hpcaitech/TensorNVMe
A Python library transfers PyTorch tensors between CPU and NVMe
When working with very large PyTorch models or datasets that don't fit into your computer's main memory (RAM), this tool helps you temporarily move parts of your data to faster NVMe solid-state drives. It takes your PyTorch tensors and offloads them to disk, then brings them back to CPU memory when needed. This is useful for machine learning engineers and researchers training large models on systems with limited RAM.
125 stars. No commits in the last 6 months.
Use this if you are a machine learning practitioner experiencing out-of-memory errors or slow training times when working with large PyTorch tensors on a Linux system with NVMe storage.
Not ideal if you are not using PyTorch, do not have NVMe storage, or are working on macOS/Windows.
Stars
125
Forks
27
Language
C++
License
—
Category
Last pushed
Nov 27, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/hpcaitech/TensorNVMe"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
pymc-devs/pytensor
PyTensor allows you to define, optimize, and efficiently evaluate mathematical expressions...
arogozhnikov/einops
Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)
lava-nc/lava-dl
Deep Learning library for Lava
tensorly/tensorly
TensorLy: Tensor Learning in Python.
tensorpack/tensorpack
A Neural Net Training Interface on TensorFlow, with focus on speed + flexibility