92 lines
2.9 KiB
Markdown
92 lines
2.9 KiB
Markdown
---
|
|
pipeline_tag: sentence-similarity
|
|
tags:
|
|
- sentence-transformers
|
|
- feature-extraction
|
|
- sentence-similarity
|
|
- transformers
|
|
- ko
|
|
---
|
|
|
|
# ddobokki/klue-roberta-small-nli-sts
|
|
|
|
한국어 Sentence Transformer 모델입니다.
|
|
|
|
<!--- Describe your model here -->
|
|
|
|
## Usage (Sentence-Transformers)
|
|
|
|
[sentence-transformers](https://www.SBERT.net) 라이브러리를 이용해 사용할 수 있습니다.
|
|
|
|
```
|
|
pip install -U sentence-transformers
|
|
```
|
|
|
|
사용법
|
|
|
|
```python
|
|
from sentence_transformers import SentenceTransformer
|
|
sentences = ["흐르는 강물을 거꾸로 거슬러 오르는", "세월이 가면 가슴이 터질 듯한"]
|
|
|
|
model = SentenceTransformer('ddobokki/klue-roberta-small-nli-sts')
|
|
embeddings = model.encode(sentences)
|
|
print(embeddings)
|
|
```
|
|
|
|
|
|
|
|
## Usage (HuggingFace Transformers)
|
|
transformers 라이브러리만 사용할 경우
|
|
|
|
```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 = ["흐르는 강물을 거꾸로 거슬러 오르는", "세월이 가면 가슴이 터질 듯한"]
|
|
|
|
# Load model from HuggingFace Hub
|
|
tokenizer = AutoTokenizer.from_pretrained('ddobokki/klue-roberta-small-nli-sts')
|
|
model = AutoModel.from_pretrained('ddobokki/klue-roberta-small-nli-sts')
|
|
|
|
# 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)
|
|
```
|
|
|
|
## Performance
|
|
- Semantic Textual Similarity test set results <br>
|
|
|
|
| Model | Cosine Pearson | Cosine Spearman | Euclidean Pearson | Euclidean Spearman | Manhattan Pearson | Manhattan Spearman | Dot Pearson | Dot Spearman |
|
|
|------------------------|:----:|:----:|:----:|:----:|:----:|:----:|:----:|:----:|
|
|
| KoSRoBERTa<sup>small</sup> | 84.27 | 84.17 | 83.33 | 83.65 | 83.34 | 83.65 | 82.10 | 81.38 |
|
|
|
|
|
|
## Full Model Architecture
|
|
```
|
|
SentenceTransformer(
|
|
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: RobertaModel
|
|
(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 --> |