Mousam

About

This is a machine learning based application which makes use of the delhi weather kaggle dataset.

The application makes use of lagged datasets

We have made use of logistic regression and linear regression for the prediction of rain and temperature respectively

The application predicts average temperature for the day in degree celsius

Our Parameter for calling a day as hot is if the average temperature for the day is above 25 degree celsius

Graphs and Plots

Some of my graphical analysis is shown below :

Hosting with Flask and Git

Hosting with heroku is pretty simple, just follow the following steps:

1. Heroku Login - This will open up the browser through which you can log in to your heroku account

2. git clone (your repo link) - This clones your repo

3. cd into the folder of interest and use the following commands:-

      1. git init - It will initialize an empty repository
      
      2. git add . - It will add the files to the repository
      
      3. git commit -m "message" - It will commit the changes
   

4. Use heroku create (your desirable app name) and then use heroku git:remote -a (the git link generated from heroku create command)

5. Finally use git push heroku master

NOTE

1. Add a requirements.txt to your repository
2. Add Procfile to your repository

Our App

Our weather prediction app is available at https://mousam-webapp.herokuapp.com

For any doubts drop an email to me