Skip to content

Prediciting Twitter users MBTI using BERT algorithm

Notifications You must be signed in to change notification settings

oumaimaaouadi123/mbti-bert

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

Introduction

This deep learning project aims to predict Twitter users personality type using MBTI as personality classification model and BERT algorithm as a classifier. In this document, we will explain to you how to open this project and test it yourself.

Opening the notebook with google colab

Google Colaboratory is a free online cloud-based Jupyter notebook environment that allows you to train your machine learning and deep learning models on CPUs, GPUs, and TPUs

To open a notebook in google colab to need to :

  • Sign in into your gmail account
  • Go to https://colab.research.google.com
  • Go to File > Upload Notebook > Choose a file
  • Select the .ipynb file you want to open from your computer

Importing files

To upload a file to google colab, you need to:

  • Go to the side bar and click in "Files" icon
  • Click on the "Upload to session storage" icon
  • Choose the file you want to upload from your computer

In our case, we need to upload two files :

Selecting the runtime type

As we explained, google colab gives you access to hardware ressources, such as RAM, disk storage, CPUs and eventually GPUs.

Running your Machine Learning / Deep Learning models on CPUs will cost you a lot of time, so it recommanded that you run them on GPUs.

In order to do so, you need to :

  • Go to "Runtime"
  • Click on "Change the runtime type"
  • Choose the "GPU" option and save

Selecting this option will help your models train faster and return results in a very short period of time.

Running the code

You can run your code cell by cell or go to Runtime > Run all

About

Prediciting Twitter users MBTI using BERT algorithm

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published