In this repository, I put into test my newly acquired Deep Learning skills, in order to solve the Kaggle's famous Image Classification Problem, called "Dogs vs. Cats". Given images of dogs and cats, I try to predict if there is a dog or a cat depicted into each one of them, using Convolutional Neural Networks. More specifically, the following Deep Learning techniques will be used in this project: (a) Binary and (b) Multi-class Image Classification, (c) Data Augmentation, and (d) Transfer Learning along with Dropout regularization. By means of (c) and (d), will be examined, if, and to what extend, they help in mediating issues of serious overfitting, particularly arising when training neural networks. Lastly, the main programming tool for deploying this project is the Python's high-level neural network library, called Keras, which runs on top of TensorFlow.