bert base chinese ws

Introduction

CKIP BERT BASE CHINESE is a project offering traditional Chinese transformer models, including ALBERT, BERT, and GPT-2, alongside NLP tools for word segmentation, part-of-speech tagging, and named entity recognition.

Architecture

The model is based on the BERT architecture, optimized for traditional Chinese language tasks. It supports token classification and can be utilized using libraries like PyTorch and JAX.

Training

The model was trained using traditional Chinese datasets and supports various NLP tasks. It is optimized for performance in token classification.

Guide: Running Locally

To run the CKIP BERT BASE CHINESE model locally, follow these steps:

1. Install Transformers Library:

   pip install transformers
   

2. Load Tokenizer and Model: Use BertTokenizerFast instead of AutoTokenizer for tokenization.

   from transformers import BertTokenizerFast, AutoModel
   
   tokenizer = BertTokenizerFast.from_pretrained('bert-base-chinese')
   model = AutoModel.from_pretrained('ckiplab/bert-base-chinese-ws')
   

3. Inference: Prepare your input and use the tokenizer and model for tasks like token classification.

Cloud GPUs

For enhanced performance, consider using cloud GPU services like Amazon EC2, Google Cloud Platform, or Microsoft Azure.

License

This project is licensed under the GPL-3.0 license.