Table 4 Best tuned hyperparameters for the machine learning models
From: RCC-Supporter: supporting renal cell carcinoma treatment decision-making using machine learning
Model | Parameters | Hyperparameter values for each feature set | ||||||
|---|---|---|---|---|---|---|---|---|
20 | 50 | 75 | 100 | 150 | All | Export50 | ||
Decision Tree | max_depth | 4 | 4 | 6 | 6 | 6 | 6 | 5 |
min_samples_split | 13 | 22 | 21 | 21 | 20 | 21 | 17 | |
Random Forest | n_estimators | 300 | 200 | 250 | 250 | 200 | 300 | 250 |
max_depth | 13 | 14 | 15 | 14 | 15 | 13 | 13 | |
min_samples_split | 9 | 10 | 9 | 8 | 8 | 8 | 10 | |
GBM | n_estimators | 200 | 250 | 250 | 300 | 300 | 300 | 300 |
max_depth | 7 | 7 | 5 | 5 | 5 | 5 | 7 | |
learning_rate | 0.05 | 0.1 | 0.1 | 0.01 | 0.05 | 0.05 | 0.01 | |