Frontida is a cutting-edge chatbot initiative designed to empower new mothers navigating the complexities of postpartum depression (PPD). By leveraging sophisticated natural language processing and tailored video recommendation technologies, Frontida delivers empathetic, real-time support and resources aimed at alleviating the impacts of PPD. This project contributes to the UN Sustainable Development Goals, emphasizing Good Health & Well-being, Gender Equality, and Industry, Innovation, & Infrastructure.

Follow these instructions to set up the Frontida project on your local machine for development and testing purposes.

Ensure you have the following prerequisites installed:

• Node.js
• Python 3.8 or higher
• A Firebase account
• Access to Hugging Face and Render.com for deployment purposes

git clone https://github.com/Mama-Frontida/FRONTIDA_FRONTEND.git frontend
git clone https://github.com/Mama-Frontida/APIs.git backend

cd backend
pip install -r requirements.txt
python app.py  # Runs the Flask server

cd frontend  # Assuming you're in the root directory
npm install   # Install dependencies
npm run dev   # Start the development server

To get started with the T5-Small model:

git clone https://github.com/Mama-Frontida/text-generation-models.git model-repo
cd model-repo  # Navigate to the model repository
# Follow the Jupyter Notebook instructions within the repo

For cloning the production repository from our Hugging Face account:

git clone https://huggingface.co/Danroy/MamaFrontida

Ensure you follow the notebook instructions for running the model and generating text. Install any necessary dependencies as specified within the notebook or the repository's documentation.

Set up your Firebase project and ensure the firebase.jsx file in the frontend directory is updated with your Firebase configuration keys.

• Frontend: React (JavaScript)
• Backend: Python, Flask
• AI Model: T5-small model, fine-tuned for PPD support
• Database & Authentication: Firebase
• Deployment: Vercel (Frontend), Render.com (Backend APIs), Hugging Face (Model)

After initiating the frontend and backend servers, open http://localhost:5173 to access the Frontida web application. Users can engage with the chatbot for support and video recommendations tailored to their needs.

For immediate access, visit our live platform at https://frontida.vercel.app.

Your contributions to Frontida are highly appreciated. Whether you're improving the application or reporting bugs, please feel free to open an issue or submit a pull request.

Frontida is made available under the MIT License. For more details, see the LICENSE.md file.

We extend our gratitude to all individuals and organizations whose contributions have made Frontida a reality. Special thanks to our dedicated team members for their relentless effort and innovation:

• Danroy Mwangi - Team Lead & NLP Expert
• Maria Muthiore - Backend Development Lead
• Nelson Kamau - Frontend Development Lead

Together, we continue to strive towards making a significant impact on the well-being of new mothers worldwide.