mirror of
https://github.com/zenlm/grpo.git
synced 2026-10-30 14:59:43 +00:00
zeekay
4f71f6f7c5
- Complete guide and implementation for LLM fine-tuning using GRPO - Hanzo AI cloud platform integration with CLI tools - Kubeflow ML pipeline configuration for scalable training - Example implementation with Skippy educational platform dataset - Full documentation including setup, usage, and deployment guides - Python package structure with trainer, dataset, and reward modules - Configuration files for both local and cloud deployment
45 lines
659 B
Plaintext
45 lines
659 B
Plaintext
# Core ML Libraries
|
|
datasets>=2.14.0
|
|
transformers>=4.35.0
|
|
trl>=0.7.0
|
|
torch>=2.0.0
|
|
peft>=0.6.0
|
|
accelerate>=0.24.0
|
|
|
|
# Hanzo AI Platform Libraries
|
|
hanzoai>=0.1.0
|
|
hanzoai-ml>=0.1.0
|
|
hanzoai-cli>=0.1.0
|
|
|
|
# Additional Dependencies
|
|
numpy>=1.24.0
|
|
pandas>=2.0.0
|
|
scikit-learn>=1.3.0
|
|
scipy>=1.10.0
|
|
matplotlib>=3.7.0
|
|
seaborn>=0.12.0
|
|
|
|
# Utilities
|
|
pyyaml>=6.0
|
|
python-dotenv>=1.0.0
|
|
rich>=13.0.0
|
|
typer>=0.9.0
|
|
click>=8.1.0
|
|
|
|
# Development Dependencies
|
|
pytest>=7.4.0
|
|
pytest-cov>=4.1.0
|
|
black>=23.0.0
|
|
isort>=5.12.0
|
|
flake8>=6.0.0
|
|
mypy>=1.5.0
|
|
|
|
# Logging and Monitoring
|
|
loguru>=0.7.0
|
|
wandb>=0.15.0
|
|
tensorboard>=2.14.0
|
|
|
|
# API and Serving
|
|
fastapi>=0.100.0
|
|
uvicorn>=0.23.0
|
|
pydantic>=2.0.0 |