DeepChoice
AI Shopping Super App for eCommerce
๐ ๏ธ Configuration
-
Download the dataset from Kaggle and place it in the
datadirectory. The dataset contains product images (optional) and metadata. -
Create a
.envfile in the root directory and set the following environment variables:GROQ_API_KEY=your_groq_api_key GOOGLE_API_KEY=your_google_api_key
-
Download the fine-tuned complaints model from the Kaggle
and place it in thecomplaints_modeldirectory at the root level.
๐ณ Docker Compose
-
To run the app using Docker Compose:
docker-compose up --build
- FastAPI backend: http://localhost:8000
- Next.js frontend: http://localhost:3000
โ๏ธ Manual Setup
-
Install the required packages:
pip install -r requirements.txt
-
Run the uvicorn dev server:
uvicorn app.main:app --reload
It will create the embeddings in
data/*.indexand run the backend server onhttp://127.0.0.1:8000. -
Run the Next.js frontend in a separate terminal:
cd frontend npm install npm run devIt will run the frontend server on
http://localhost:3000. -
Open the frontend in your browser: http://localhost:3000
Relevant Resources:
- Products Dataset Drive link (36MB): H&M Personalized Fashion Recommendations
- Product Images - Optional (30.1GB): H&M Product Images
