Oldpan/Pytorch-Memory-Utils

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1012 stars · 152 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A small set of Python scripts for tracking GPU memory usage in PyTorch, including per-line tensor allocation reports and model parameter/intermediate-variable size estimation. It is a lightweight copy-in utility rather than an installable package.

Use cases

• track gpu memory usage during pytorch training
• find which lines of code allocate the most gpu memory
• estimate the memory footprint of a pytorch model
• debug cuda out of memory errors
• measure intermediate variable memory with and without backward pass

When to choose

• you need a quick, dependency-free script to profile pytorch gpu memory
• you want line-by-line tensor allocation reports during training
• you need to estimate model parameter and activation memory before training

When to avoid

• you need a maintained installable profiler with active updates
• you use frameworks other than pytorch
• you need cpu memory profiling or multi-gpu aggregation

Facets

library · maturity maintenance

monitoring developer-tools deep-learning developer-tools performance python pytorch gpu-memory memory-profiling debugging gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Oldpan/Pytorch-Memory-Utils")