Commit
•
00276ea
1
Parent(s):
f28bd03
Update README.md
Browse files
README.md
CHANGED
|
@@ -37,6 +37,21 @@ The overall performance of CSC **test**:
|
|
| 37 |
| chinese-text-correction-7b-lora | Qwen/Qwen2.5-7B-Instruct | [🤗 Hugging Face](https://huggingface.co/shibing624/chinese-text-correction-7b-lora) |
|
| 38 |
|
| 39 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
|
| 41 |
## Usage (pycorrector)
|
| 42 |
|
|
|
|
| 37 |
| chinese-text-correction-7b-lora | Qwen/Qwen2.5-7B-Instruct | [🤗 Hugging Face](https://huggingface.co/shibing624/chinese-text-correction-7b-lora) |
|
| 38 |
|
| 39 |
|
| 40 |
+
### 评估结果
|
| 41 |
+
- 评估指标：F1
|
| 42 |
+
- CSC(Chinese Spelling Correction): 拼写纠错模型，表示模型可以处理音似、形似、语法等长度对齐的错误纠正
|
| 43 |
+
- CTC(CHinese Text Correction): 文本纠错模型，表示模型支持拼写、语法等长度对齐的错误纠正，还可以处理多字、少字等长度不对齐的错误纠正
|
| 44 |
+
- GPU：Tesla V100，显存 32 GB
|
| 45 |
+
|
| 46 |
+
| Model Name | Model Link | Base Model | Avg | SIGHAN-2015 | EC-LAW | MCSC | GPU/CPU | QPS |
|
| 47 |
+
|:-----------------|:------------------------------------------------------------------------------------------------------------------------|:---------------------------|:-----------|:------------|:-------|:-------|:--------|:--------|
|
| 48 |
+
| Kenlm-CSC | [shibing624/chinese-kenlm-klm](https://huggingface.co/shibing624/chinese-kenlm-klm) | kenlm | 0.3409 | 0.3147 | 0.3763 | 0.3317 | CPU | 9 |
|
| 49 |
+
| Mengzi-T5-CSC | [shibing624/mengzi-t5-base-chinese-correction](https://huggingface.co/shibing624/mengzi-t5-base-chinese-correction) | mengzi-t5-base | 0.3984 | 0.7758 | 0.3156 | 0.1039 | GPU | 214 |
|
| 50 |
+
| ERNIE-CSC | [PaddleNLP/ernie-csc](https://github.com/PaddlePaddle/PaddleNLP/tree/develop/legacy/examples/text_correction/ernie-csc) | PaddlePaddle/ernie-1.0-base-zh | 0.4353 | 0.8383 | 0.3357 | 0.1318 | GPU | 114 |
|
| 51 |
+
| MacBERT-CSC | [shibing624/macbert4csc-base-chinese](https://huggingface.co/shibing624/macbert4csc-base-chinese) | hfl/chinese-macbert-base | 0.3993 | 0.8314 | 0.1610 | 0.2055 | GPU | **224** |
|
| 52 |
+
| ChatGLM3-6B-CSC | [shibing624/chatglm3-6b-csc-chinese-lora](https://huggingface.co/shibing624/chatglm3-6b-csc-chinese-lora) | THUDM/chatglm3-6b | 0.4538 | 0.6572 | 0.4369 | 0.2672 | GPU | 3 |
|
| 53 |
+
| Qwen2.5-1.5B-CTC | [shibing624/chinese-text-correction-1.5b](https://huggingface.co/shibing624/chinese-text-correction-1.5b) | Qwen/Qwen2.5-1.5B-Instruct | 0.6802 | 0.3032 | 0.7846 | 0.9529 | GPU | 6 |
|
| 54 |
+
| Qwen2.5-7B-CTC | [shibing624/chinese-text-correction-7b](https://huggingface.co/shibing624/chinese-text-correction-7b) | Qwen/Qwen2.5-7B-Instruct | **0.8225** | 0.4917 | 0.9798 | 0.9959 | GPU | 3 |
|
| 55 |
|
| 56 |
## Usage (pycorrector)
|
| 57 |
|