Mapping Relationship Discovery of Multidimensional Architectures in Autonomous Transportation System Based on Text-Matching Model

Table 4

Parameters of BERT and its derivative models.

Pretraining model	Number of hidden layers	Size of hidden layer	Size of output
BERT	12	768	768
ALBERT	12	768	128
Roberta	12	768	768
BERT-wwm	12	768	768