https://flyte.org
Join Slack
Hi, community, I've come across two different parameters both called "image_config" in different par...
# flyte-support
d
Hi, community, I've come across two different parameters both called "image_config" in different parts of the code. I'm a little bit confused about why they are not identical. Should they represent the same use case, or do they have different purposes? Here are the instances I am referring to:
Copy code
remote = FlyteRemote(config=Config.auto(), \
                      default_project="flytesnacks", default_domain="development")


execution = remote.execute(wf, \
                           inputs={"name": "Kermit"},\
                           image_config=ImageConfig("futureoutlier/flyte-practice:latest"))
Copy code
flyte_workflow = remote.register_workflow(
    entity=flyte_entity,
    serialization_settings=SerializationSettings(
        image_config=ImageConfig.from_images("futureoutlier/flyte-practice:latest"),
        project="flytesnacks", 
        domain="development",
        version="v1"  # You need to specify the version here.
    ),
    version="v1",
)
Any clarification would be greatly appreciated. Thank you very much!
g
the first example is outdated, could you help us update the doc?
d
Sure, no problem !
Hi kevin, I would love to help you update the doc. But here's the problem. The first example came from the comment in the code. https://github.com/flyteorg/flytekit/blob/master/flytekit/configuration/__init__.py#L323C2-L339 It suggest users to use default image if there's no special requirement. Should I modify the comment in the code?
t
yes please
thank you.
the old type hints for 
default_image
in the 
ImageConfig
dataclass in the comments should also just be removed.
d
No problem, I will do it today !
I've done both in flytesnacks and flytekit. Please review it, thank you very much. https://github.com/flyteorg/flytesnacks/pull/1024 https://github.com/flyteorg/flytekit/pull/1734
Open in Slack
PreviousNext
Powered by

Flyte

Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Powered by