pytorch-patterns
PyTorch deep learning patterns and best practices for building robust, efficient, and reproducible training pipelines, model architectures, and data loading. Use when writing or reviewing PyTorch training loops, model architectures, or data loading, or when a run will not reproduce.
- installs 8w
- 1,857
- 30-day movement
- starts with the next reading
- Related entries
- 6
- Connections
- 0
An agent skill (SKILL.md) by affaan-m that provides PyTorch patterns and best practices for training pipelines, model architectures, and data loading. It covers device-agnostic code, reproducibility, architecture, training loops, data pipelines, checkpointing, and performance optimization.
Reach for it when writing or reviewing PyTorch code, debugging training loops or data pipelines, or when a run will not reproduce.
Use it to
- Write new PyTorch models or training scripts
- Review deep learning code
- Debug training loops or data pipelines
- Optimize GPU memory usage or training speed
- Fix non-reproducible runs
For Developers writing, reviewing, or debugging PyTorch deep learning code
- Host repository
- affaan-m/ECC
- Installs, lifetime
- 3,100
- Installs, 8 weeks
- 1,857
- Host stars
- 261k
- Host language
- JavaScript