What Akshay demonstrated

Built an end-to-end gas-turbine TEY predictor using leakage-free sensor features, a time-based train/test split, and honest held-out evaluation. Compared a LinearRegression baseline against a tuned RandomForest, then reported MAE, RMSE, and R2 on unseen later-period data and documented the model’s intended use, limits, and monitoring approach.

• Leakage-Free Feature Selection
• Honest Held-Out Evaluation
• Baseline-to-Ensemble Comparison