ml-training-recipes
orchestra-research835 installs
Train and debug your PyTorch neural networks.
What it does
Guides neural-network training with PyTorch recipes for model selection, optimization, debugging, and controlled experiments.
When to use it
Use when training or fine-tuning neural networks, investigating loss spikes or out-of-memory errors, selecting architectures, or improving GPU throughput.
How to use it
It changes how the agent selects, trains, tunes, and debugs neural networks in the project.
What you provide
- An existing project
Latest review
What you need · 2
Declares torch version 2.0.0 or newer as a dependency.
The shown training loop uses a CUDA GPU for mixed precision and synchronization.
Access · 0
No additional actions listed.