This is a multiclass image classifier using a CNN model that has been trained on the Keras integrated CIFAR-10 (labelled subset of tiny images) dataset. The inbuilt CNN uses three layers roughly modelled after the VGG blocks in the CNN introduced by this paper. There is built-in autotuning, made possible by KerasTuner.
This classifier was programmed in order to join the Waterloo Aerial Robotics Group's Computer Vision team. For more details see docs.uwarg.com.