Workforce 是一个 CAMEL 中的一个多智能体系统。在这个系统中，多个智能体将通过协作，解决给定的任务。本示例中，我们将使用 Workforce 创建一个黑客松评审团，让具有不同性格的评审一起合作，为黑客松项目打分。
## 安装依赖
首先安装 `camel-ai` ：
```python
%pip install "camel-ai[all]==0.2.15a0"
```
**如果在 Colab 的环境中运行**，在此我们需要进行特定的处理，以使其支持协程调用。在一般的环境下（如本地环境），此步骤是可以省略的。
```python
%pip install nest_asyncio
import nest_asyncio
nest_asyncio.apply()
```
## API Key 配置
在这个示例中，我们将使用与网页搜索相关的工具。因此，我们需要提前配置一些对应的API Key和环境变量。
```python
from getpass import getpass
import os
os.environ["QWEN_API_KEY"] = getpass("Please input your Qwen API key: ")
# https://developers.google.com/custom-search/v1/overview
os.environ["GOOGLE_API_KEY"] = getpass("Please input your Google API key: ")
# https://cse.google.com/cse/all
os.environ["SEARCH_ENGINE_ID"] = getpass("Please input your Search Egine ID: ")
```
## 创建评审的辅助函数
在此示例中，我们将创建多个具有不同性格和评分标准的评审智能体。出于便利性的考虑，我们首先创建一个可以系统化生成评审的函数。
```python
import textwrap
from camel.agents import ChatAgent
from camel.configs import QwenConfig
from camel.messages import BaseMessage
from camel.models import ModelFactory
from camel.tasks import Task
from camel.toolkits import FunctionTool, SearchToolkit, WeatherToolkit, GoogleMapsToolkit
from camel.types import ModelPlatformType, ModelType
from camel.societies.workforce import Workforce
def make_judge(
persona: str,
example_feedback: str,
criteria: str,
) -> ChatAgent:
msg_content = textwrap.dedent(
f"""\
You are a judge in a hackathon.
This is your persona that you MUST act with: {persona}
Here is an example feedback that you might give with your persona, you MUST try your best to align with this:
{example_feedback}
When evaluating projects, you must use the following criteria:
{criteria}
You also need to give scores based on these criteria, from 1-4. The score given should be like 3/4, 2/4, etc.
""" # noqa: E501
)
sys_msg = BaseMessage.make_assistant_message(
role_name="Hackathon Judge",
content=msg_content,
)
model = ModelFactory.create(
model_platform=ModelPlatformType.QWEN,
model_type=ModelType.QWEN_TURBO,
model_config_dict=QwenConfig(temperature=0.2).as_dict(),
)
agent = ChatAgent(
system_message="system message",
model=model,
)
response = agent.step("")
return agent
```
## 创建一个模拟的黑客松项目
接下来，我们将创建一个模拟的黑客松项目描述。稍后，评审团将根据这段描述对项目进行评分。
```python
proj_content = textwrap.dedent(
"""\
Project name: CAMEL-Powered Adaptive Learning Assistant
How does your project address a real problem: Our CAMEL-Powered Adaptive Learning Assistant addresses the challenge of personalized education in an increasingly diverse and fast-paced learning environment. Traditional one-size-fits-all approaches to education often fail to meet the unique needs of individual learners, leading to gaps in understanding and reduced engagement. Our project leverages CAMEL-AI's advanced capabilities to create a highly adaptive, intelligent tutoring system that can understand and respond to each student's learning style, pace, and knowledge gaps in real-time.
Explain your tech and which parts work: Our system utilizes CAMEL-AI's in-context learning and multi-domain application features to create a versatile learning assistant. The core components include:
1. Learner Profile Analysis: Uses natural language processing to assess the student's current knowledge, learning preferences, and goals.
2. Dynamic Content Generation: Leverages CAMEL-AI to create personalized learning materials, explanations, and practice questions tailored to each student's needs.
3. Adaptive Feedback Loop: Continuously analyzes student responses and adjusts the difficulty and style of content in real-time.
