Open-source Model Erlangshen-MegatronBert-1.3B-Sentiment - Achieve Precise Chinese Sentiment Analysis for Free

Erlangshen MegatronBert 1.3B Sentiment

Developed by IDEA-CCNL

A Chinese sentiment analysis model based on the MegatronBert architecture, fine-tuned on multiple sentiment analysis tasks

Text Classification

Transformers

ChineseOpen Source License:Apache-2.0 #Chinese Sentiment Analysis #High-precision Classification #Multi-task Fine-tuning

Downloads 93

Release Time : 5/11/2022

Model Overview

This model is a Chinese sentiment analysis model based on the MegatronBert architecture, fine-tuned on 8 Chinese sentiment analysis datasets, suitable for sentiment analysis in natural language understanding tasks.

Model Features

High-performance Sentiment Analysis

Excellent performance on multiple Chinese sentiment analysis datasets with an accuracy rate of up to 98.1%.

Large-scale Pre-training

Based on the 1.3B-parameter MegatronBert architecture, it possesses strong natural language understanding capabilities.

Multi-task Fine-tuning

Fine-tuned on 8 Chinese sentiment analysis datasets (totaling 227,347 samples), suitable for various sentiment analysis scenarios.

Model Capabilities

Chinese Sentiment Analysis

Natural Language Understanding

Text Classification

Use Cases

Sentiment Analysis

Social Media Sentiment Analysis

Analyze the sentiment tendencies of user comments on social media

Accuracy rate up to 98.1%

Product Review Analysis

Analyze the sentiment tendencies of product reviews on e-commerce platforms

Accuracy rate up to 97.8%

🚀 Erlangshen-MegatronBert-1.3B-Sentiment

A fine-tuned Chinese BERT model for sentiment analysis, which topped FewCLUE and ZeroCLUE benchmarks in 2021.

Github: Fengshenbang-LM
Docs: Fengshenbang-Docs

🚀 Quick Start

This is the fine-tuned version of the Chinese BERT model on several sentiment analysis datasets, which topped FewCLUE and ZeroCLUE benchmark in 2021.

✨ Features

Model Taxonomy

Property	Details
Demand	General
Task	Natural Language Understanding (NLU)
Series	Erlangshen
Model	MegatronBert
Parameter	1.3B
Extra	Sentiment Analysis

Model Information

Based on Erlangshen-MegatronBert-1.3B, we fine-tuned a sentiment analysis version on 8 Chinese sentiment analysis datasets, with totaling 227,347 samples.

Performance

Model	ASAP-SENT	ASAP-ASPECT	ChnSentiCorp
Erlangshen-Roberta-110M-Sentiment	97.77	97.31	96.61
Erlangshen-Roberta-330M-Sentiment	97.9	97.51	96.66
Erlangshen-MegatronBert-1.3B-Sentiment	98.1	97.8	97

💻 Usage Examples

Basic Usage

from transformers import AutoModelForSequenceClassification
from transformers import BertTokenizer
import torch

tokenizer=BertTokenizer.from_pretrained('IDEA-CCNL/Erlangshen-MegatronBert-1.3B-Sentiment')
model=AutoModelForSequenceClassification.from_pretrained('IDEA-CCNL/Erlangshen-MegatronBert-1.3B-Sentiment')

text='今天心情不好'

output=model(torch.tensor([tokenizer.encode(text)]))
print(torch.nn.functional.softmax(output.logits,dim=-1))

📚 Documentation

Citation

If you are using the resource for your work, please cite the our paper:

@article{fengshenbang,
  author    = {Jiaxing Zhang and Ruyi Gan and Junjie Wang and Yuxiang Zhang and Lin Zhang and Ping Yang and Xinyu Gao and Ziwei Wu and Xiaoqun Dong and Junqing He and Jianheng Zhuo and Qi Yang and Yongfeng Huang and Xiayu Li and Yanghan Wu and Junyu Lu and Xinyu Zhu and Weifeng Chen and Ting Han and Kunhao Pan and Rui Wang and Hao Wang and Xiaojun Wu and Zhongshen Zeng and Chongpei Chen},
  title     = {Fengshenbang 1.0: Being the Foundation of Chinese Cognitive Intelligence},
  journal   = {CoRR},
  volume    = {abs/2209.02970},
  year      = {2022}
}

You can also cite our website:

@misc{Fengshenbang-LM,
  title={Fengshenbang-LM},
  author={IDEA-CCNL},
  year={2021},
  howpublished={\url{https://github.com/IDEA-CCNL/Fengshenbang-LM}},
}

📄 License

This project is licensed under the Apache-2.0 license.

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