From the course: Accessibility and AI: Bridging the Disability Divide
Join today to access over 24,700 courses taught by industry experts.
Enriching APIs with your own data
From the course: Accessibility and AI: Bridging the Disability Divide
Enriching APIs with your own data
Ready-to-use APIs are often the option for leveraging the latest advances in AI. But sometimes, you might want to train the model on your own data, or add a feature or custom functionality that the regular models don't support. Now, you might be asking why. Well, one key benefit is better personalization and localization, like supporting local languages or dialects and addressing cultural norms. Another could be supporting new domains that regular models don't work as well with, such as answering questions on accessibility. So let's get into fine-tuning or training based models on your data. To train a model from scratch involves three main steps. First, gathering and organizing the data you need. Second, choosing the design or setup of the model, also known as model architecture, and third, running the actual training process. But there is another approach known as fine-tuning. Fine-tuning is a process that builds on top of an existing model by enhancing it with data that the…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.