An application that predicts the category of a youtube video using it's decription as an input
Model was trained on data from 6 categories, hence it works well only on them. They are as follows:
- 'Art&Music'
- 'Food'
- 'History'
- 'Science&Technology'
- 'manufacturing'
- 'travel blog'
- To deploy a containerized application for a machine learning model
- to run this app on your local machine:
- clone this repository
- navigate to this folder in terminal
- $ docker build -t yt01 . #build a docker image to run this app
- $ docker run -d -p 5000:5000 yt01 #run the above built image on port 5000