---
library_name: peft
license: mit
datasets:
- multi_nli
- snli
language:
- en
metrics:
- spearmanr
---


# AnglE📐: Angle-optimized Text Embeddings

> It is Angle 📐, not Angel 👼.

🔥 A New SOTA Model for Semantic Textual Similarity! 

Github: https://github.com/SeanLee97/AnglE

<a href="https://arxiv.org/abs/2309.12871">
    <img src="https://img.shields.io/badge/Arxiv-2306.06843-yellow.svg?style=flat-square" alt="https://arxiv.org/abs/2309.12871" />
</a>

[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/angle-optimized-text-embeddings/semantic-textual-similarity-on-sick-r-1)](https://paperswithcode.com/sota/semantic-textual-similarity-on-sick-r-1?p=angle-optimized-text-embeddings)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/angle-optimized-text-embeddings/semantic-textual-similarity-on-sts16)](https://paperswithcode.com/sota/semantic-textual-similarity-on-sts16?p=angle-optimized-text-embeddings)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/angle-optimized-text-embeddings/semantic-textual-similarity-on-sts15)](https://paperswithcode.com/sota/semantic-textual-similarity-on-sts15?p=angle-optimized-text-embeddings)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/angle-optimized-text-embeddings/semantic-textual-similarity-on-sts14)](https://paperswithcode.com/sota/semantic-textual-similarity-on-sts14?p=angle-optimized-text-embeddings)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/angle-optimized-text-embeddings/semantic-textual-similarity-on-sts13)](https://paperswithcode.com/sota/semantic-textual-similarity-on-sts13?p=angle-optimized-text-embeddings)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/angle-optimized-text-embeddings/semantic-textual-similarity-on-sts12)](https://paperswithcode.com/sota/semantic-textual-similarity-on-sts12?p=angle-optimized-text-embeddings)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/angle-optimized-text-embeddings/semantic-textual-similarity-on-sts-benchmark)](https://paperswithcode.com/sota/semantic-textual-similarity-on-sts-benchmark?p=angle-optimized-text-embeddings)


**STS Results**


| Model | STS12 | STS13 | STS14 | STS15 | STS16 | STSBenchmark | SICKRelatedness |  Avg. |
| ------- |-------|-------|-------|-------|-------|--------------|-----------------|-------|
| [SeanLee97/angle-llama-7b-nli-20231027](https://huggingface.co/SeanLee97/angle-llama-7b-nli-20231027) | 78.68 | 90.58 | 85.49 | 89.56 | 86.91 |    88.92     |      81.18      | 85.90 |
| [SeanLee97/angle-llama-7b-nli-v2](https://huggingface.co/SeanLee97/angle-llama-7b-nli-v2) | 79.00 | 90.56 | 85.79 | 89.43 | 87.00 |    88.97     |      80.94      | **85.96** |



## Usage

1) use AnglE

```bash
python -m pip install -U angle-emb
```

```python
from angle_emb import AnglE, Prompts

# init
angle = AnglE.from_pretrained('NousResearch/Llama-2-7b-hf', pretrained_lora_path='SeanLee97/angle-llama-7b-nli-v2')

# set prompt
print('All predefined prompts:', Prompts.list_prompts())
angle.set_prompt(prompt=Prompts.A)
print('prompt:', angle.prompt)

# encode text
vec = angle.encode({'text': 'hello world'}, to_numpy=True)
print(vec)
vecs = angle.encode([{'text': 'hello world1'}, {'text': 'hello world2'}], to_numpy=True)
print(vecs)
```

2) use transformers
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel, PeftConfig

peft_model_id = 'SeanLee97/angle-llama-7b-nli-20231027'
config = PeftConfig.from_pretrained(peft_model_id)
tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)
model = AutoModelForCausalLM.from_pretrained(config.base_model_name_or_path).bfloat16().cuda()
model = PeftModel.from_pretrained(model, peft_model_id).cuda()

def decorate_text(text: str):
    return f'Summarize sentence "{text}" in one word:"'

inputs = 'hello world!'
tok = tokenizer([decorate_text(inputs)], return_tensors='pt')
for k, v in tok.items():
    tok[k] = v.cuda()
vec = model(output_hidden_states=True, **tok).hidden_states[-1][:, -1].float().detach().cpu().numpy()
print(vec)
```

## Citation

You are welcome to use our code and pre-trained models. If you use our code and pre-trained models, please support us by citing our work as follows:

```bibtex
@article{li2023angle,
  title={AnglE-Optimized Text Embeddings},
  author={Li, Xianming and Li, Jing},
  journal={arXiv preprint arXiv:2309.12871},
  year={2023}
}
```