From the course: Power BI Copilot Tips and Troubleshooting
Join today to access over 24,800 courses taught by industry experts.
Optimizing Copilot input prompts
From the course: Power BI Copilot Tips and Troubleshooting
Optimizing Copilot input prompts
- [Instructor] When we interact with Copilot, we're not interacting with an actual human, but rather data-trained on content and reference materials that humans trained that Copilot then matches. When we send a text prompt to our Copilot AI model in Power BI, it doesn't look for exact matches on the entire text input. Instead, Copilot and other large language models break down the text into smaller pieces of text even within a word, then pattern match this input against existing text pieces in the trained AI model that we already have structured outputs to make predictions about what these unknown outputs should be. We can use prompt engineering to build better AI model inputs for Generative AI tools like Copilot to optimize the predicted outputs they can return. Like many other concepts in data and AI, it's both an art and a science. Think back to the Copilot prompts as a chat window that we discussed earlier in this chapter. In Copilot, we start with the existing semantic model…
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.