EliteKnightDeveloper / ShopAI

Fashion Shopping AI which converses and gives recommendations (with links)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Themer logo

ShopAI

Your own personalised wardrobe recommender!

Screenshots

Onboarding Search Page Chat Page

Team Members

How it works

Architecture

Dev

Prerequisites

Packages

Configs Required

  • GCP Project Config
    • Make sure Vertex AI and Vision AI is enabled in API & Services
    • Rename the file as gcpconfig.json and place it in the backend directory
  • Create an .env file with the following content
    OPEN_AI_API_KEY=<OPENAI_API_KEY>
    TWILIO_SID=<TWILIO_SID>
    TWILIO_AUTH_TOKEN=<TWILIO_AUTH_TOKEN>
    GCP_ACCESS_TOKEN=<GCP_ACCESS_TOKEN> # gcloud auth print-access-token
    GCP_PROJECT_ID=<GCP_PROJECT_ID> # Copy and Paste from gcpconfig.json
    

Install Dependencies

Note: Create a Virtual Env before running the below commands

  1. Install frontend packages

    cd frontend && yarn && cd ..
  2. Install backend packages

    cd backend && pip install -r requirements.txt && cd ..

Running the app

  1. Scraping the Content (a cron job can be setup if running in production)

    cd backend && python instaScraper.py && cd ..
  2. Analyzing the trend (a cron job can be setup if running in production)

    cd backend && python trendsAnalyzer.py && cd ..
  3. Start the Frontend Server

    cd frontend && yarn build && yarn preview
  4. In second terminal, start the backend server

    cd backend && uvicorn main:app --reload

Running the ChatBot

  1. Make sure the backend server is on. You can turn it on by running the following command in the backend directory

    cd backend && uvicorn main:app --reload
  2. Tunnel the backend through ngrok (in second terminal)

    ngrok http 8000
  3. Change the Sandbox Configuration in Try Whatsapp section of your twilio app. Click here for steps

References

About

Fashion Shopping AI which converses and gives recommendations (with links)


Languages

Language:Python 48.3%Language:TypeScript 37.9%Language:JavaScript 8.2%Language:CSS 5.2%Language:HTML 0.4%