Objective

In this tutorial, I’m going to walk you through using a pre-trained neural network to extract a feature vector from images and cluster the images based on how similar the feature vectors are.

The model

The pre-trained model that will be used in this tutorial is the VGG16 convolutional neural network (CNN), which is considered to be state of the art for image recognition tasks. We are going to be using this model as a feature extractor only, meaning that we will remove the final (prediction) layer so that we can obtain a feature vector.

The Data

This implementation will use the flowers dataset from Kaggle which you can download here. The dataset contains 210 images of 10 different species of flowers that will be downloaded as _png _files.

#transfer-learning #machine-learning #clustering #feature-extraction