Image classification is a typical deep learning application. This task had an initial increase of interest thanks to the ImageNet (http://image-net.org/)  image database organized according to the WordNet (http://wordnet.princeton.edu/) hierarchy (currently only the nouns), in which each node of the hierarchy is depicted by hundreds and thousands of images. More precisely, ImageNet aimed to label and categorize images into almost 22,000 separate object categories. In the context of deep learning, ImageNet refers generally to the work contained in the paper ImageNet Large Scale Visual Recognition Challenge (http://www.image-net.org/challenges/LSVRC/), or ILSVRC for short. In this case, the goal is to train a model that can classify an input image into 1,000 separate object categories. In this recipe, we will use pre-trained models over 1.2 million training images with 50,000 validation images and 100,000 testing images.