From the course: Fine-Tune Your LLMs
Unlock the full course today
Join today to access over 22,700 courses taught by industry experts or purchase this course individually.
Solution: Develop a chatbot based on a fine-tuned model
From the course: Fine-Tune Your LLMs
Solution: Develop a chatbot based on a fine-tuned model
(upbeat dance music) - [Instructor] How did you do? Let's look at how I solved this challenge. Here we are in code spaces. I've already executed all the cells in this notebook for efficiency. Don't forget to put your API key in a local environment variable or hard code it within the notebook before executing the code. Let's scroll down to cell 57, all the way at the bottom. The section is the Customer Support Chatbot. The first step is to set the persona for the AI assistant. I'm using an array to keep track of the conversation history. The first message I set is the system message, and it says, "This is a customer support chatbot designed to help with common inquiries for Kesha's Boutique." Remember, the goal is to manage a multi-turn conversation, which is a back and forth dialogue with ChatGPT. When holding a multi-turn conversation, we'll need to send the entire conversation history with each request. I've created this collect messages function that keeps track of the messages…
Contents
-
-
-
-
-
-
Train a new fine-tuned model4m 9s
-
(Locked)
Challenge: Fine-tune a pretrained LLM48s
-
(Locked)
Solution: Fine-tune a pretrained LLM3m 23s
-
(Locked)
Retrieve and use a fine-tuned model4m 3s
-
(Locked)
Challenge: Develop a chatbot based on a fine-tuned model1m 29s
-
(Locked)
Solution: Develop a chatbot based on a fine-tuned model4m 10s
-
-
-