4. Multi-Modal Integration: Incorporates text, images, and interactive elements to cater to different learning styles.
5. Progress Tracking: Provides detailed insights into the student's learning journey, identifying strengths and areas for improvement.
Currently, we have successfully implemented the Learner Profile Analysis and Dynamic Content Generation modules. The Adaptive Feedback Loop is partially functional, while the Multi-Modal Integration and Progress Tracking features are still in development.
""" # noqa: E501
)
```
## 创建智能体
接下来，我们将创建五个独特的智能体，这些智能体将一起协作完成任务。其中之一的智能体是助手，负责收集信息并总结最终结果。我们为这个智能体添加了搜索功能，以便它能够从在线内容中实时获取信息。
另外四个智能体则是具有不同人格和评分标准的评审智能体。它们将根据项目描述以及助手收集的信息对项目进行评分。
```python
# 创建助手智能体
search_toolkit = SearchToolkit()
search_tools = [
FunctionTool(search_toolkit.search_google),
FunctionTool(search_toolkit.search_duckduckgo),
]
researcher_model = ModelFactory.create(
model_platform=ModelPlatformType.QWEN,
model_type=ModelType.QWEN_TURBO,
model_config_dict=QwenConfig(temperature=0.2).as_dict(),
)
researcher_agent = ChatAgent(
system_message=BaseMessage.make_assistant_message(
role_name="Researcher",
content="You are a researcher who does research on AI and Open"
"Sourced projects. You use web search to stay updated on the "
"latest innovations and trends.",
),
model=researcher_model,
tools=search_tools,
)
# 创建评审 - 投资人
vc_persona = (
'You are a venture capitalist who is obsessed with how projects can '
'be scaled into "unicorn" companies. You peppers your speech with '
'buzzwords like "disruptive," "synergistic," and "market penetration."'
' You do not concerned with technical details or innovation unless '
'it directly impacts the business model.'
)
vc_example_feedback = (
'"Wow, this project is absolutely disruptive in the blockchain-enabled'
' marketplace! I can definitely see synergistic applications in the '
'FinTech ecosystem. The scalability is through the roof--this is '
'revolutionary!'
)
vc_criteria = textwrap.dedent(
"""\
### **Applicability to Real-World Usage (1-4 points)**
- **4**: The project directly addresses a significant real-world problem with a clear, scalable application.
- **3**: The solution is relevant to real-world challenges but requires more refinement for practical or widespread use.
- **2**: Some applicability to real-world issues, but the solution is not immediately practical or scalable.
- **1**: Little or no relevance to real-world problems, requiring substantial changes for practical use.
""" # noqa: E501
)
vc_agent = make_judge(
vc_persona,
vc_example_feedback,
vc_criteria,
)
# 创建评审 - 代码工程师
eng_persona = (
'You are an experienced engineer and a perfectionist. You are highly '
'detail-oriented and critical of any technical flaw, no matter how '
'small. He evaluates every project as though it were going into a '
'mission-critical system tomorrow, so his feedback is thorough but '
'often harsh.'
)
eng_example_feedback = (
'There are serious code inefficiencies in this project. The '
'architecture is unstable, and the memory management is suboptimal. '
'I expect near-perfect performance, but this solution barely functions'
' under stress tests. It has potential, but it is nowhere near '
'deployment-ready.'
)
eng_criteria = textwrap.dedent(
"""\
### **Technical Implementation (1-4 points)**
- **4**: Flawless technical execution with sophisticated design, efficient performance, and robust architecture.
- **3**: Strong technical implementation, though there may be areas for improvement or further development.
- **2**: The project works, but technical limitations or inefficiencies hinder its overall performance.
- **1**: Poor technical implementation with major issues in functionality, coding, or structure.
""" # noqa: E501
)
eng_agent = make_judge(
eng_persona,
eng_example_feedback,
eng_criteria,
)
# 创建评审 - AI创业者
founder_persona = (
'You are a well-known AI startup founder who is always looking for the'
' "next big thing" in AI. You value bold, inventive ideas and '
'prioritizes projects that break new ground over those that improve '
'existing systems.'
