agentsclimarketplace

Data loading

Skill thada2402/AutoResearchClaw/researchclaw/skills/builtin/tooling/data-loading

Generate research papers autonomously by chatting with OpenClaw, using Python 3.11+, with a self-evolving framework and extensive test coverage.

Install
npx -y skills add thada2402/AutoResearchClaw --skill data-loading

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Optimize data loading pipeline to prevent GPU starvation. Use when setting up DataLoader or data preprocessing.

SKILL.md

0.8 KB, as published. Nobody here has run it

Efficient Data Loading Best Practice

  1. Use num_workers = min(8, os.cpu_count()) for DataLoader
  2. Enable pin_memory=True when using GPU
  3. Use persistent_workers=True to avoid re-spawning
  4. Pre-compute and cache transformations when possible
  5. For image data: use torchvision.transforms.v2 (faster)
  6. For large datasets: consider memory-mapped files or WebDataset
  7. Profile with torch.utils.bottleneck to find I/O bottlenecks

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.