Table 14 The training time with and without imbalance learning (s)
From: Exploring best-matched embedding model and classifier for charging-pile fault diagnosis
Classifier* | GloVe | Word2vec | BERT | |
|---|---|---|---|---|
RF | 0 | 0.88 | 1.00 | 1.12 |
1 | 1.00 | 0.88 | 1.25 | |
SVM | 0 | 4.72 | 2.27 | 9.53 |
1 | 5.86 | 2.89 | 11.75 | |
KNN | 0 | 0.23 | 0.18 | 0.24 |
1 | 0.24 | 0.19 | 0.30 | |
MLP | 0 | 4.18 | 3.62 | 3.82 |
1 | 3.36 | 2.82 | 5.40 | |
RNN | 0 | 133.86 | 42.28 | 340.02 |
1 | 133.49 | 42.42 | 354.34 | |
AB | 0 | 7.88 | 2.78 | 18.25 |
1 | 9.13 | 3.19 | 25.75 | |
GBDT | 0 | 35.85 | 17.30 | 83.37 |
1 | 42.43 | 20.84 | 140.38 | |
DT | 0 | 0.08 | 0.04 | 0.11 |
1 | 0.10 | 0.06 | 0.15 | |
ET | 0 | 0.02 | 0.01 | 0.04 |
1 | 0.03 | 0.01 | 0.07 | |
VOTE | 0 | 53.92 | 27.40 | 122.15 |
1 | 61.54 | 30.76 | 176.98 | |