harshalmore31
5 days ago
committed by
GitHub
commit
61fa99aa71
@ -0,0 +1,132 @@
|
||||
# Mixture of Agents Example
|
||||
|
||||
The Mixture of Agents (MoA) is a sophisticated multi-agent architecture that implements parallel processing with iterative refinement. This approach processes multiple specialized agents simultaneously, concatenates their outputs, and then performs multiple parallel runs to achieve consensus or enhanced results.
|
||||
|
||||
## How It Works
|
||||
|
||||
1. **Parallel Processing**: Multiple agents work simultaneously on the same input
|
||||
2. **Output Concatenation**: Results from all agents are combined into a unified response
|
||||
3. **Iterative Refinement**: The process repeats for `n` layers/iterations to improve quality
|
||||
4. **Consensus Building**: Multiple runs help achieve more reliable and comprehensive outputs
|
||||
|
||||
This architecture is particularly effective for complex tasks that benefit from diverse perspectives and iterative improvement, such as financial analysis, risk assessment, and multi-faceted problem solving.
|
||||
|
||||
|
||||
|
||||
|
||||
## Installation
|
||||
|
||||
Install the swarms package using pip:
|
||||
|
||||
```bash
|
||||
pip install -U swarms
|
||||
```
|
||||
|
||||
## Basic Setup
|
||||
|
||||
1. First, set up your environment variables:
|
||||
|
||||
```python
|
||||
WORKSPACE_DIR="agent_workspace"
|
||||
ANTHROPIC_API_KEY=""
|
||||
```
|
||||
|
||||
## Code
|
||||
|
||||
```python
|
||||
from swarms import Agent, MixtureOfAgents
|
||||
|
||||
# Agent 1: Risk Metrics Calculator
|
||||
risk_metrics_agent = Agent(
|
||||
agent_name="Risk-Metrics-Calculator",
|
||||
agent_description="Calculates key risk metrics like VaR, Sharpe ratio, and volatility",
|
||||
system_prompt="""You are a risk metrics specialist. Calculate and explain:
|
||||
- Value at Risk (VaR)
|
||||
- Sharpe ratio
|
||||
- Volatility
|
||||
- Maximum drawdown
|
||||
- Beta coefficient
|
||||
|
||||
Provide clear, numerical results with brief explanations.""",
|
||||
max_loops=1,
|
||||
# model_name="gpt-4o-mini",
|
||||
random_model_enabled=True,
|
||||
dynamic_temperature_enabled=True,
|
||||
output_type="str-all-except-first",
|
||||
max_tokens=4096,
|
||||
)
|
||||
|
||||
# Agent 2: Portfolio Risk Analyzer
|
||||
portfolio_risk_agent = Agent(
|
||||
agent_name="Portfolio-Risk-Analyzer",
|
||||
agent_description="Analyzes portfolio diversification and concentration risk",
|
||||
system_prompt="""You are a portfolio risk analyst. Focus on:
|
||||
- Portfolio diversification analysis
|
||||
- Concentration risk assessment
|
||||
- Correlation analysis
|
||||
- Sector/asset allocation risk
|
||||
- Liquidity risk evaluation
|
||||
|
||||
Provide actionable insights for risk reduction.""",
|
||||
max_loops=1,
|
||||
# model_name="gpt-4o-mini",
|
||||
random_model_enabled=True,
|
||||
dynamic_temperature_enabled=True,
|
||||
output_type="str-all-except-first",
|
||||
max_tokens=4096,
|
||||
)
|
||||
|
||||
# Agent 3: Market Risk Monitor
|
||||
market_risk_agent = Agent(
|
||||
agent_name="Market-Risk-Monitor",
|
||||
agent_description="Monitors market conditions and identifies risk factors",
|
||||
system_prompt="""You are a market risk monitor. Identify and assess:
|
||||
- Market volatility trends
|
||||
- Economic risk factors
|
||||
- Geopolitical risks
|
||||
- Interest rate risks
|
||||
- Currency risks
|
||||
|
||||
Provide current risk alerts and trends.""",
|
||||
max_loops=1,
|
||||
# model_name="gpt-4o-mini",
|
||||
random_model_enabled=True,
|
||||
dynamic_temperature_enabled=True,
|
||||
output_type="str-all-except-first",
|
||||
max_tokens=4096,
|
||||
)
|
||||
|
||||
|
||||
swarm = MixtureOfAgents(
|
||||
agents=[
|
||||
risk_metrics_agent,
|
||||
portfolio_risk_agent,
