Apart from designing custom CNN architectures, we can use architectures that have already been built. ResNet is one such popular architecture. In most cases, we can achieve better performance by using such architectures. In this chapter, we will learn how to perform distributed training of a ResNet model in Flax.

In this code example, we will train the model from scratch‚ meaning that we will not use the pretrained weights. In a separate chapter, we’ve covered how to perform transfer learning with ResNet.

Prior to training the ResNet model, it’s important to process the data, which was covered previously.

Get hands-on with 1200+ tech skills courses.