Table 9 The hyper-parameters of classifiers
From: Exploring best-matched embedding model and classifier for charging-pile fault diagnosis
Classifier | Hyper-parameters |
|---|---|
RF | n_estimators = 500, n_jobs = -1 |
SVM | kernel = rbf, decision_function_shape = ovr, max_iter = 1000 |
KNN | n_neighbors = 8 |
MLP | activation = relu, solver = adam, momentum = 0.9, learning_rate_init = 0.001, random_state = 1 |
RNN | loss = CrossEntropy, hidden_dim = 128, layer = 2, optimizer = adam |
AB | n_estimators = 500 |
GBDT | n_estimators = 500, learning_rate = 0.1, max_depth = 1, random_state = 1 |
DT | criterion = gini |
ET | criterion = gini |
VOTE | voting = hard, including six classifiers; same hyper-parameters as RF, SVM, KNN, MLP, AB, and GBDT |