From the course: LLMOps in Practice: A Deep Dive

Unlock the full course today

Join today to access over 24,900 courses taught by industry experts.

Challenge: Implementing RLHF and user feedback

Challenge: Implementing RLHF and user feedback - Python Tutorial

From the course: LLMOps in Practice: A Deep Dive

Start my 1-month free trial Buy for my team

Challenge: Implementing RLHF and user feedback

“

For this challenge, you're going to implement what we just discussed, capturing human feedback on the quality of the chatbots and using that to make a decision. When you're done, your chatbot should look a little bit like this, with the core difference being that when you receive a message back from the chatbot, it also has these three buttons: good, neutral, or bad. If the user presses bad, then a new completion will be triggered from the assistant. Note that it says trying again and then gives you a new message. The system should also log that negative feedback. At some future point when you fire up the server, you might see excessive negative feedback and use that to trigger a server change. The challenge to implement this comes in three parts. The first part of the challenge is to implement these buttons on the front end. I'll give you a hint, and here's some code that will go into your front end. It will create an element in your HTML to contain your feedback buttons and render…

Contents