PromCSE-bert-base-zh Open-source Model - Super Practical for Free Calculation of Chinese Sentence Similarity!

Promcse Bert Base Zh

Developed by hellonlp

PromCSE is a supervised learning-based sentence embedding model specifically designed for calculating Chinese sentence similarity.

Text Embedding

Transformers

ChineseOpen Source License:MIT #Sentence Similarity Calculation #Chinese Semantic Matching #Supervised Contrastive Learning

Downloads 1,761

Release Time : 12/7/2023

Model Overview

PromCSE is a supervised learning-based sentence embedding model that optimizes sentence representations through contrastive learning, specifically designed for calculating Chinese sentence similarity. It excels in multiple Chinese sentence similarity tasks.

Model Features

High-Performance Sentence Embedding

Achieves optimal performance on multiple Chinese sentence similarity datasets

Supervised Learning Optimization

Uses supervised learning methods to optimize sentence representations

Easy to Use

Provides a simple API interface for easy integration into existing systems

Model Capabilities

Sentence Vector Generation

Sentence Similarity Calculation

Semantic Search

Use Cases

Information Retrieval

Semantic Search

Using sentence embeddings for semantic similarity search

Obtains more relevant results compared to traditional keyword search

Question Answering Systems

Question Matching

Matching user questions with similar questions in the knowledge base

Improves the accuracy of question answering systems

🚀 PromCSE(sup)

A tool for sentence similarity tasks, which uses specific datasets and models to achieve high - performance sentence encoding and similarity calculation.

📚 Documentation

🔍 Data List

The following datasets are all in Chinese.

Data	size(train)	size(valid)	size(test)
ATEC	62477	20000	20000
BQ	100000	10000	10000
LCQMC	238766	8802	12500
PAWSX	49401	2000	2000
STS-B	5231	1458	1361
SNLI	146828	2699	2618
MNLI	122547	2932	2397

📦 Model List

The evaluation dataset is in Chinese, and we used the same language model RoBERTa base on different methods. In addition, considering that the test set of some datasets is small, which may lead to a large deviation in evaluation accuracy, the evaluation data here uses train, valid and test at the same time, and the final evaluation result adopts the weighted average (w - avg) method.

Model	STS - B(w - avg)	ATEC	BQ	LCQMC	PAWSX	Avg.
BERT - Whitening	65.27	-	-	-	-	-
SimBERT	70.01	-	-	-	-	-
SBERT - Whitening	71.75	-	-	-	-	-
[BAAI/bge - base - zh](https://huggingface.co/BAAI/bge - base - zh)	78.61	-	-	-	-	-
[hellonlp/simcse - base - zh](https://huggingface.co/hellonlp/simcse - roberta - base - zh)	80.96	-	-	-	-	-
[hellonlp/promcse - base - zh](https://huggingface.co/hellonlp/promcse - bert - base - zh)	81.57	-	-	-	-	-

💻 Usage Examples

🚀 Quick Start

To use the tool, first install the promcse package from PyPI

pip install promcse

Basic Usage

After installing the package, you can load our model by two lines of code

from promcse import PromCSE
model = PromCSE("hellonlp/promcse-bert-base-zh", "cls", 10)

Then you can use our model for encoding sentences into embeddings

embeddings = model.encode("武汉是一个美丽的城市。")
print(embeddings.shape)
#torch.Size([768])

Advanced Usage

Compute the cosine similarities between two groups of sentences

sentences_a = ['你好吗']
sentences_b = ['你怎么样','我吃了一个苹果','你过的好吗','你还好吗','你',
               '你好不好','你好不好呢','我不开心','我好开心啊', '你吃饭了吗',
               '你好吗','你现在好吗','你好个鬼']
similarities = model.similarity(sentences_a, sentences_b)
print(similarities)
# [(1.0, '你好吗'),
#  (0.9167, '你好不好'),
#  (0.8956, '你好不好呢'),
#  (0.8431, '你还好吗'),
#  (0.7919, '你怎么样'),
#  (0.7649, '你现在好吗'),
#  (0.7458, '你过的好吗'),
#  (0.6844, '你好个鬼'),
#  (0.6177, '你'),
#  (0.5654, '你吃饭了吗'),
#  (0.3612, '我好开心啊'),
#  (0.1875, '我不开心'),
#  (0.0866, '我吃了一个苹果')]

📄 License

This project is licensed under the MIT 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