Python - Error in "vetiver.prepare_docker()" with the examples.

[coforfe]

Hi,
Thanks again for your continuous effort to improve this library and for making equivalent to use it betwen R and Python.

It is fantastic to generate so easisy a FastAPI end-point and submit data for prediction!.

I am trying to reproduce the whole example in the documentation with Python, and I do not have any problem, except when I try to get the DockerFile with vetiver.prepare_docker().

I am using two approches to generate the DockerFile:

• one with the code in your cars_mpg example

vetiver.prepare_docker(model_board, "cars_mpg")

• and the other one by using what you have in the superbowl_ml example.

vetiver.prepare_docker(
    model_board, 
    "cars_mpg",
    version = "20230619T163528Z-22feb",
    port = 8080
)

and both with errors...

Attached you can find some screenshots of the different errors I get.
From the errors, I do not know if I should provide an URL where the docker will be called....

Thanks,
Carlos.
P.S: I am using Vetiver's 0.2.1 version.

![Captura de pantalla 2023-06-19 a las 16 41 02](https://github.com/rstudio/vetiver-r/assets/8349230/0f034fcd-ac91-42c1-b59c-5

1740855e027)

[coforfe]

I was just curious about if with R, I got the same error or something different and It run flawlessly, generating the three files as you stated in the documentation....

[isabelizimm]

Hello--first off, thank you for reporting this error! Are you able to to tell me what version of Python pins you are using? It's possible that an early version of pins + the newest vetiver has some conflicts, as the url field was not always available.

[coforfe]

Hi Isabel,
Yes, of course...

It is 0.7.0...
I did update vetiver but not pins... Let me update it and try again.

Thanks for your always quick answers!.
Carlos.

[coforfe]

Hi Isabel,

Yes!.
It worked after updating pins !.
It generated the:

• Dockerfile
• vetiver_requirements.txt
• app.py

I will try to run the Dockerfile and generate a prediction with all of this.

Thanks again!
Carlos.

[isabelizimm]

Great to hear it!