Atharva-77/E-Commerce-ImageSearch-PricePredict

MERN stack E-commerce website with Image Search and Price prediction of items seller intends to resell

E-Commerce Website With ImageSearch & PricePredict

This is website in MERN(MongoDB, Express, Reactjs, Nodejs):- https://e-commerce-stores.vercel.app/ .
Flask framework used for image search, price prediction.
(See last line for image search, price prediction)

You can Add/Remove items from Basket, Write Product review.
Unique Feature (which isn't on most e-commerce sites):- User(buyer) can Upload an image of product and the website shows similar products to the user. Seller also have option to Predict Price of a product that the seller intends to resell.

Buyers can Fill shipping address and Place orders too.
Paypal Payment gateway has been integrated for buyers.

Features:-

1. Add/Remove items to a basket
2. Write Product review
3. Image Search Feature. Unique Feature (which isn't on most e-commerce sites)
4. Predict Price of a Product for seller Unique Feature (which isn't on most e-commerce sites)
5. Paypal Payment gateway integrated.
6. Separate screen for sellers to track products.
7. Separate screen for Admins to track products,delete users,etc..

Techstack:-

MERN(MongoDB Express, Reactjs, Nodejs) Stack is used to build the website.

Frontend is made in Reactjs. Redux is used to fetch most of the data from Backend.
Backend is made in Nodejs, express and Mongodb is the database used.

Firebase is used to images of Product. Flask framework used for image search, price prediction.

For Image search, resnet-50 model used. For Price prediction, lasso-regression model used.

Unfortunately you cannot access Machine learning parts(Image search, Price Prediction) as they take a lot of space, hence they couldn't be deployed/ hosted in the free tier of heroku or other such websites. But you can download the file and run on local machine.

---

For Image Search:-

At the header(search bar), click on the downarrow besides the camera logo(as shown below)

Step 1

Upload the picture as shown below.
Step2

How to run project locally?:-

1. Go to backend directory
2. Do npm run install. Server gets started.
3. Come out of backend directory. i.e. come in same level where src folder is.
4. Do npm run install. Frontend gets started.
5. In backend/config_folder create config.env
6. In config.env file, put:-
                          a. PORT                        (Port number where server is to be run)
                          b. MONGO_URI              (MongoDb url of database to connect)
                          c. JWT_SECRET_TOKEN       (Can put any string as the token)
                          d. PAYPAL_CLIENTID         (Paypal client id)