Metadata-Version: 2.4
Name: gradscope
Version: 0.1.0
Summary: Gradient and training anomaly monitoring for PyTorch and TensorFlow
Author: Prabhnoor Singh
License: MIT
Project-URL: Homepage, https://github.com/your-org/gradscope
Project-URL: Source, https://github.com/your-org/gradscope
Requires-Python: >=3.9
Description-Content-Type: text/markdown
Requires-Dist: typing-extensions; python_version < "3.11"
Provides-Extra: server
Requires-Dist: fastapi; extra == "server"
Requires-Dist: uvicorn; extra == "server"
Provides-Extra: system
Requires-Dist: psutil; extra == "system"

# GradScope

GradScope is a small, focused library that watches your training loops and tells you when gradients, metrics, or weight updates go bad.

- 2-line attach for PyTorch and TensorFlow
- Automatic collection of gradients, updates, drift, and metrics
- Rich alerting, summaries, diffing, export, CLI, and web API
- Lightweight FastAPI dashboard for quick inspection of runs
- Git-aware run metadata (commit, branch, dirty state) for reproducibility

For full documentation and examples, see [docs/README.md](docs/README.md).
