From the course: Applied AI: Getting Started with Hugging Face Transformers

Course coverage and prerequisites

From the course: Applied AI: Getting Started with Hugging Face Transformers

Start my 1-month free trial Buy for my team

Course coverage and prerequisites

“

- [Instructor] Before we get started with the course, let's review the course coverage and prerequisites. We will begin this course with the review of the machine learning process for natural language processing or NLP tasks. Then we will discuss transformers and their architecture and components. Next, we will discuss Hugging Face, explore its website, and review various capabilities available for the community. We will then get to building NLP tasks with pre-trained transformers from Hugging Face. We will explore two such tasks, namely sentiment analysis, and named entity recognition. It is important to note what we will not cover, so we set the expectations right. Transformers are built on complex mathematics behind the scenes. We will not discuss the mathematics behind transformers, but just explore its architecture at a block diagram and data flow level. It is recommended to read the original transformer's paper, if students are interested further in the mathematics behind it. We are not going to build transformers from scratch, we will leverage the pre-trained transformers available in Hugging Face to implement our tasks. Since Hugging Face is a community platform, anyone can contribute to its repository, but we will not discuss the steps needed for doing so. This course has a number of prerequisites. I strongly recommend that you get familiar with these concepts before taking this course. You should be familiar with general machine learning concepts and technologies. You should also be aware of natural language processing concepts and the machine learning process to build such applications. Deep learning concepts and architectures is a key domain to be familiar with. Recurrent neural networks and embeddings are also essential concepts as we use those as part of the course. This course uses examples in Python with Jupyter Notebooks, so familiarity is also required there. Using Keras and TensorFlow frameworks is also essential to execute the examples in this course. Some of the key recommended prerequisite courses are Deep Learning: Getting Started and Recurrent Neural Networks. They cover important concepts and examples that are foundations for this course. Let's now set up the environment required for the course.

Contents