Transfer Learning
In a convolutional neural network, the main task of the convolutional layers is to enhance the important features of an image. If a particular filter is used to identify the straight lines in an image then it will work for other images as well this is particularly what we do in transfer learning. There are models which are developed by researchers by regress hyperparameter tuning and training for weeks on millions of images belonging to 1000 different classes like imagenet dataset. A model that works well for one computer vision task proves to be good for others as well. Because of this reason, we leverage those trained convolutional layers parameters and tuned hyperparameters for our task to obtain higher accuracy.
Dog Breed Classification using Transfer Learning
In this article, we will learn how to build a classifier using the Transfer Learning technique which can classify among different breeds of dogs. This project has been developed using collab and the dataset has been taken from Kaggle whose link has been provided as well.