Text Analytics Toolbox Model for BERT-Small Network

Pretrained BERT-Small Network for MATLAB.

BERT-Small is a pretrained language model based on Deep Learning Transformer architecture that can be used for a wide variety of Natural Language Processing (NLP) tasks. This model has 4 self-attention layers and a hidden size of 512.

To load a BERT-Small model, you can run the following code:

[net, tokenizer] = bert(Model="small");