Update README.md
Browse files
README.md
CHANGED
@@ -1,82 +1,82 @@
|
|
1 |
-
---
|
2 |
-
license: cc-by-sa-4.0
|
3 |
-
---
|
4 |
-
|
5 |
-
# Compare-Answer Model
|
6 |
-
|
7 |
-
Welcome to the repository for the Compare-Answer Model, an innovative tool designed to enhance the accuracy and efficiency of mathematical answer comparison tasks. This model leverages advanced techniques to provide precise comparisons across a wide range of mathematical problems.
|
8 |
-
|
9 |
-
## Features
|
10 |
-
|
11 |
-
- **High Accuracy**: Utilizes state-of-the-art technology to ensure high reliability in answer comparison.
|
12 |
-
- **Broad Compatibility**: Supports a variety of mathematical problem types and formats.
|
13 |
-
- **Easy Integration**: Designed for easy integration with existing systems and workflows.
|
14 |
-
|
15 |
-
## Installation
|
16 |
-
|
17 |
-
To get started with the Compare-Answer Model, clone this repository and load model with Transformers.
|
18 |
-
|
19 |
-
# Quick Start
|
20 |
-
To use the model, import it and call the main comparison function with the required parameters:
|
21 |
-
```python
|
22 |
-
from transformers import AutoModelForCausalLM, AutoTokenizer
|
23 |
-
device = "cuda" # the device to load the model onto
|
24 |
-
model = AutoModelForCausalLM.from_pretrained(
|
25 |
-
model_path, torch_dtype="auto", device_map="auto"
|
26 |
-
)
|
27 |
-
tokenizer = AutoTokenizer.from_pretrained(model_path)
|
28 |
-
|
29 |
-
def build_user_query(question, pred_answer, answer, base_prompt):
|
30 |
-
input_text = base_prompt.replace("{{question}}", question)
|
31 |
-
input_text = input_text.replace("{{pred_step}}", pred_answer)
|
32 |
-
input_text = input_text.replace("{{answer}}", answer)
|
33 |
-
input_text = input_text.replace("{{analysis}}", "") # default set analysis to blank, if exist, you can pass in the corresponding parameter.
|
34 |
-
return input_text
|
35 |
-
|
36 |
-
chat_prompt = """<|im_start|>system
|
37 |
-
You are a helpful assistant.<|im_end|>
|
38 |
-
<|im_start|>human
|
39 |
-
{}<|im_end|>
|
40 |
-
<|im_start|>gpt
|
41 |
-
"""
|
42 |
-
|
43 |
-
basic_prompt = """## 任务描述\n\n你是一个数学老师,学生提交了题目的解题步骤,你需要参考`题干`,`解析`和`答案`,判断`学生解题步骤`的结果是否正确。忽略`学生解题步骤`中的错误,只关注最后的答案。答案可能出现在`解析`中,也可能出现在`答案`中。\n\n## 输入内容\n\n题干:\n\n```\n{{question}}\n```\n\n解析:\n\n```\n{{analysis}}\n\n```\n\n答案:\n\n```\n{{answer}}\n```\n\n学生解题步骤:\n\n```\n{{pred_step}}\n```\n\n输出:"""
|
44 |
-
base_prompt = chat_prompt.format(basic_prompt)
|
45 |
-
|
46 |
-
def build_user_query(question, pred_answer, answer, base_prompt):
|
47 |
-
input_text = base_prompt.replace("{{question}}", question)
|
48 |
-
input_text = input_text.replace("{{pred_step}}", pred_answer)
|
49 |
-
input_text = input_text.replace("{{answer}}", answer)
|
50 |
-
input_text = input_text.replace("{{analysis}}", "") # default set analysis to blank, if exist, you can pass in the corresponding parameter.
|
51 |
-
return input_text
|
52 |
-
|
53 |
-
prompt = build_user_query("1+1=", "3", "2", base_prompt)
|
54 |
-
|
55 |
-
model_inputs = tokenizer([prompt], return_tensors="pt").to(device)
|
56 |
-
|
57 |
-
generated_ids = model.generate(model_inputs.input_ids, temperature=0, max_new_tokens=16, eos_token_id=100005)
|
58 |
-
generated_ids = [
|
59 |
-
output_ids[len(input_ids) :]
|
60 |
-
for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
|
61 |
-
]
|
62 |
-
|
63 |
-
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)[0]
|
64 |
-
```
|
65 |
-
|
66 |
-
## Documentation
|
67 |
-
For more detailed information about the model's API and functionalities, please contact us.
|
68 |
-
|
69 |
-
# Contributing
|
70 |
-
Contributions to the Compare-Answer Model are welcome! If you have suggestions or improvements, please fork the repository and submit a pull request.
|
71 |
-
|
72 |
-
# License
|
73 |
-
This project is licensed under the MIT License - see the LICENSE.md file for details.
|
74 |
-
|
75 |
-
# Acknowledgements
|
76 |
-
Thanks to all contributors who have helped in developing this model.
|
77 |
-
Special thanks to MathEval for providing the datasets and challenges that inspired this project.
|
78 |
-
|
79 |
-
# Contact
|
80 |
-
For any inquiries, please reach out via email at liutianqiao1@tal.com or open an issue in this repository.
|
81 |
-
|
82 |
Thank you for using or contributing to the Compare-Answer Model!
