Bert-base-vi-cased Open-Source Vietnamese Model - Supports Case Handling and Replicates Original Output

Bert Base Vi Cased

Developed by Geotrend

A customized Vietnamese lite version of the BERT model based on bert-base-multilingual-cased, supporting case-sensitive processing and fully reproducing the original model's representation output.

Large Language Model OtherOpen Source License:Apache-2.0 #Vietnamese BERT #Multilingual Lite #Lossless Representation

Downloads 15

Release Time : 3/2/2022

Model Overview

This model is an optimized base version of BERT for Vietnamese, retaining the accuracy of the original multilingual BERT model while being customized for efficiency.

Model Features

Customized Lite

A Vietnamese-optimized base version of BERT, more lightweight compared to the full multilingual model

Lossless Accuracy

Fully reproduces the representation output of the original bert-base-multilingual-cased model

Case-Sensitive

Supports case-sensitive processing for Vietnamese

Model Capabilities

Vietnamese text understanding

Vietnamese text representation

Vietnamese downstream task fine-tuning

Use Cases

Natural Language Processing

Vietnamese Text Classification

Can be used for tasks like Vietnamese news classification and sentiment analysis

Vietnamese Question Answering System

Serves as a base model for building Vietnamese question-answering systems

🚀 bert-base-vi-cased

This project shares smaller versions of bert-base-multilingual-cased that support a specific set of languages. Unlike distilbert-base-multilingual-cased, these versions produce the same representations as the original model, maintaining its accuracy.

🚀 Quick Start

The bert-base-vi-cased is a specialized version of the multilingual BERT model. It offers the same accuracy as the original while being more focused on specific languages.

✨ Features

Same Representations: Produces exactly the same representations as the original bert-base-multilingual-cased model, preserving the original accuracy.
Custom Language Support: Handles a custom number of languages.

📦 Installation

There is no specific installation command provided in the original README. However, you need to have the transformers library installed to use this model. You can install it using the following command:

pip install transformers

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-vi-cased")
model = AutoModel.from_pretrained("Geotrend/bert-base-vi-cased")

Advanced Usage

To generate other smaller versions of multilingual transformers, please visit our Github repo.

📚 Documentation

For more information, please visit our paper: Load What You Need: Smaller Versions of Multilingual BERT.

How to cite

@inproceedings{smallermbert,
  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
  booktitle={SustaiNLP / EMNLP},
  year={2020}
}

📄 License

This project is licensed under the Apache-2.0 license.

🔧 Technical Details

The model is based on the bert-base-multilingual-cased architecture. It is trained on the Wikipedia dataset in Vietnamese.

Property	Details
Model Type	bert-base-vi-cased
Training Data	wikipedia

Contact

Please contact amine@geotrend.fr for any question, feedback or request.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご