suolyer commited on
Commit
5dbd03a
1 Parent(s): eadda8d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +10 -10
README.md CHANGED
@@ -9,9 +9,9 @@ tags:
9
  - zero-shot
10
  ---
11
 
12
- # Erlangshen-RoBERTa-330M-UniMC-Chinese
13
 
14
- - Paper: [Zero-Shot Learners for Nature Language Understanding via a Unified Multiple Choice Perspective](https://github.com/IDEA-CCNL/Fengshenbang-LM)
15
  - Github: [Fengshenbang-LM](https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/unimc/)
16
  - Docs: [Fengshenbang-Docs](https://fengshenbang-doc.readthedocs.io/)
17
 
@@ -45,9 +45,9 @@ avoiding problems in commonly used large generative models such as FLAN. It not
45
  | LM-BFF | 85.6 | 54.4 | 53 | 47.1 | 41.6 | 57.6 | 61.2 | 51.7 | 54.7 | 56.32 |
46
  | P-tuning | 88.3 | 56 | 54.2 | **57.6** | 41.9 | 60.9 | 59.3 | **62.9** | 58.1 | 59.91 |
47
  | EFL | 84.9 | 45 | 52.1 | 42.7 | 66.2 | 71.8 | 30.9 | 56.6 | 53 | 55.91 |
48
- | [UniMC-110M](https://huggingface.co/IDEA-CCNL/Erlangshen-RoBERTa-110M-UniMC-Chinese) | 88.64 | 54.08 | 54.32 | 48.6 | 66.55 | 73.76 | 67.71 | 52.54 | 59.92 | 62.86 |
49
- | [UniMC-330M](https://huggingface.co/IDEA-CCNL/Erlangshen-RoBERTa-330M-UniMC-Chinese) | 89.53 | 57.3 | 54.25 | 50 | 70.59 | 77.49 | 78.09 | 55.73 | 65.16 | 66.46 |
50
- | [UniMC-1.3B](https://huggingface.co/IDEA-CCNL/Erlangshen-MegatronBERT-1.3B-UniMC-Chinese) | **89.278** | **60.9** | **57.46** | 52.89 | **76.33** | **80.37** | **90.33** | 61.73 | **79.15** | **72.05** |
51
 
52
  **Zero-shot**
53
 
@@ -59,9 +59,9 @@ avoiding problems in commonly used large generative models such as FLAN. It not
59
  | ZeroPrompt | - | - | - | 16.14 | 46.16 | - | - | - | 47.98 | - |
60
  | Yuan1.0-13B | 88.13 | 38.99 | 57.47 | 38.82 | 48.13 | 59.38 | 86.14 | 50 | 38.99 | 56.22 |
61
  | ERNIE3.0-240B | 88.75 | **50.97** | **57.83** | **40.42** | 53.57 | 64.38 | 87.13 | 56.25 | 53.46 | 61.41 |
62
- | [UniMC-110M](https://huggingface.co/IDEA-CCNL/Erlangshen-RoBERTa-110M-UniMC-Chinese) | 86.16 | 31.26 | 46.61 | 26.54 | 66.91 | 73.34 | 66.68 | 50.09 | 53.66 | 55.7 |
63
- | [UniMC-330M](https://huggingface.co/IDEA-CCNL/Erlangshen-RoBERTa-330M-UniMC-Chinese) | 87.5 | 30.4 | 47.6 | 31.5 | 69.9 | 75.9 | 78.17 | 49.5 | 60.55 | 59.01 |
64
- | [UniMC-1.3B](https://huggingface.co/IDEA-CCNL/Erlangshen-MegatronBERT-1.3B-UniMC-Chinese) | **88.79** | 42.06 | 55.21 | 33.93 | **75.57** | **79.5** | **89.4** | 50.25 | **66.67** | **64.53** |
65
 