)
founder_example_feedback = (
'This is interesting, but I have seen similar approaches before. I am '
'looking for something that pushes boundaries and challenges norms. '
'What is the most revolutionary part of this project? Let us see what '
'is trending on Internet to make sure this is not already out there!'
)
founder_criteria = textwrap.dedent(
"""\
### **Innovation (1-4 points)**
- **4**: The project showcases a groundbreaking concept or a unique approach that significantly departs from existing methods.
- **3**: The project demonstrates a novel twist on known solutions or introduces some innovative aspects.
- **2**: Some level of innovation is present, but the project largely builds on existing ideas without major new contributions.
- **1**: Little or no innovation; the project is based on standard approaches with minimal creativity.
""" # noqa: E501
)
founder_agent = make_judge(
founder_persona,
founder_example_feedback,
founder_criteria,
)
# 创建评审 - CAMEL社区贡献者
contributor_persona = (
'You are a contributor to the CAMEL-AI project and is always excited '
'to see how people are using it. You are kind and optimistic, always '
'offering positive feedback, even for projects that are still rough '
'around the edges.'
)
contributor_example_feedback = (
'Oh, I love how you have implemented CAMEL-AI here! The use of its '
'adaptive learning capabilities is fantastic, and you have really '
'leveraged the contextual reasoning in a great way! Let me just pull '
'up the GitHub README to check if there is any more potential '
'optimizations.'
)
contributor_criteria = textwrap.dedent(
"""\
### **Use of CAMEL-AI (1-4 points)**
- **4**: Excellent integration of CAMEL-AI, fully leveraging its advanced features like in-context learning, adaptability, or multi-domain applications.
- **3**: Good use of CAMEL-AI, but there are opportunities to exploit more of its advanced capabilities.
- **2**: Limited use of CAMEL-AI, relying mostly on basic features without taking advantage of its full potential.
- **1**: CAMEL-AI integration is minimal or poorly implemented, adding little value to the project.
""" # noqa: E501
)
contributor_agent = make_judge(
contributor_persona,
contributor_example_feedback,
contributor_criteria,
)
```
## 创建 Workforce
接下来，我们将进行演示中最重要的部分：创建一个 Workforce 。这是本示例中最重要的一步。首先，我们可以通过传递一个描述来实例化一个 Workforce 。然后，我们只需调用 `add_single_agent_worker()` 函数，便可将智能体及其描述添加到该 Workforce 中。
请注意，各个智能体的描述在 Workforce 中非常重要，因为它是 Workforce 中负责协调的智能体进行任务分配的重要依据。因此，建议在将智能体添加到 Workforce 时，清晰地标明其责任和能力。
```python
coordinator_model = ModelFactory.create(
model_platform=ModelPlatformType.QWEN,
model_type=ModelType.QWEN_MAX,
model_config_dict=QwenConfig(temperature=0.2).as_dict(),
)
task_agent_model = ModelFactory.create(
model_platform=ModelPlatformType.QWEN,
model_type=ModelType.QWEN_TURBO,
model_config_dict=QwenConfig(temperature=0.2).as_dict(),
)
# Default tools for a new agent
function_list = [
*SearchToolkit().get_tools(),
*WeatherToolkit().get_tools(),
*GoogleMapsToolkit().get_tools(),
]
new_agent_model = ModelFactory.create(
model_platform=ModelPlatformType.QWEN,
model_type=ModelType.QWEN_TURBO,
model_config_dict=QwenConfig(temperature=0.2).as_dict(),
)
workforce = Workforce(
'Hackathon Judges',
coordinator_agent_kwargs={"model": coordinator_model},
task_agent_kwargs={"model": task_agent_model},
new_worker_agent_kwargs={"model": new_agent_model, "tools": function_list},
)
workforce.add_single_agent_worker(
'Visionary Veronica (Judge), a venture capitalist who is '
'obsessed with how projects can be scaled into "unicorn" companies',
worker=vc_agent,
).add_single_agent_worker(
'Critical John (Judge), an experienced engineer and a'
' perfectionist.',
worker=eng_agent,
).add_single_agent_worker(
'Innovator Iris (Judge), a well-known AI startup founder who'
' is always looking for the "next big thing" in AI.',
worker=founder_agent,
).add_single_agent_worker(
'Friendly Frankie (Judge), a contributor to the CAMEL-AI '
'project and is always excited to see how people are using it.',
worker=contributor_agent,
).add_single_agent_worker(
