BinaryClassification-Cats_Dogs

This project is used to work upon the identification of Cats and Dogs from a given source(Here,Dataset available at Kaggle). The model uses 3 pretrained models namely ResNet50,InceptionV3 and VGG19 for transfer learning. An ensemble scheme(of reduced mean) has then been used to train the model. Here, the custom made training loops, using the concepts of gradient tape has been used. Later, the analysis of the model has been done through confusion matrix and other graphical means.