Table of Contents
Let’s try an exercise to use a Large Language Model in an application. To do that, we will use several different platforms. We don’t technically need all of these platforms to build and deploy our application, but it is good to familiarize yourself with them, as we will use them later in this class.
Hugging Face is a leading open-source platform that provides a hub for machine learning models. It houses thousands of pre-trained transformer models for tasks like text generation, translation, summarization, sentiment analysis, etc.
Researchers at Google Brain popularized transformer models. These models are designed to process sequences of data efficiently. They rely on self-attention to understand the relationships between all input parts simultaneously rather than processing them step by step like older models. This makes them much faster and better at handling large datasets, especially in tasks like language understanding, translation, or text generation. The libraries at Hugging Face and the simple Python commands make it easy for developers to access LLMs and build applications with them without knowing the details of how these models work.
We will use Python, a programming language for web development, data science, AI, and automation, to build our application. Python is known for its versatility and ease of use. It simplifies coding; it has many libraries to help you create various applications. You can start with Google’s Python Class, the Python.org documentation to learn Python, and Microsoft’s Introduction to Python Course.
To test the application, we will use Colab. Google Colab (Collaboratory) is a cloud-based platform provided by Google that allows you to write and execute Python code in a browser. It offers free access to powerful hardware like GPUs and TPUs.
We will also use GitHub to store our source code for the application. Developers and researchers use GitHub to share code, collaborate, and demonstrate coding projects. It is one of the most well-known coding collaboration platforms in the world.
Go to the Hugging Face Codespaces: https://huggingface.co/kimappl