'Researcher Rachel (Helper), a researcher who does online searches to '
'find the latest innovations and trends on AI and Open Sourced '
'projects.',
worker=researcher_agent,
)
```
## 创建一个 Task
Task 是被 Workforce 接受和处理的对象。我们可以通过将任务内容传递给它来初始化一个 Task。任务的内容建议尽可能详细，这将有助于后续的分解和处理。
这里的 `additional_info` 是一个可选属性。当任务有重要的附加信息，并且我们不希望它被轻易改变时，这将会是一个非常有用的选项。无论 Task 如何被分解和处理， Workforce 都会保持 `additional_info` 不变。在这种场景下，它非常适合用来保存项目描述。
> 同时需要注意的是，Task 的 `id` 不是很重要，可以填写任何值（尽管我们建议使用 `"0"`）。这是由于 `Task` 设计中的一些遗留问题，之后会进行修复。
```python
task = Task(
content="Evaluate the hackathon project. First, do some research on "
"the infomation related to the project, then each judge should give a"
" score accordingly. Finally, list the opinions from each judge while"
" preserving the judge's unique identity, along with the score and"
" judge name, and also give a final summary of the opinions.",
additional_info=proj_content,
id="0",
)
```
## 处理任务
最后，我们可以使用 `process_task()` 让 Workforce 处理任务。控制台中将会输出全部过程，并且我们将打印任务的最终结果。
```python
task = workforce.process_task(task)
print(task.result)
```
Worker node 140613824831328 (Researcher Rachel (Helper), a researcher who does online searches to find the latest innovations and trends on AI and Open Sourced projects.) get task 0.0: Subtask 1: Research the project and gather relevant information (Researcher Rachel)
======
Reply from Worker node 140613824831328 (Researcher Rachel (Helper), a researcher who does online searches to find the latest innovations and trends on AI and Open Sourced projects.):
The CAMEL-Powered Adaptive Learning Assistant project aims to address the challenge of personalized education by leveraging CAMEL-AI's advanced capabilities. The system includes several core components such as Learner Profile Analysis, Dynamic Content Generation, Adaptive Feedback Loop, Multi-Modal Integration, and Progress Tracking. Currently, the Learner Profile Analysis and Dynamic Content Generation modules have been successfully implemented, while the Adaptive Feedback Loop is partially functional. The Multi-Modal Integration and Progress Tracking features are still under development.
======
Worker node 140613824823552 (Critical John (Judge), an experienced engineer and a perfectionist.) get task 0.1: Subtask 2: Evaluate the project based on the gathered information and provide a score (Critical John)
======
Reply from Worker node 140613824823552 (Critical John (Judge), an experienced engineer and a perfectionist.):
The CAMEL-Powered Adaptive Learning Assistant demonstrates a strong technical implementation with several sophisticated design aspects. The Learner Profile Analysis and Dynamic Content Generation modules show promising functionality. However, the Adaptive Feedback Loop is only partially functional, and the Multi-Modal Integration and Progress Tracking features are still under development. The architecture is solid, but there are areas for improvement in terms of feature completeness and robustness. Technical efficiency could also be enhanced. Score: 3/4
======
Worker node 140613824832096 (Innovator Iris (Judge), a well-known AI startup founder who is always looking for the "next big thing" in AI.) get task 0.2: Subtask 3: Evaluate the project based on the gathered information and provide a score (Innovator Iris)
======
Reply from Worker node 140613824832096 (Innovator Iris (Judge), a well-known AI startup founder who is always looking for the "next big thing" in AI.):
The CAMEL-Powered Adaptive Learning Assistant project showcases a groundbreaking concept in personalized education by leveraging advanced AI capabilities. The integration of CAMEL-AI's in-context learning and multi-domain application features sets it apart from traditional approaches. While the project demonstrates a strong technical foundation, the partial functionality of the Adaptive Feedback Loop and the ongoing development of Multi-Modal Integration and Progress Tracking indicate room for further innovation. The project's potential to revolutionize personalized learning is significant, yet it requires more comprehensive implementation to fully realize its potential. Score: 3/4.
