From the course: LLMOps in Practice: A Deep Dive
Unlock this course with a free trial
Join today to access over 25,500 courses taught by industry experts.
Hands-on metrics - Python Tutorial
From the course: LLMOps in Practice: A Deep Dive
Hands-on metrics
So previously, we've been looking at our application, our chat application. We've added RAG to it and we've been looking at extending that RAG with ops. We saw logging and then we saw how to use logging and to do hands-on logging. And now just recently, we looked at metrics and in particular, the Prometheus client for metrics. So in this video, we're going to get hands-on and show how to configure that client. And in your directory, you'll have this metrics.js file and we're going to be working in that one as well as working in our server. So here's our metrics.js file. And you can see it requires the Prometheus client. And that Prometheus client is something that we'll be installing a little bit later as we run the server. But what I've done is within the Prometheus client, we wanted to capture some different metrics for different things that are going on. For example, the HTTP request duration, the RAG query duration, how long it took to do calls to OpenAI, and then feedback. So as…
Contents
-
-
-
-
-
-
(Locked)
Extending RAG with ops2m 9s
-
(Locked)
Logging3m 38s
-
(Locked)
Hands-on logging10m 23s
-
(Locked)
Metrics5m 28s
-
(Locked)
Hands-on metrics10m 20s
-
(Locked)
Version management4m 50s
-
(Locked)
Hands-on version management13m 32s
-
(Locked)
RAG Ops: Updating the data4m 37s
-
(Locked)
Hands-on RAG ops9m 22s
-
(Locked)
RAG in action: Exercise1m 4s
-
(Locked)
-