BAAI
/

AquilaChat2-34B / README_zh.md
shunxing1234's picture
Update README_zh.md
dcb2a8a
|
raw
history blame
1.79 kB
---
license: other
---
![Aquila_logo](./log.jpeg)
<h4 align="center">
<p>
<a href="https://huggingface.co/BAAI/AquilaChat2-34B/blob/main/README.md">English</a>
<b>简体中文</b> |
</p>
</h4>
# 悟道·天鹰(Aquila2)
我们开源了我们的 **Aquila2** 系列,现在包括基础语言模型 **Aquila2-7B****Aquila2-34B** ,对话模型 **AquilaChat2-7B****AquilaChat2-34B**,长文本对话模型**AquilaChat2-7B-16k****AquilaChat2-34B-16k**
悟道 · 天鹰 Aquila 模型的更多细节将在官方技术报告中呈现。请关注官方渠道更新。
## 对话模型性能
<br>
<p align="center">
<img src="chat_metrics_CN.jpeg" width="1024"/>
<p>
<br>
## 快速开始使用 AquilaChat2-34B
## 使用方式/How to use
### 1. 推理/Inference
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
device = torch.device("cuda")
model_info = "BAAI/AquilaChat2-34B"
tokenizer = AutoTokenizer.from_pretrained(model_info, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_info, trust_remote_code=True)
model.eval()
model.to(device)
text = "请给出10个要到北京旅游的理由。"
tokens = tokenizer.encode_plus(text)['input_ids']
tokens = torch.tensor(tokens)[None,].to(device)
stop_tokens = ["###", "[UNK]", "</s>"]
with torch.no_grad():
out = model.generate(tokens, do_sample=True, max_length=512, eos_token_id=100007, bad_words_ids=[[tokenizer.encode(token)[0] for token in stop_tokens]])[0]
out = tokenizer.decode(out.cpu().numpy().tolist())
print(out)
```
## 证书/License
Aquila2系列开源模型使用 [智源Aquila系列模型许可协议](https://huggingface.co/BAAI/AquilaChat2-34B/blob/main/BAAI-Aquila-Model-License%20-Agreement.pdf)