From the course: Creating Advanced AI Applications with Python, APIs, and GitHub Models

Unlock this course with a free trial

Join today to access over 25,500 courses taught by industry experts.

Solution: Using tools when required

Solution: Using tools when required

From the course: Creating Advanced AI Applications with Python, APIs, and GitHub Models

Start my 1-month free trial Buy for my team

Solution: Using tools when required

“

(upbeat music) - [Instructor] Now as part of our solution, we need to be able to distinguish between when we're using function calling and tools versus when we can just use the large language model. So let's go ahead and update our run_pipeline function. So this is what my run_pipeline function currently looks like, so it looks like there's a section for if I'm going to be using tool_calls or if I'm going to be using just the large language model, and this is this section over here. So let's move some of the code from here to the pipeline function, so response and the call to the large language model. And so we need to provide the input as a parameter. And so in this case, we're just going to be calling the run_pipeline instead. Now let's head over to the OpenAI documentation, and in the function calling section under the Advanced usage, if you scroll down to the bottom, you can see that if we have the finish_reason as being tool_calls, that means we are using function callings and…

Contents