Hugging Face's logo

uer
/
roberta-base-word-chinese-cluecorpussmall

Fill-Mask
Transformers
PyTorch
TensorFlow
JAX
Chinese
bert
Inference Endpoints
Model card Files Community
2
uer commited on
Commit
80bdcf2
•
1 Parent(s): 9122f60

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +11 -11
README.md CHANGED
@@ -30,18 +30,18 @@ You can download the 5 Chinese RoBERTa miniatures either from the [UER-py Modelz
30
 
31
  Compared with [char-based models](https://huggingface.co/uer/chinese_roberta_L-2_H-128), word-based models achieve better results in most cases. Here are scores on the devlopment set of six Chinese tasks:
32
 
33
- | Model | Score | douban | chnsenticorp | lcqmc | tnews(CLUE) | iflytek(CLUE) | ocnli(CLUE) |
34
  | -------------- | :---: | :----: | :----------: | :---: | :---------: | :-----------: | :---------: |
35
- | RoBERTa-Tiny(char) | 72.3 | 83.0 | 91.4 | 81.8 | 62.0 | 55.0 | 60.3 |
36
- | **RoBERTa-Tiny(word)** | **74.3(+2.0)** | **86.4** | **93.2** | **82.0** | **66.4** | **58.2** | **59.6** |
37
- | RoBERTa-Mini(char) | 75.7 | 84.8 | 93.7 | 86.1 | 63.9 | 58.3 | 67.4 |
38
- | **RoBERTa-Mini(word)** | **76.7(+1.0)** | **87.6** | **94.1** | **85.4** | **66.9** | **59.2** | **67.3** |
39
- | RoBERTa-Small(char) | 76.8 | 86.5 | 93.4 | 86.5 | 65.1 | 59.4 | 69.7 |
40
- | **RoBERTa-Small(word)** | **78.1(+1.3)** | **88.5** | **94.7** | **87.4** | **67.6** | **60.9** | **69.8** |
41
- | RoBERTa-Medium(char) | 77.8 | 87.6 | 94.8 | 88.1 | 65.6 | 59.5 | 71.2 |
42
- | **RoBERTa-Medium(word)** | **78.9(+1.1)** | **89.2** | **95.1** | **88.0** | **67.8** | **60.6** | **73.0** |
43
- | RoBERTa-Base(char) | 79.5 | 89.1 | 95.2 | 89.2 | 67.0 | 60.9 | 75.5 |
44
- | **RoBERTa-Base(word)** | **80.2(+0.7)** | **90.3** | **95.7** | **89.4** | **68.0** | **61.5** | **76.8** |
45
 
46
  For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained with the sequence length of 128:
47
 
 
30
 
31
  Compared with [char-based models](https://huggingface.co/uer/chinese_roberta_L-2_H-128), word-based models achieve better results in most cases. Here are scores on the devlopment set of six Chinese tasks:
32
 
33
+ | Model | Score | book_review | chnsenticorp | lcqmc | tnews(CLUE) | iflytek(CLUE) | ocnli(CLUE) |
34
  | -------------- | :---: | :----: | :----------: | :---: | :---------: | :-----------: | :---------: |
35
+ | RoBERTa-Tiny(char) | 72.3 | 83.4 | 91.4 | 81.8 | 62.0 | 55.0 | 60.3 |
36
+ | **RoBERTa-Tiny(word)** | **74.4(+2.1)** | **86.7** | **93.2** | **82.0** | **66.4** | **58.2** | **59.6** |
37
+ | RoBERTa-Mini(char) | 75.9 | 85.7 | 93.7 | 86.1 | 63.9 | 58.3 | 67.4 |
38
+ | **RoBERTa-Mini(word)** | **76.9(+1.0)** | **88.5** | **94.1** | **85.4** | **66.9** | **59.2** | **67.3** |
39
+ | RoBERTa-Small(char) | 76.9 | 87.5 | 93.4 | 86.5 | 65.1 | 59.4 | 69.7 |
40
+ | **RoBERTa-Small(word)** | **78.4(+1.5)** | **89.7** | **94.7** | **87.4** | **67.6** | **60.9** | **69.8** |
41
+ | RoBERTa-Medium(char) | 78.0 | 88.7 | 94.8 | 88.1 | 65.6 | 59.5 | 71.2 |
42
+ | **RoBERTa-Medium(word)** | **79.1(+1.1)** | **90.0** | **95.1** | **88.0** | **67.8** | **60.6** | **73.0** |
43
+ | RoBERTa-Base(char) | 79.7 | 90.1 | 95.2 | 89.2 | 67.0 | 60.9 | 75.5 |
44
+ | **RoBERTa-Base(word)** | **80.4(+0.7)** | **91.1** | **95.7** | **89.4** | **68.0** | **61.5** | **76.8** |
45
 
46
  For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained with the sequence length of 128:
47