In a previous post I looked at how to create an age/gender classifier using Keras. In this post we'll look at the final product - a web app that uses the classifier to generate a life insurance quote. To get from the python model to the finished product involves a few steps:

In the next two posts I'll cover how to use the Keras python library and the R Shiny package to create a web app capable of producing a life insurance quote from an image of a face. To do this, we'll need to have a model that can estimate a persons age and gender from our input image - for this we will use a convolutional neural network (CNN).

Following on from a previous post, this article discusses:

• Assessing how well the GLM generalises to new data with cross validation
• Automated GLM selection using grid search

Previously, we looked at using H2O's GLM function to set a rating plan for life insurance contracts. Importing and preparing the data, the use of offsets and fitting the GLM with h2o were all covered here more ...