Driven Data Competition - Warm Up: Machine Learning with a Heart

Link with dataset : https://www.drivendata.org/competitions/54/machine-learning-with-a-heart/

The file SamplePipeline.ipynb contains a baseline pipeline for working with the dataset

Getting access to this notebook -

1. Fork this repo into your personal github profile
2. Ensure you can view/open the notebook by either opening it in Github or using https://nbviewer.jupyter.org/

   NOTE : NBViewer requires the repo to be public

3. Clone the repo into your local desktop/laptop and run the IPYNB file using jupyter

   NOTE : Make sure the dataset and paths are matching.

Making changes to the notebook (example) -

1. Uncomment lines within the EDA section in the notebook, run the whole notebook again

   NOTE: The pairplot will take some time

2. commit the changes with a comment (mandatory) and push them into your personal repo
3. Check in your online github profile to see if the notebook is rendering the changes (or use nbviewer)

Required improvments -

1. Code documentation in steps such as "Are there any missing data points?"
2. Train and test split MUST be done BEFORE encoding and scaling

---

Merging changes from this repo to your forked repo

This article is where I learnt it from : https://digitaldrummerj.me/git-sync-fork-to-master/

In a nutshell, Open the terminal in your working folder (the folder where your forked repo is), then type

1. git remote add upstream [original repo path].git
2. git fetch upstream
3. git merge upstream/master
   • Resolve merge conflicts if any
4. git push