Deploying Your Model
By now, you have learned and put into practice the key concepts and tips for building exceptional deep learning models for regular regression and classification problems. In real life, models are not just built for learning purposes. On the contrary, when training models for purposes other than research, the main idea is to be able to reuse them in the future to perform predictions over new data that, although the model was not trained on, the model should perform similarly well with.
In a small organization, the ability to serialize and deserialize models suffices. However, when models are to be used by large corporations, by users, or to alter a massively important and large task, it is a better practice to convert the model into a format that can be used in most production environments (such as APIs, websites, and online and offline applications).
In this section, we will learn how to save and load models, as well as how to use PyTorch's most recent...