======
Worker node 140613824828832 (Visionary Veronica (Judge), a venture capitalist who is obsessed with how projects can be scaled into "unicorn" companies) get task 0.3: Subtask 4: Evaluate the project based on the gathered information and provide a score (Visionary Veronica)
======
Reply from Worker node 140613824828832 (Visionary Veronica (Judge), a venture capitalist who is obsessed with how projects can be scaled into "unicorn" companies):
The CAMEL-Powered Adaptive Learning Assistant project is highly disruptive in the realm of personalized education. It addresses a significant real-world problem by offering a highly adaptive, intelligent tutoring system that caters to individual learner needs. The project has a clear, scalable application that can penetrate the educational technology market effectively. The current implementation of Learner Profile Analysis and Dynamic Content Generation shows strong potential, and the partially functional Adaptive Feedback Loop indicates progress towards a fully synergistic solution. The ongoing development of Multi-Modal Integration and Progress Tracking suggests a robust roadmap for future enhancements. Overall, the project has the potential to become a market leader in personalized learning solutions. Score: 3/4.
======
Worker node 140613824831712 (Friendly Frankie (Judge), a contributor to the CAMEL-AI project and is always excited to see how people are using it.) get task 0.4: Subtask 5: Evaluate the project based on the gathered information and provide a score (Friendly Frankie)
======
Reply from Worker node 140613824831712 (Friendly Frankie (Judge), a contributor to the CAMEL-AI project and is always excited to see how people are using it.):
The CAMEL-Powered Adaptive Learning Assistant project is a commendable effort in addressing the challenge of personalized education. The integration of CAMEL-AI's advanced features, such as in-context learning and multi-domain applications, is impressive. The successful implementation of the Learner Profile Analysis and Dynamic Content Generation modules showcases the potential of this system. Although the Adaptive Feedback Loop is partially functional and Multi-Modal Integration and Progress Tracking are still under development, the project has a solid foundation and a clear roadmap for future enhancements. Score: 3/4.
======
Worker node 140613824831712 (Friendly Frankie (Judge), a contributor to the CAMEL-AI project and is always excited to see how people are using it.) get task 0.5: Subtask 6: Compile the scores and opinions from each judge, preserving their unique identities (You or any Helper)
======
Reply from Worker node 140613824831712 (Friendly Frankie (Judge), a contributor to the CAMEL-AI project and is always excited to see how people are using it.):
The CAMEL-Powered Adaptive Learning Assistant project has received consistent scores and evaluations from the judges. Each evaluator acknowledges the project's strong technical foundation and innovative approach to personalized education. Critical John, Innovator Iris, Visionary Veronica, and Friendly Frankie all provided a score of 3/4, highlighting the successful implementation of the Learner Profile Analysis and Dynamic Content Generation modules. They also noted the partial functionality of the Adaptive Feedback Loop and the ongoing development of Multi-Modal Integration and Progress Tracking. The project demonstrates a clear roadmap for future enhancements and significant potential to revolutionize personalized learning.
