Uses Deep Conv Neural Network for Large-Scale Image Classifier in python, pytorch and torchvision. This project uses deep learning model pretrained on hundreds of thousands of images as part of the overall application architecture. It uses and compares VGG13, VGG19, Densenet, Resnet152 FC for transfer learning in order to predict different types of flowers. The application is portable over to a mobile device e.g. phone in order to capture images of unknown flowers to make accurate predictions of their names and types.
There are four files to use, one is the train.py, and the other is the predict.py, and two utility files, namely arguments and utils.