Instructions

English | 中文

UER-py supports a wide range of pre-training models and downstream tasks. This section shows the comprehensive use-cases when using the UER-py. In many cases, we use BERT model to demonstrate how to use UER-py by default.

• Preprocess the data

• Pretrain the model

• More pretraining models

• Finetune the tasks

• Tokenization and vocabulary

• Scripts