Census Income Dataset Classification

Data Science Notebook on a Classification Task

Objective

In the Jupyter Notebook included in this page, we will using the Census Income Dataset to predict whether an individual's income exceeds $50K/yr based on census data.

The Dataset can be found here:

https://archive.ics.uci.edu/ml/datasets/adult

The Notebook can be found here:

https://github.com/dformoso/sklearn-classification/blob/master/Data%20Science%20Workbook%20-%20Census%20Income%20Dataset.ipynb

Companion Mindmap/Cheatsheet

This Jupyter Notepad has a companion Mindmap/Cheatsheet that lists most of the Data Science steps that can be found at the following link:

https://github.com/dformoso/machine-learning-mindmap

Steps

In this Notebook, we'll perform:

Feature Exploration (Uni and Bi-variate)
Feature Imputation
Feature Selection
Feature Encoding
Feature Ranking
Machine Learning with sklearn and Tensorflow
Random Search
Accuracy, Precision, Recall, and f1 calculations
ROC Curve

Setup

This Notebook has been designed to be run on top of the Jupyter Tensorflow Docker instance found in the link below:

https://github.com/jupyter/docker-stacks/tree/master/tensorflow-notebook

If you haven't downloaded Docker at this point, please visit:

https://www.docker.com/get-docker

Then, open a shell or terminal session and copy/paste the following:

docker run -itd \
  --restart always \
  --name jupyter \
  --hostname jupyter \
  -p 8888:8888 \
  -p 6006:6006 \
  jupyter/tensorflow-notebook:latest \
  start-notebook.sh --NotebookApp.token=''

Upon running the command, docker will automatically pull the images it needs and get the containers going for us.

Give it a minute or so for Jupyter to start, and head to the following URL: http://localhost:8888

You should now have Jupyter running. If after a minute you can't reach the URL, check that the containers are running correctly and the network has been created by typing:

### Check the containers are running
docker ps -a

Loading the Notebook

Download it from this link:

https://github.com/dformoso/sklearn-classification/blob/master/Data%20Science%20Workbook%20-%20Census%20Income%20Dataset.ipynb

Go back to:

http://localhost:8888, load your Notebook into Jupyter and run it. That's it!

Troubleshooting Docker

Here's a few useful commands in case something goes wrong with your docker instance:

# Restart Jupyter Docker Container
docker restart jupyter

# Stop Jupyter Docker Container
docker stop jupyter

# Remove Jupyter Docker Container
docker rm jupyter

Feature Exploration (Uni and Bi-variate) Feature Imputation Feature Selection Feature Encoding Feature Ranking Machine Learning Training Random Search Accuracy, Precision, Recall, and f1 calculations ROC Curve