/BE-Major-Project

Primary LanguageJavaScript

Serving the tensorflow model using Tensorflow-serving in docker

This will serve model as HTTP/REST API at:localhost:8501

docker run -p 8501:8501 --mount type=bind,source=C:\Users\Asus\Desktop\project_practice\model_py,target=/models/my_model  -e MODEL_NAME=my_model -t tensorflow/serving

Create a sqlite3 database and store movies data

In the root folder run the following commands to create database

mkdir database
cd database
touch database.db

Start python shell in sentiment_analysis folder and run

import model
model.db.create_all()

From sentiment_analysis folder run the following command to store movies

python movies.py

Start the flask server

The flask server will serve the site at localhost:5000

cd sentiment_analysis
python index.py

Demo:

The main page consists the rating of each movie,scaled between 1 and 5

Demo1

Demo2

On this page user can enter reviews.The reviews are sent to the ML model and the review is classified as positive or negative.All the reviews for each movie is stored in the database and also the overall rating of movie is also updated accordingly.

Demo3