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-sa-4.0
|
3 |
+
---
|
4 |
+
|
5 |
+
# Compare-Answer Model
|
6 |
+
|
7 |
+
Welcome to the repository for the Compare-Answer Model, an innovative tool designed to enhance the accuracy and efficiency of mathematical answer comparison tasks. This model leverages advanced techniques to provide precise comparisons across a wide range of mathematical problems.
|
8 |
+
|
9 |
+
## Features
|
10 |
+
|
11 |
+
- **High Accuracy**: Utilizes state-of-the-art technology to ensure high reliability in answer comparison.
|
12 |
+
- **Broad Compatibility**: Supports a variety of mathematical problem types and formats.
|
13 |
+
- **Easy Integration**: Designed for easy integration with existing systems and workflows.
|
14 |
+
|
15 |
+
## Installation
|
16 |
+
|
17 |
+
To get started with the Compare-Answer Model, clone this repository and load model with Transformers.
|
18 |
+
|
19 |
+
# Quick Start
|
20 |
+
To use the model, import it and call the main comparison function with the required parameters:
|
21 |
+
```python
|
22 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
23 |
+
device = "cuda" # the device to load the model onto
|
24 |
+
model = AutoModelForCausalLM.from_pretrained(
|
25 |
+
model_path, torch_dtype="auto", device_map="auto"
|
26 |
+
)
|
27 |
+
tokenizer = AutoTokenizer.from_pretrained(model_path)
|
28 |
+
|
29 |
+
def build_user_query(question, pred_answer, answer, base_prompt):
|
30 |
+
input_text = base_prompt.replace("{{question}}", question)
|
31 |
+
input_text = input_text.replace("{{pred_step}}", pred_answer)
|
32 |
+
input_text = input_text.replace("{{answer}}", answer)
|
33 |
+
input_text = input_text.replace("{{analysis}}", "") # default set analysis to blank, if exist, you can pass in the corresponding parameter.
|
34 |
+
return input_text
|
35 |
+
|
36 |
+
chat_prompt = """<|im_start|>system
|
37 |
+
You are a helpful assistant.<|im_end|>
|
38 |
+
<|im_start|>human
|
39 |
+
{}<|im_end|>
|
40 |
+
<|im_start|>gpt
|
41 |
+
"""
|
42 |
+
|
43 |
+
basic_prompt = """## 任务描述\n\n你是一个数学老师,学生提交了题目的解题步骤,你需要参考`题干`,`解析`和`答案`,判断`学生解题步骤`的结果是否正确。忽略`学生解题步骤`中的错误,只关注最后的答案。答案可能出现在`解析`中,也可能出现在`答案`中。\n\n## 输入内容\n\n题干:\n\n```\n{{question}}\n```\n\n解析:\n\n```\n{{analysis}}\n\n```\n\n答案:\n\n```\n{{answer}}\n```\n\n学生解题步骤:\n\n```\n{{pred_step}}\n```\n\n输出:"""
|
44 |
+
base_prompt = chat_prompt.format(basic_prompt)
|
45 |
+
|
46 |
+
def build_user_query(question, pred_answer, answer, base_prompt):
|
47 |
+
input_text = base_prompt.replace("{{question}}", question)
|
48 |
+
input_text = input_text.replace("{{pred_step}}", pred_answer)
|
49 |
+
input_text = input_text.replace("{{answer}}", answer)
|
50 |
+
input_text = input_text.replace("{{analysis}}", "") # default set analysis to blank, if exist, you can pass in the corresponding parameter.
|
51 |
+
return input_text
|
52 |
+
|
53 |
+
prompt = build_user_query("1+1=", "3", "2", base_prompt)
|
54 |
+
|
55 |
+
model_inputs = tokenizer([prompt], return_tensors="pt").to(device)
|
56 |
+
|
57 |
+
generated_ids = model.generate(model_inputs.input_ids, temperature=0, max_new_tokens=16, eos_token_id=100005)
|
58 |
+
generated_ids = [
|
59 |
+
output_ids[len(input_ids) :]
|
60 |
+
for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
|
61 |
+
]
|
62 |
+
|
63 |
+
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)[0]
|
64 |
+
```
|
65 |
+
|
66 |
+
## Documentation
|
67 |
+
For more detailed information about the model's API and functionalities, please contact us.
|
68 |
+
|
69 |
+
# Contributing
|
70 |
+
Contributions to the Compare-Answer Model are welcome! If you have suggestions or improvements, please fork the repository and submit a pull request.
|
71 |
+
|
72 |
+
# License
|
73 |
+
This project is licensed under the MIT License - see the LICENSE.md file for details.
|
74 |
+
|
75 |
+
# Acknowledgements
|
76 |
+
Thanks to all contributors who have helped in developing this model.
|
77 |
+
Special thanks to MathEval for providing the datasets and challenges that inspired this project.
|
78 |
+
|
79 |
+
# Contact
|
80 |
+
For any inquiries, please reach out via email at liutianqiao1@tal.com or open an issue in this repository.
|
81 |
+
|
82 |
Thank you for using or contributing to the Compare-Answer Model!
|