Predict Census Income using Deep Learning Models

In this project, we are going to work on Deep Learning using H2O to predict Census income.
Each project comes with 2-5 hours of micro-videos explaining the solution.
Code & Dataset
Get access to 50+ solved projects with iPython notebooks and datasets.
Project Experience
Add project experience to your Linkedin/Github profiles.

What will you learn

  • Apply DL models
  • Use of H2O
  • Compare the DL models
  • Grid Search on DL models
  • Tuning the accuracy in DL models

Project Description

This data was extracted from the census bureau database found at:
Donor: Ronny Kohavi and Barry Becker,
            Data Mining and Visualization
            Silicon Graphics.
            e-mail: for questions.
Split into train-test using MLC++ GenCVFiles (2/3, 1/3 random).
48842 instances, mix of continuous and discrete    (train=32561, test=16281)
45222 if instances with unknown values are removed (train=30162, test=15060)
Duplicate or conflicting instances : 6
Class probabilities for adult.all file
Probability for the label '>50K'  : 23.93% / 24.78% (without unknowns)
Probability for the label '<=50K' : 76.07% / 75.22% (without unknowns)

Extraction was done by Barry Becker from the 1994 Census database.  
A set of reasonably clean records was extracted using the following conditions:
   ((AAGE>16) && (AGI>100) && (AFNLWGT>1)&& (HRSWK>0))
Prediction task is to determine whether a person makes over 50K a year.

Curriculum For This Mini Project

  Problem Statement
  Import Data Sets
  What is Deep Learning?
  Understanding H2O
  Data Sanity Check
  Remove leading white space
  Impute Missing values
  Initializing H2O
  Train H2O model without hidden layer (C )
  Hyperparameter optimization
  Random Grid Search