Experiment Reproducibility Routine #20

Code Example

import pandas as pd

df = pd.DataFrame({
  "feature_a": [12, 18, 25, 31, 28, 22],
  "feature_b": [3, 4, 8, 7, 6, 5],
  "target": [0, 0, 1, 1, 1, 0],
})

def build_summary(frame: pd.DataFrame):
  metrics = {
    "rows": len(frame),
    "mean_a": float(frame["feature_a"].mean()),
    "mean_b": float(frame["feature_b"].mean()),
    "target_rate": float(frame["target"].mean()),
  }
  return metrics

print("run:", "python script.py")
print(build_summary(df))

Commands & References

• python script.py
• df.describe()
• plt.plot()

Lab Steps

1. Prepare environment using: python script.py
2. Load a small sample dataset and validate schema.
3. Run the core code workflow and collect metrics.
4. Compare results and write one improvement note.

Exercises

• Change one hyperparameter and compare impact.
• Add one validation rule to reduce bad inputs.
• Document one failure mode and mitigation.