Image classification has been one of the leading research fields in the last five years. This is not surprising because being able to successfully classify images solves many business problems across a variety of industries. For example, the entire autonomous vehicle industry is dependent on the accuracy of these image classification and object detection models.
In this chapter, we will look at how Amazon SageMaker drastically simplifies the image classification problem. Aside from gathering a rich set of images for training, we will look at how to specify hyperparameters (parameters internal to the algorithm), train Docker images, and use infrastructure specifications for training.
In this chapter, we will cover the following topics:
- Walking through convolutional neural and residual networks
- Classifying images through transfer learning...
