, size, value distribution, and special values like NaNs or infinities. It's designed for data scientists, machine learning engineers, and researchers who frequently inspect numerical data structures.

1,360 stars. Used by 3 other packages. Actively maintained with 1 commit in the last 30 days. Available on PyPI.

Use this if you need a quick, human-readable summary of your numerical tensors (PyTorch, JAX, or NumPy) during development or debugging, including visual representations of data distributions.

Not ideal if you need to deeply inspect every individual value of a large tensor, or if you are not working with PyTorch, JAX, or NumPy tensors.

deep-learning machine-learning-engineering scientific-computing numerical-analysis data-inspection
Maintenance 13 / 25
Adoption 13 / 25
Maturity 25 / 25
Community 10 / 25

How are scores calculated?

Stars

1,360

Forks

22

Language

Jupyter Notebook

License

MIT

Last pushed

Mar 11, 2026

Commits (30d)

1

Dependencies

2

Reverse dependents

3

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/xl0/lovely-tensors"

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