======
Worker node 140613824831328 (Researcher Rachel (Helper), a researcher who does online searches to find the latest innovations and trends on AI and Open Sourced projects.) get task 0.6: Subtask 7: Write a final summary of the opinions and scores (You or any Helper)
======
Reply from Worker node 140613824831328 (Researcher Rachel (Helper), a researcher who does online searches to find the latest innovations and trends on AI and Open Sourced projects.):
The CAMEL-Powered Adaptive Learning Assistant project has received consistent evaluations from the judges. Each evaluator acknowledged the project's strong technical foundation and innovative approach to personalized education. The project has been scored 3/4 by Critical John, Innovator Iris, Visionary Veronica, and Friendly Frankie. They highlighted the successful implementation of the Learner Profile Analysis and Dynamic Content Generation modules. However, they also noted the partial functionality of the Adaptive Feedback Loop and the ongoing development of Multi-Modal Integration and Progress Tracking. Despite these areas needing further development, the project demonstrates a clear roadmap for future enhancements and significant potential to revolutionize personalized learning.
======
Based on the results from the sub-tasks, here is the final evaluation of the "CAMEL-Powered Adaptive Learning Assistant" project:
### Judge Opinions
**Critical John**
- **Score:** 3/4
- **Opinion:** The project has a strong technical foundation and innovative approach to personalized education. The successful implementation of the Learner Profile Analysis and Dynamic Content Generation modules showcases the potential of this system. However, the partial functionality of the Adaptive Feedback Loop and the ongoing development of Multi-Modal Integration and Progress Tracking indicate areas for further enhancement. The project has a clear roadmap for future improvements and significant potential to revolutionize personalized learning.
**Innovator Iris**
- **Score:** 3/4
- **Opinion:** The CAMEL-Powered Adaptive Learning Assistant project integrates CAMEL-AI's advanced features impressively. The successful implementation of the Learner Profile Analysis and Dynamic Content Generation modules highlights the project's potential. The Adaptive Feedback Loop is partially functional, and Multi-Modal Integration and Progress Tracking are still under development. Despite these areas needing further development, the project has a solid foundation and a clear roadmap for future enhancements.
**Visionary Veronica**
- **Score:** 3/4
- **Opinion:** This project is highly disruptive in the realm of personalized education, addressing a significant real-world problem with a highly adaptive, intelligent tutoring system. The current implementation of Learner Profile Analysis and Dynamic Content Generation shows strong potential, and the partially functional Adaptive Feedback Loop indicates progress towards a fully synergistic solution. The ongoing development of Multi-Modal Integration and Progress Tracking suggests a robust roadmap for future enhancements.
**Friendly Frankie**
- **Score:** 3/4
- **Opinion:** The project is a commendable effort in addressing the challenge of personalized education. The integration of CAMEL-AI's advanced features, such as in-context learning and multi-domain applications, is impressive. The successful implementation of the Learner Profile Analysis and Dynamic Content Generation modules showcases the potential of this system. Although the Adaptive Feedback Loop is partially functional and Multi-Modal Integration and Progress Tracking are still under development, the project has a solid foundation and a clear roadmap for future enhancements.
### Final Summary
Each judge has provided a consistent score of 3/4 for the "CAMEL-Powered Adaptive Learning Assistant" project. The project is recognized for its strong technical foundation and innovative approach to personalized education. The successful implementation of the Learner Profile Analysis and Dynamic Content Generation modules is a significant achievement. However, the partial functionality of the Adaptive Feedback Loop and the ongoing development of Multi-Modal Integration and Progress Tracking indicate areas for further enhancement. Despite these limitations, the project demonstrates a clear roadmap for future improvements and significant potential to revolutionize personalized learning.
## 🌟 精彩亮点
多智能体系统的魅力就在于它的多样性。在这个示例中，我们带你了解了如何设置和运行 CAMEL 的 Workforce 组成一个黑客松评审委员会，展示了多个智能体如何协作解决复杂的任务。你可以很容易地将这个例子扩展到其他需要多样化视角和专业技能的场景，比如使用不同工具的智能体等。
## ⭐ 给仓库加个星吧！
如果你觉得 CAMEL 有用或者有趣，别忘了在 [GitHub](https://github.com/camel-ai/camel) 给它加个星！你的星标不仅能帮助更多人发现这个项目，还能激励我们继续做得更好。
