Hugging Face's logo

IDEA-CCNL
/
Erlangshen-DeBERTa-v2-186M-Chinese-SentencePiece

Fill-Mask
Transformers
PyTorch
Safetensors
Chinese
deberta-v2
bert
Inference Endpoints
Model card Files Community
1
wanng commited on
Commit
2ce5650
•
1 Parent(s): 33623e3

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -10
README.md CHANGED
@@ -45,7 +45,7 @@ We present the results (dev set) on the following tasks:
45
  | Model | OCNLI | CMNLI |
46
  | ---------------------------------------------------- | ------ | ------ |
47
  | RoBERTa-base | 0.743 | 0.7973 |
48
- | **Erlangshen-DeBERTa-v2-186M-Chinese-SentencePiece** | 0.7625 | 0.81 |
49
 
50
  ## 使用 Usage
51
 
@@ -60,15 +60,6 @@ fillmask_pipe = FillMaskPipeline(model, tokenizer)
60
  print(fillmask_pipe(text, top_k=10))
61
  ```
62
 
63
- ## Finetune
64
-
65
- We present the dev results on some tasks.
66
-
67
- | Model | OCNLI | CMNLI |
68
- | ---------------------------------------------------- | ------ | ------ |
69
- | RoBERTa-base | 0.743 | 0.7973 |
70
- | **Erlangshen-DeBERTa-v2-186M-Chinese-SentencePiece** | 0.7625 | 0.81 |
71
-
72
  ## 引用 Citation
73
 
74
  如果您在您的工作中使用了我们的模型，可以引用我们的[论文](https://arxiv.org/abs/2209.02970)：
 
45
  | Model | OCNLI | CMNLI |
46
  | ---------------------------------------------------- | ------ | ------ |
47
  | RoBERTa-base | 0.743 | 0.7973 |
48
+ | **Erlangshen-DeBERTa-v2-186M-Chinese-SentencePiece** | 0.7625 | 0.8100 |
49
 
50
  ## 使用 Usage
51
 
 
60
  print(fillmask_pipe(text, top_k=10))
61
  ```
62
 
 
 
 
 
 
 
 
 
 
63
  ## 引用 Citation
64
 
65
  如果您在您的工作中使用了我们的模型，可以引用我们的[论文](https://arxiv.org/abs/2209.02970)：