Id |
Subject |
Object |
Predicate |
Lexical cue |
T422 |
26-91 |
Sentence |
denotes |
The CNNCF consisted of stage I and stage II, as shown in Fig. 3a. |
T423 |
92-163 |
Sentence |
denotes |
Stage I was duplicated Q times in the framework (in this study, Q = 1). |
T424 |
164-286 |
Sentence |
denotes |
It consisted of multiple ResBlock-A with a number of M (in this study, M = 2), one ResBlock-B, and one Control Gate Block. |
T425 |
287-390 |
Sentence |
denotes |
Stage II consisted of multiple ResBlock-A with a number of N (in this study, N = 2) and one ResBlock-B. |
T426 |
391-537 |
Sentence |
denotes |
The weighted cross-entropy loss function was used and was minimized using the SGD optimizer with a learning rate of a1 (in this study, a1 = 0.01). |
T427 |
538-788 |
Sentence |
denotes |
A warm-up strategy58 was used in the initialization of the learning rate for a smooth training start, and a reduction factor of b1 (in this study, b1 = 0.1) was used to reduce the learning rate after every c1 (in this study, c1 = 10) training epochs. |
T428 |
789-927 |
Sentence |
denotes |
The model was trained for d1 (in this study, d1 = 40) epochs, and the model parameters saved in the last epoch was used in the test phase. |