movierecommendationsystem_streamlit's Introduction

Recommendation System

Developed A python App on Pycharm (app.py) which shows Recommended movies on the basis of movies selected.

There are 3 types of recommendation systems:

Content Based : Recommendations are provided on the basis of Type of content acessed by user.This is done by creating tags.
Collaborative Filtering Based : Recommendations on the basis of common interests shared by multiple users.
Hybrid : Which used both content and collaborative Filtering

I have used Streamlit library to create this app (to run use: streamlit run app.py)
And pickle library to make pickle files to be used to extract data in app.py and data is being uploaded through RecommendationSystem.ipynb

In app.py file

similarity=pickle.load(open('similarity.pkl','rb'))

In RecommendationSystem.ipynb

pickle.dump(similarity,open('similarity.pkl','wb'))

API

I have Used Api from TMDB to extract poster of Movie through movie id from it

Deployment

I have used Heroku to deploy my project :

Heroku requires : requirements.txt file, ProcFile, setup.sh and .gitignore (The code of each is written Down)

In the terminal of pycharm write following lines to deploy the project :

git init
heroku login
heroku create recomendatnsys
git add .
git commit -m "message"
git push heroku master

codes of each file

requirements.txt

write this code in terminal to generate requirements.txt file

pip freeze > requirements.txt

Procfile

web: sh setup.sh && streamlit run app.py

setup.sh

mkdir -p ~/.streamlit/

echo "\
[general]\n\
email = \"[email protected]\"\n\
" > ~/.streamlit/credentials.toml

echo "\
[server]\n\
headless = true\n\
enableCORS=false\n\
port = $PORT\n\
" > ~/.streamlit/config.toml