From the course: Prompt Flow: Hands-On
Unlock this course with a free trial
Join today to access over 25,300 courses taught by industry experts.
Create a Docker image for a flow - Azure AI Services Tutorial
From the course: Prompt Flow: Hands-On
Create a Docker image for a flow
- [Instructor] To run our flow using Docker, we first need to create a Docker image from the Docker file. That's done through the command line, and now we're working with Docker instead of Prompt Flow. First, in terminal, navigate to the parent folder of the Docker folder. So that's cd reader. Then I'll run the following command: docker build, and then I point at the folder, which in this case is dist. Say -t, and then I need to give my Docker container a name. So I'll call this one reader-serve. Hit return, and now Docker starts building out the container. Now this process is involved and takes quite some time because we're now building out an entire environment based on the different settings inside the prompt flow and based on the requirements that are provided in the project. So if you've been sitting here following along with me all the way up until now, this is a good time to step aside and take a break and come back when the process is done.
Contents
-
-
-
Idea: Build an article discussion app for kids2m 32s
-
(Locked)
Creating a new Prompt Flow from scratch2m 15s
-
(Locked)
Creating a connection to access AI services2m 17s
-
(Locked)
Testing the flow1m 39s
-
(Locked)
Adding custom Python to the flow4m 24s
-
(Locked)
Continuous incremental testing1m 40s
-
(Locked)
Adding LLM tools to the flow3m
-
(Locked)
Adding new input fields to LLM tools4m 26s
-
(Locked)
Debugging the flow3m 10s
-
(Locked)
Creating conditional flow patterns7m 22s
-
(Locked)
Adding required Python dependencies before deployment2m 40s
-
(Locked)
Building a flow using Docker3m 5s
-
(Locked)
Create a Docker image for a flow59s
-
(Locked)
Run the Docker image3m 10s
-
(Locked)
Bonus: Add Mistral AI as a custom connection5m 10s
-
-