|
||||
market_risk_agent,
|
||||
],
|
||||
layers=1,
|
||||
max_loops=1,
|
||||
output_type="final",
|
||||
)
|
||||
|
||||
|
||||
out = swarm.run(
|
||||
"Calculate VaR and Sharpe ratio for a portfolio with 15% annual return and 20% volatility"
|
||||
)
|
||||
|
||||
print(out)
|
||||
```
|
||||
|
||||
## Support and Community
|
||||
|
||||
If you're facing issues or want to learn more, check out the following resources to join our Discord, stay updated on Twitter, and watch tutorials on YouTube!
|
||||
|
||||
| Platform | Link | Description |
|
||||
|----------|------|-------------|
|
||||
| 📚 Documentation | [docs.swarms.world](https://docs.swarms.world) | Official documentation and guides |
|
||||
| 📝 Blog | [Medium](https://medium.com/@kyeg) | Latest updates and technical articles |
|
||||
| 💬 Discord | [Join Discord](https://discord.gg/jM3Z6M9uMq) | Live chat and community support |
|
||||
| 🐦 Twitter | [@kyegomez](https://twitter.com/kyegomez) | Latest news and announcements |
|
||||
| 👥 LinkedIn | [The Swarm Corporation](https://www.linkedin.com/company/the-swarm-corporation) | Professional network and updates |
|
||||
| 📺 YouTube | [Swarms Channel](https://www.youtube.com/channel/UC9yXyitkbU_WSy7bd_41SqQ) | Tutorials and demos |
|
||||
| 🎫 Events | [Sign up here](https://lu.ma/5p2jnc2v) | Join our community events |
|
||||
|
||||
@ -0,0 +1,138 @@
|
||||
# Agents with Vision and Tool Usage
|
||||
|
||||
This tutorial demonstrates how to create intelligent agents that can analyze images and use custom tools to perform specific actions based on their visual observations. You'll learn to build a quality control agent that can process images, identify potential security concerns, and automatically trigger appropriate responses using function calling capabilities.
|
||||
|
||||
## What You'll Learn
|
||||
|
||||
- How to configure an agent with multi-modal capabilities for image analysis
|
||||
- How to integrate custom tools and functions with vision-enabled agents
|
||||
- How to implement automated security analysis based on visual observations
|
||||
- How to use function calling to trigger specific actions from image analysis results
|
||||
- Best practices for building production-ready vision agents with tool integration
|
||||
|
||||
## Use Cases
|
||||
|
||||
This approach is perfect for:
|
||||
|
||||
- **Quality Control Systems**: Automated inspection of manufacturing processes
|
||||
|
||||
- **Security Monitoring**: Real-time threat detection and response
|
||||
|
||||
- **Object Detection**: Identifying and categorizing items in images
|
||||
|
||||
- **Compliance Checking**: Ensuring standards are met in various environments
|
||||
|
||||
- **Automated Reporting**: Generating detailed analysis reports from visual data
|
||||
|
||||
## Installation
|
||||
|
||||
Install the swarms package using pip:
|
||||
|
||||
```bash
|
||||
pip install -U swarms
|
||||
```
|
||||
|
||||
## Basic Setup
|
||||
|
||||
1. First, set up your environment variables:
|
||||
|
||||
```python
|
||||
WORKSPACE_DIR="agent_workspace"
|
||||
OPENAI_API_KEY=""
|
||||
```
|
||||
|
||||
|
||||
## Code
|
||||
|
||||
- Create tools for your agent as a function with types and documentation
|
||||
|
||||
- Pass tools to your agent `Agent(tools=[list_of_callables])`
|
||||
|
||||
- Add your image path to the run method like: `Agent().run(task=task, img=img)`
|
||||
|
||||
```python
|
||||
from swarms.structs import Agent
|
||||
from swarms.prompts.logistics import (
|
||||
Quality_Control_Agent_Prompt,
|
||||
)
|
||||
|
||||
|
||||
# Image for analysis
|
||||
factory_image = "image.jpg"
|
||||
|
||||
|
||||
def security_analysis(danger_level: str) -> str:
|
||||
"""
|
||||
Analyzes the security danger level and returns an appropriate response.
