File size: 847 Bytes
b492544 ca8c98f af92b3d ca8c98f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 |
## bert-from-clip-chinese pretrained model
该预训练权重是从BertCLIP预训练模型中单独取出来的Bert的权重,BertCLIP是中文版本的CLIP预训练模型,基于LiT-tuning(Locked-image Text tuning)的策略,
使用140万中文图文对数据进行多模态对比学习预训练。
Github: [CLIP-Chinese](https://github.com/yangjianxin1/CLIP-Chinese)
Bolg: [CLIP-Chinese:中文多模态对比学习CLIP预训练模型](https://mp.weixin.qq.com/s/6gQX91M-Lt7eiMimhYRJEw)
## Usage
可以直接使用Huggingface的BertModel直接加载该预训练权重,进行下游的任务。
```python
from transformers import BertTokenizer, BertModel
model_name_or_path = 'YeungNLP/bert-from-clip-chinese-1M'
tokenizer = BertTokenizer.from_pretrained(model_name_or_path)
model = BertModel.from_pretrained(model_name_or_path)
``` |