Deep Learning with PyTorch

Building the network architecture

For most of the real-world use cases, particularly in computer vision, we rarely build our own architecture. There are different architectures that can be quickly used to solve our real-world problems. For our example, we use a popular deep learning algorithm called ResNet, which won the first prize in 2015 in different competitions, such as ImageNet, related to computer vision. For a simpler understanding, let's assume that this algorithm is a bunch of different PyTorch layers carefully tied together and not focus on what happens inside this algorithm. We will see some of the key building blocks of the ResNet algorithm in Chapter 5, Deep Learning for Computer Vision, when we learn about CNNs. PyTorch makes ...