Code to create a classifier for a kaggle image processing competition
This script was used to train a binary classifier using data from a Kaggle competition. Training and testing data consisted of GIST and CNN image data of a scene, in which the classifier had to ascertain whether a live object was present. This was accompanied by additional GIST and CNN data containing empty cells that had to be imputed, data detailing the confidence with which annotators assigned classes to images and the proportion of positive and negative classes within the final test data. This data was used to train a Gaussian Support Vector Machine using the Classification Learner.