Table 1 Optimizing experimental outcomes.
From: Cattle identification based on multiple feature decision layer fusion
System specifications | The operating system is Ubuntu 22.04.2, the CPU is Intel(R) Core(TM) i7-7800X, the running memory is 64GB, the primary frequency is 3.50 GHz, and the GPUs are 2 NVIDIA GeForce RTX 2080 Ti. | ||
|---|---|---|---|
experimental environment | CUDA version: 11.3 Python: 3.8.10 PyTorch:1.12.0 | ||
Parameter tuning methods | Grid search and cross-validation techniques were used to tune model parameters. | ||
Model optimal hyperparameters. (Highlight the best indicators in bold) | |||
Decision Trees(DT) | ‘max_depth’: [None, 10, 20, 30] ‘min_samples_split’: [2, 5, 10] ‘min_samples_leaf’: [1, 2, 4] | Bagging | ‘n_estimators’: [2,10,15,20] ‘max_samples’: [0.5, 1.0] ‘bootstrap’: [True, False] |
Logistic regression (Lr) | ‘penalty’: [‘l1’, ‘l2’] ‘C’: [0.01, 0.1, 1, 10] ‘solver’: [‘liblinear’, ‘lbfgs’,‘saga’] | Gradient_Boosting_Classifier | ‘n_estimators’:[50,100, 150] ‘learning_rate’:[0.01,0.1,0.2] ‘max_depth’: [3, 5,7] |
Gaussian_NB (GS) | ‘var_smoothing’: [1e-9, 1e-8, 1e-7, 1e-6] | LightGBM | ‘n_estimators’:[50,100, 150] ‘learning_rate’:[0.01,0.1,0.2] ‘max_depth’: [3, 5,7] |
Random_Forests(RF) | ‘n_estimators’:[50, 100, 150, 200] ‘criterion’: [‘gini’, ‘entropy’] ‘max_depth’: [None, 10, 20, 30] ‘min_samples_split’: [2, 5, 10] ‘min_samples_leaf’: [1, 2, 4] ‘max_features’:[None,‘sqrt’,‘log2’] | XGboost | ‘n_estimators’: [50,100, 150] ‘learning_rate’:[0.01,0.1,0.2] ‘max_depth’: [3, 5,7] |
Voting Classifier | ‘voting’: [‘hard’, ‘soft’] ‘weights’: {‘DT’:2, ‘LG’:4, ‘GS’: 1, ‘RF’: 3} | Stacking | ‘cv’: [5, 10] ‘stack_method’: ‘auto’ |
