Shopping Assistant is a chatbot, which can assist consumers in deciding the right product and bridge the gap between online and offline shopping.
- Suggests products to the consumer depending upon his needs, just like a salesperson.
- Helps the consumer to virtually experience fashion products. E.g. If a consumer needs to try a T-shirt or a spectacle our shopping assistant gives him real time experience of how that product would look on him/her.
- Provides a summary of all the reviews about a product, which prevents users from doing the tedious job of going through hundreds of reviews of that product.
- Clone the repository by using the below command:
git clone https://github.com/cjchirag7/shopping-assistant
- Download the 4 model files from here and copy them all to the directory
shopping-assistant/services/api/
. Please note that if you download themodel_files
folder in .zip format, then you need to extract all the files out of it and then copy them to the above directory.
Note: If you are using Windows, make sure Docker Desktop is running.
- Make sure you are in the root of the project (i.e.,
./shopping-assistant/
folder). - Run
docker-compose up
to spin up the containers. If you are using Linux or Mac, you may need to usesudo
for this command to work. web-app
would then be available locally at http://localhost:3000 ,server
at http://localhost:8000 and theAPI documentation
would be available at http://localhost:8000/redoc
- If you don't already have
pipenv
installed, install it using the following commands:
pip install --upgrade setuptools wheel
pip install --user pipenv
- Activate the virtual environment in the
api
folder by using the following command:
cd services/api
pipenv shell
- In the activated virtual environment, run the following command to install all the dependencies:
pipenv install
- In the activated virtual environment, run the following command to run the API:
uvicorn main:app
- The
server
would run at http://127.0.0.1:8000/ and theAPI documentation
would be available at http://127.0.0.1:8000/docs
- Make sure you have
node
installed with version >= 14. Check using following command:
node -v
- In the
web-app
folder, install all the dependencies using the following command:
cd services/web-app
npm install
- In the web-app folder, run the React App using:
npm start
- The web app would start running at http://localhost:3000
- Available at https://www.youtube.com/watch?v=x_BFtcoaTks
- Screenshots have been attached in
screenshots/
- Training files have been attached as
training-files/
- Server code has been attached in
services/api/
- Frontend code has been attached in
services/web-app/