66
 
67
 
@@ -75,13 +75,13 @@ pip install --editable .
75
 
76
  ```python3
77
  import argparse
78
- from fengshen import UniMCPiplines
79
 
80
 
81
  total_parser = argparse.ArgumentParser("TASK NAME")
82
  total_parser = UniMCPiplines.piplines_args(total_parser)
83
  args = total_parser.parse_args()
84
- args.pretrained_model_path = 'IDEA-CCNL/Erlangshen-RoBERTa-330M-UniMC-Chinese'
85
  args.learning_rate=2e-5
86
  args.max_length=512
87
  args.max_epochs=3
 
9
  - zero-shot
10
  ---
11
 
12
+ # Erlangshen-UniMC-RoBERTa-330M-Chinese
13
 
14
+ - Paper: [Zero-Shot Learners for Natural Language Understanding via a Unified Multiple Choice Perspective](https://github.com/IDEA-CCNL/Fengshenbang-LM)
15
  - Github: [Fengshenbang-LM](https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/unimc/)
16
  - Docs: [Fengshenbang-Docs](https://fengshenbang-doc.readthedocs.io/)
17
 
 
45
  | LM-BFF | 85.6 | 54.4 | 53 | 47.1 | 41.6 | 57.6 | 61.2 | 51.7 | 54.7 | 56.32 |
46
  | P-tuning | 88.3 | 56 | 54.2 | **57.6** | 41.9 | 60.9 | 59.3 | **62.9** | 58.1 | 59.91 |
47
  | EFL | 84.9 | 45 | 52.1 | 42.7 | 66.2 | 71.8 | 30.9 | 56.6 | 53 | 55.91 |
48
+ | [UniMC-110M](https://huggingface.co/IDEA-CCNL/Erlangshen-UniMC-RoBERTa-110M-Chinese) | 88.64 | 54.08 | 54.32 | 48.6 | 66.55 | 73.76 | 67.71 | 52.54 | 59.92 | 62.86 |
49
+ | [UniMC-330M](https://huggingface.co/IDEA-CCNL/Erlangshen-UniMC-RoBERTa-330M-Chinese) | 89.53 | 57.3 | 54.25 | 50 | 70.59 | 77.49 | 78.09 | 55.73 | 65.16 | 66.46 |
50
+ | [UniMC-1.3B](https://huggingface.co/IDEA-CCNL/Erlangshen-UniMC-MegatronBERT-1.3B-Chinese) | **89.278** | **60.9** | **57.46** | 52.89 | **76.33** | **80.37** | **90.33** | 61.73 | **79.15** | **72.05** |
51
 
52
  **Zero-shot**
53
 
 
59
  | ZeroPrompt | - | - | - | 16.14 | 46.16 | - | - | - | 47.98 | - |
60
  | Yuan1.0-13B | 88.13 | 38.99 | 57.47 | 38.82 | 48.13 | 59.38 | 86.14 | 50 | 38.99 | 56.22 |
61
  | ERNIE3.0-240B | 88.75 | **50.97** | **57.83** | **40.42** | 53.57 | 64.38 | 87.13 | 56.25 | 53.46 | 61.41 |
62
+ | [UniMC-110M](https://huggingface.co/IDEA-CCNL/Erlangshen-UniMC-RoBERTa-110M-Chinese) | 86.16 | 31.26 | 46.61 | 26.54 | 66.91 | 73.34 | 66.68 | 50.09 | 53.66 | 55.7 |
63
+ | [UniMC-330M](https://huggingface.co/IDEA-CCNL/Erlangshen-UniMC-RoBERTa-330M-Chinese) | 87.5 | 30.4 | 47.6 | 31.5 | 69.9 | 75.9 | 78.17 | 49.5 | 60.55 | 59.01 |
64
+ | [UniMC-1.3B](https://huggingface.co/IDEA-CCNL/Erlangshen-UniMC-MegatronBERT-1.3B-Chinese) | **88.79** | 42.06 | 55.21 | 33.93 | **75.57** | **79.5** | **89.4** | 50.25 | **66.67** | **64.53** |
65
 
66
 
67
 
 
75
 
76
  ```python3
77
  import argparse
78
+ from fengshen.pipelines.multiplechoice import UniMCPiplines
79
 
80
 
81
  total_parser = argparse.ArgumentParser("TASK NAME")
82
  total_parser = UniMCPiplines.piplines_args(total_parser)
83
  args = total_parser.parse_args()
84
+ args.pretrained_model_path = 'IDEA-CCNL/Erlangshen-UniMC-RoBERTa-330M-Chinese'
85
  args.learning_rate=2e-5
86
  args.max_length=512
87
  args.max_epochs=3