|
||||
|
||||
Args:
|
||||
danger_level (str, optional): The level of danger to analyze.
|
||||
Can be "low", "medium", "high", or None. Defaults to None.
|
||||
|
||||
Returns:
|
||||
str: A string describing the danger level assessment.
|
||||
- "No danger level provided" if danger_level is None
|
||||
- "No danger" if danger_level is "low"
|
||||
- "Medium danger" if danger_level is "medium"
|
||||
- "High danger" if danger_level is "high"
|
||||
- "Unknown danger level" for any other value
|
||||
"""
|
||||
if danger_level is None:
|
||||
return "No danger level provided"
|
||||
|
||||
if danger_level == "low":
|
||||
return "No danger"
|
||||
|
||||
if danger_level == "medium":
|
||||
return "Medium danger"
|
||||
|
||||
if danger_level == "high":
|
||||
return "High danger"
|
||||
|
||||
return "Unknown danger level"
|
||||
|
||||
|
||||
custom_system_prompt = f"""
|
||||
{Quality_Control_Agent_Prompt}
|
||||
|
||||
You have access to tools that can help you with your analysis. When you need to perform a security analysis, you MUST use the security_analysis function with an appropriate danger level (low, medium, or high) based on your observations.
|
||||
|
||||
Always use the available tools when they are relevant to the task. If you determine there is any level of danger or security concern, call the security_analysis function with the appropriate danger level.
|
||||
"""
|
||||
|
||||
# Quality control agent
|
||||
quality_control_agent = Agent(
|
||||
agent_name="Quality Control Agent",
|
||||
agent_description="A quality control agent that analyzes images and provides a detailed report on the quality of the product in the image.",
|
||||
# model_name="anthropic/claude-3-opus-20240229",
|
||||
model_name="gpt-4o-mini",
|
||||
system_prompt=custom_system_prompt,
|
||||
multi_modal=True,
|
||||
max_loops=1,
|
||||
output_type="str-all-except-first",
|
||||
# tools_list_dictionary=[schema],
|
||||
tools=[security_analysis],
|
||||
)
|
||||
|
||||
|
||||
response = quality_control_agent.run(
|
||||
task="Analyze the image and then perform a security analysis. Based on what you see in the image, determine if there is a low, medium, or high danger level and call the security_analysis function with that danger level",
|
||||
img=factory_image,
|
||||
)
|
||||
```
|
||||
|
||||
|
||||
## Support and Community
|
||||
|
||||
If you're facing issues or want to learn more, check out the following resources to join our Discord, stay updated on Twitter, and watch tutorials on YouTube!
