Ddroidlabs-mixture-of-agents
small agentic model designed as a coding assistant
Mixture of Agents Model (MAM) - Full-Stack Development Team
Overview
The Mixture of Agents Model (MAM) is an AI-driven full-stack development team that integrates specialized agents for front-end development, back-end development, database management, DevOps, and project management. This unified model leverages a pretrained transformer and fine-tuned datasets to handle a variety of software development tasks efficiently.
Folder Structure
mixture_of_agents/
βββ app.py
βββ colab_notebook.ipynb
βββ dataset/
β βββ code_finetune_dataset.json
βββ agents/
β βββ front_end_agent.py
β βββ back_end_agent.py
β βββ database_agent.py
β βββ devops_agent.py
β βββ project_management_agent.py
βββ integration/
β βββ integration_layer.py
βββ model/
βββ load_pretrained_model.py
βββ fine_tune_model.py
Setup Instructions
Prerequisites
- Python 3.7 or higher
- Flask
- Google Colab account (for running the notebook)
- Libraries:
transformers
,datasets
,numpy
,pandas
Installation
Clone the Repository:
git clone https://github.com/your-repo/mixture_of_agents.git cd mixture_of_agents
Install Required Libraries:
pip install -r requirements.txt
Upload to Google Drive:
- Upload the
mixture_of_agents
folder to your Google Drive.
- Upload the
Open Colab Notebook:
- Open
colab_notebook.ipynb
in Google Colab.
- Open
Running the Model
Mount Google Drive:
- Mount your Google Drive in Colab by running the first cell of the notebook:
from google.colab import drive drive.mount('/content/drive')
- Mount your Google Drive in Colab by running the first cell of the notebook:
Install Necessary Packages:
- Install the required packages in the Colab environment:
!pip install transformers datasets
- Install the required packages in the Colab environment:
Load and Fine-Tune the Model:
- Follow the steps in the Colab notebook to load the pretrained model and fine-tune it using the provided dataset:
from model.load_pretrained_model import load_model_and_tokenizer model, tokenizer = load_model_and_tokenizer() from model.fine_tune_model import fine_tune_model fine_tune_model(model, tokenizer, '/content/drive/MyDrive/mixture_of_agents/dataset/code_finetune_dataset.json')
- Follow the steps in the Colab notebook to load the pretrained model and fine-tune it using the provided dataset:
Initialize and Use the Agents:
- Initialize the agents and use the integration layer to process tasks:
from agents.front_end_agent import FrontEndAgent from agents.back_end_agent import BackEndAgent from agents.database_agent import DatabaseAgent from agents.devops_agent import DevOpsAgent from agents.project_management_agent import ProjectManagementAgent from integration.integration_layer import IntegrationLayer front_end_agent = FrontEndAgent(model, tokenizer) back_end_agent = BackEndAgent(model, tokenizer) database_agent = DatabaseAgent(model, tokenizer) devops_agent = DevOpsAgent(model, tokenizer) project_management_agent = ProjectManagementAgent(model, tokenizer) integration_layer = IntegrationLayer(front_end_agent, back_end_agent, database_agent, devops_agent, project_management_agent) task_data = {'task': 'Create a responsive website layout'} result = integration_layer.process_task('front_end', task_data) print(result)
- Initialize the agents and use the integration layer to process tasks:
Running the Web Application
Ensure All Agent Files and Integration Layer Are Available:
- Make sure the
agents
andintegration
directories with their respective Python files (front_end_agent.py
,back_end_agent.py
,database_agent.py
,devops_agent.py
,project_management_agent.py
, andintegration_layer.py
) are in the same directory asapp.py
.
- Make sure the
Run the Application:
- Execute the
app.py
script to start the Flask web server:python app.py
- Execute the
Using the API:
- Open your web browser and navigate to
http://127.0.0.1:5000/
to see the welcome message. - Use a tool like
curl
or Postman to send a POST request to the/process
endpoint with JSON payload to process tasks.
- Open your web browser and navigate to
Example POST Request
You can use the following example JSON payload to test the /process
endpoint:
{
"task_type": "front_end",
"task_data": {
"task": "Create a responsive website layout"
}
}
Using curl
:
curl -X POST http://127.0.0.1:5000/process -H "Content-Type: application/json" -d '{"task_type": "front_end", "task_data": {"task": "Create a responsive website layout"}}'
Agent Descriptions
Front-End Agent
- File:
agents/front_end_agent.py
- Responsibilities: UI/UX design, HTML, CSS, JavaScript frameworks (React, Vue).
Back-End Agent
- File:
agents/back_end_agent.py
- Responsibilities: Server-side logic, API development, frameworks like Node.js, Django.
Database Agent
- File:
agents/database_agent.py
- Responsibilities: Database design, query optimization, data migration.
DevOps Agent
- File:
agents/devops_agent.py
- Responsibilities: CI/CD pipelines, server management, deployment automation.
Project Management Agent
- File:
agents/project_management_agent.py
- Responsibilities: Requirement gathering, task management, progress tracking.
Integration Layer
- File:
integration/integration_layer.py
- Responsibilities: Ensures seamless communication and coordination between agents.
Fine-Tuning Dataset
Dataset File
- File:
dataset/code_finetune_dataset.json
- Description: Contains examples of various coding tasks to fine-tune the model for development-related tasks.
Contributing
Contributions are welcome! Please fork the repository and create a pull request with your changes. Ensure your code follows the project's style guidelines and includes appropriate tests.
License
This project is licensed under the apache-2.0 License.
Contact
For any questions or issues, please open an issue on GitHub or contact the repository maintainer.
- Downloads last month
- 0