ml-experiment

Reproducibility Checklist

• Random seeds set (Python, NumPy, PyTorch/TensorFlow)
• Data version tracked
• Code version tracked (git commit)
• Environment captured (requirements.txt)
• Hyperparameters logged
• Metrics logged
• Model artifacts saved

Setting Random Seeds

import random
import numpy as np

def set_seeds(seed: int = 42) -> None:
    """Set all random seeds for reproducibility."""
    random.seed(seed)
    np.random.seed(seed)