ko sbert nli

提供商jhgan
分类natural-language-inference
许可证Apache-2.0
下载量103.4K
星标0

简介

ko-sbert-nli 是一款专门针对韩语优化的 Sentence-BERT 模型，核心能力在于处理自然语言推理（NLI）任务。它能精准判断两段韩语文本之间是蕴含、矛盾还是中立关系。对于需要处理韩语语料的开发者来说，它比通用多语言模型在语义匹配和相似度计算上更地道。该模型上手简单，可直接集成到基于 Transformers 的 pipeline 中，非常适合用于构建韩语语义搜索、智能问答系统或文本聚类分析，是处理韩语 NLP 任务的高效轻量化选择。

核心亮点

  • 专为韩语优化，语义匹配精度高于通用模型
  • 擅长处理 NLI 任务，精准判定文本逻辑关系
  • 兼容 Transformers 库，开发者快速部署上手
  • 适用于韩语语义搜索、问答及文本聚类场景

使用方法

安装依赖
# 安装 Hugging Face transformers
pip install transformers torch
SDK 使用
# 使用 transformers 加载模型
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("jhgan/ko-sbert-nli")
tokenizer = AutoTokenizer.from_pretrained("jhgan/ko-sbert-nli")

Hugging Face 下载

我们推荐使用命令行或者 Hugging Face Hub SDK 来进行模型的下载。

操作指引：在下载前，请先通过如下命令安装 huggingface_hub：

操作指引 
pip install -U huggingface_hub

命令行下载

下载完整模型库

下载完整模型库 
huggingface-cli download jhgan/ko-sbert-nli

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例）

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例） 
huggingface-cli download jhgan/ko-sbert-nli config.json --local-dir ./dir

更多命令行下载选项，可参见官方文档

SDK 下载

SDK 下载 
# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('jhgan/ko-sbert-nli')

Git 下载

请确保 lfs 已经被正确安装

Git 下载 
git lfs install
git clone https://huggingface.co/jhgan/ko-sbert-nli

如果您希望跳过 lfs 大文件下载，可以使用如下命令

跳过 LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/jhgan/ko-sbert-nli

模型文件托管在 Hugging Face Hub，使用 HF CLI / SDK / Git 直接下载，不经过本站。

PyTorch / Transformers 使用

安装 Transformers

安装 Transformers 
pip install -U transformers torch

模型加载和推理

模型加载和推理 
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('jhgan/ko-sbert-nli')
tokenizer = AutoTokenizer.from_pretrained('jhgan/ko-sbert-nli')

完整文档

来源: HuggingFace

---
pipeline_tag: sentence-similarity
tags:

  • sentence-transformers

  • feature-extraction

  • sentence-similarity

  • transformers

---

ko-sbert-nli

This is a sentence-transformers model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.

<!--- Describe your model here -->

Usage (Sentence-Transformers)

Using this model becomes easy when you have sentence-transformers installed:

code
pip install -U sentence-transformers

Then you can use the model like this:

python
from sentence_transformers import SentenceTransformer
sentences = ["안녕하세요?", "한국어 문장 임베딩을 위한 버트 모델입니다."]

model = SentenceTransformer('jhgan/ko-sbert-nli')
embeddings = model.encode(sentences)
print(embeddings)

Usage (HuggingFace Transformers)

Without sentence-transformers, you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
python
from transformers import AutoTokenizer, AutoModel
import torch

#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
    token_embeddings = model_output[0] #First element of model_output contains all token embeddings
    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
    return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)


Sentences we want sentence embeddings for

sentences = ['This is an example sentence', 'Each sentence is converted']

Load model from HuggingFace Hub

tokenizer = AutoTokenizer.from_pretrained('jhgan/ko-sbert-nli')
model = AutoModel.from_pretrained('jhgan/ko-sbert-nli')

Tokenize sentences

encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')

Compute token embeddings

with torch.no_grad():
    model_output = model(encoded_input)

Perform pooling. In this case, mean pooling.

sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])

print("Sentence embeddings:")
print(sentence_embeddings)

Evaluation Results

<!--- Describe how your model was evaluated -->

KorNLI 학습 데이터셋으로 학습한 후 KorSTS 평가 데이터셋으로 평가한 결과입니다.

  • Cosine Pearson: 82.24
  • Cosine Spearman: 83.16
  • Euclidean Pearson: 82.19
  • Euclidean Spearman: 82.31
  • Manhattan Pearson: 82.18
  • Manhattan Spearman: 82.30
  • Dot Pearson: 79.30
  • Dot Spearman: 78.78

Training

The model was trained with the parameters:

DataLoader:

sentence_transformers.datasets.NoDuplicatesDataLoader.NoDuplicatesDataLoader of length 8885 with parameters:

code
{'batch_size': 64}

Loss**:

sentence_transformers.losses.MultipleNegativesRankingLoss.MultipleNegativesRankingLoss with parameters:

code
{'scale': 20.0, 'similarity_fct': 'cos_sim'}

Parameters of the fit()-Method:

code
{
    "epochs": 1,
    "evaluation_steps": 1000,
    "evaluator": "sentence_transformers.evaluation.EmbeddingSimilarityEvaluator.EmbeddingSimilarityEvaluator",
    "max_grad_norm": 1,
    "optimizer_class": "<class 'transformers.optimization.AdamW'>",
    "optimizer_params": {
        "lr": 2e-05
    },
    "scheduler": "WarmupLinear",
    "steps_per_epoch": null,
    "warmup_steps": 889,
    "weight_decay": 0.01
}

Full Model Architecture

code
SentenceTransformer(
  (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)

Citing & Authors

<!--- Describe where people can find more information -->

  • Ham, J., Choe, Y. J., Park, K., Choi, I., & Soh, H. (2020). Kornli and korsts: New benchmark datasets for korean natural language understanding. arXiv preprint arXiv:2004.03289

  • Reimers, Nils and Iryna Gurevych. “Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks.” ArXiv abs/1908.10084 (2019)

  • Reimers, Nils and Iryna Gurevych. “Making Monolingual Sentence Embeddings Multilingual Using Knowledge Distillation.” EMNLP (2020).

查看官方来源