๐ ko-sbert-sts
This is a sentence-transformers model that maps sentences and paragraphs to a 768-dimensional dense vector space, which can be used for tasks such as clustering or semantic search.
๐ Quick Start
This model can be used for sentence similarity tasks. It effectively maps sentences and paragraphs into a 768-dimensional dense vector space, facilitating tasks like clustering and semantic search.
โจ Features
- Sentence-Transformers Integration: Works seamlessly with the sentence-transformers library.
- Feature Extraction: Capable of extracting features from sentences and paragraphs.
- Sentence Similarity: Can be used to calculate the similarity between sentences.
๐ฆ Installation
Using this model becomes easy when you have sentence-transformers installed:
pip install -U sentence-transformers
๐ป Usage Examples
Basic Usage
from sentence_transformers import SentenceTransformer
sentences = ["์๋
ํ์ธ์?", "ํ๊ตญ์ด ๋ฌธ์ฅ ์๋ฒ ๋ฉ์ ์ํ ๋ฒํธ ๋ชจ๋ธ์
๋๋ค."]
model = SentenceTransformer('jhgan/ko-sbert-sts')
embeddings = model.encode(sentences)
print(embeddings)
Advanced Usage
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.
from transformers import AutoTokenizer, AutoModel
import torch
def mean_pooling(model_output, attention_mask):
token_embeddings = model_output[0]
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 = ['This is an example sentence', 'Each sentence is converted']
tokenizer = AutoTokenizer.from_pretrained('jhgan/ko-sbert-sts')
model = AutoModel.from_pretrained('jhgan/ko-sbert-sts')
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
with torch.no_grad():
model_output = model(**encoded_input)
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
print("Sentence embeddings:")
print(sentence_embeddings)
๐ Documentation
Evaluation Results
This is the result of evaluating the model on the KorSTS evaluation dataset after training on the KorSTS training dataset.
Similarity Measure |
Pearson |
Spearman |
Cosine |
81.55 |
81.23 |
Euclidean |
79.94 |
79.79 |
Manhattan |
79.90 |
79.75 |
Dot |
76.02 |
75.31 |
Training
The model was trained with the following parameters:
DataLoader:
torch.utils.data.dataloader.DataLoader
of length 719 with parameters:
{'batch_size': 8, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
Loss:
sentence_transformers.losses.CosineSimilarityLoss.CosineSimilarityLoss
Parameters of the fit()-Method:
{
"epochs": 5,
"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": 360,
"weight_decay": 0.01
}
Full Model Architecture
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
- 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. โSentence-BERT: Sentence Embeddings using Siamese BERT-Networks.โ ArXiv abs/1908.10084 (2019)
- Reimers, Nils and Iryna Gurevych. โMaking Monolingual Sentence Embeddings Multilingual Using Knowledge Distillation.โ EMNLP (2020)