How to Perform Large-Scale Image Classification

This course focuses on teaching participants how to perform large-scale image classification, specifically in the context of winning the Google Landmark Recognition 2020 Kaggle competition. The learning outcomes include understanding the challenges of landmark recognition with a vast number of classes, exploring different modeling techniques, and implementing efficient code for image classification. The course covers topics such as classical approaches, validation strategies, model architecture, fine-tuning, post-processing, ensembling, and augmentation techniques like cutout. The teaching method involves a video presentation by industry experts from NVIDIA, sharing their winning solution and insights. This course is intended for data scientists, machine learning practitioners, and individuals interested in deep learning competitions, particularly in the field of computer vision.