|
||||
|
||||
| Platform | Link | Description |
|
||||
|----------|------|-------------|
|
||||
| 📚 Documentation | [docs.swarms.world](https://docs.swarms.world) | Official documentation and guides |
|
||||
| 📝 Blog | [Medium](https://medium.com/@kyeg) | Latest updates and technical articles |
|
||||
| 💬 Discord | [Join Discord](https://discord.gg/jM3Z6M9uMq) | Live chat and community support |
|
||||
| 🐦 Twitter | [@kyegomez](https://twitter.com/kyegomez) | Latest news and announcements |
|
||||
| 👥 LinkedIn | [The Swarm Corporation](https://www.linkedin.com/company/the-swarm-corporation) | Professional network and updates |
|
||||
| 📺 YouTube | [Swarms Channel](https://www.youtube.com/channel/UC9yXyitkbU_WSy7bd_41SqQ) | Tutorials and demos |
|
||||
| 🎫 Events | [Sign up here](https://lu.ma/5p2jnc2v) | Join our community events |
|
||||
|
||||
@ -0,0 +1,80 @@
|
||||
from swarms import Agent, MixtureOfAgents
|
||||
|
||||
# Agent 1: Risk Metrics Calculator
|
||||
risk_metrics_agent = Agent(
|
||||
agent_name="Risk-Metrics-Calculator",
|
||||
agent_description="Calculates key risk metrics like VaR, Sharpe ratio, and volatility",
|
||||
system_prompt="""You are a risk metrics specialist. Calculate and explain:
|
||||
- Value at Risk (VaR)
|
||||
- Sharpe ratio
|
||||
- Volatility
|
||||
- Maximum drawdown
|
||||
- Beta coefficient
|
||||
|
||||
Provide clear, numerical results with brief explanations.""",
|
||||
max_loops=1,
|
||||
# model_name="gpt-4o-mini",
|
||||
random_model_enabled=True,
|
||||
dynamic_temperature_enabled=True,
|
||||
output_type="str-all-except-first",
|
||||
max_tokens=4096,
|
||||
)
|
||||
|
||||
# Agent 2: Portfolio Risk Analyzer
|
||||
portfolio_risk_agent = Agent(
|
||||
agent_name="Portfolio-Risk-Analyzer",
|
||||
agent_description="Analyzes portfolio diversification and concentration risk",
|
||||
system_prompt="""You are a portfolio risk analyst. Focus on:
|
||||
- Portfolio diversification analysis
|
||||
- Concentration risk assessment
|
||||
- Correlation analysis
|
||||
- Sector/asset allocation risk
|
||||
- Liquidity risk evaluation
|
||||
|
||||
Provide actionable insights for risk reduction.""",
|
||||
max_loops=1,
|
||||
# model_name="gpt-4o-mini",
|
||||
random_model_enabled=True,
|
||||
dynamic_temperature_enabled=True,
|
||||
output_type="str-all-except-first",
|
||||
max_tokens=4096,
|
||||
)
|
||||
|
||||
# Agent 3: Market Risk Monitor
|
||||
market_risk_agent = Agent(
|
||||
agent_name="Market-Risk-Monitor",
|
||||
agent_description="Monitors market conditions and identifies risk factors",
|
||||
system_prompt="""You are a market risk monitor. Identify and assess:
|
||||
- Market volatility trends
|
||||
- Economic risk factors
|
||||
- Geopolitical risks
|
||||
- Interest rate risks
|
||||
- Currency risks
|
||||
|
||||
Provide current risk alerts and trends.""",
|
||||
max_loops=1,
|
||||
# model_name="gpt-4o-mini",
|
||||
random_model_enabled=True,
|
||||
dynamic_temperature_enabled=True,
|
||||
output_type="str-all-except-first",
|
||||
max_tokens=4096,
|
||||
)
|
||||
|
||||
|
||||
swarm = MixtureOfAgents(
|
||||
agents=[
|
||||
risk_metrics_agent,
|
||||
portfolio_risk_agent,
|
||||
market_risk_agent,
|
||||
],
|
||||
layers=1,
|
||||
max_loops=1,
|
||||
output_type="final",
|
||||
)
|
||||
|
||||
|
||||
out = swarm.run(
|
||||
"Calculate VaR and Sharpe ratio for a portfolio with 15% annual return and 20% volatility"
|
||||
)
|
||||
|
||||
print(out)
|
||||
Loading…
Reference in new issue