admin管理员组

文章数量:1414613

I want to have a downstream component that has an optional dsl.Dataset input. That optional input is the output of an upstream task within a dsl.If. All of these are within a dsl.ParallelFor, so that I can consume everything with a dsl.Collected afterwards. Something like:

from typing import Optional
from kfp import dsl

@dslponent
def component_A_op() -> dsl.Dataset:
    return Dataset()

@dslponent
def component_B_op(a_data: Optional[Dataset] = None) -> Dataset:
    if a_data:
        return Dataset()
    else:
        return Dataset()

@dslponent
def component_C_op(datas: List[Dataset]):
    pass

@dsl.pipeline()
def pipeline:
    with dsl.ParallelFor(items=["item_1", "item_2"]) as _items:
        with dsl.If(_items == "item_1"):
            component_A = component_A_op()
        component_B = component_B_op(
            a_data=  # if component_A is present then component_A.output else None
        )
    component_C = component_C_op(datas=dsl.Collected(component_B)

Is this possible?

本文标签：

版权声明：本文标题：google cloud vertex ai - Is this possible: Use output of upstream component within a dsl.If; otherwise, use None - Stack Overflo 内容由网友自发贡献，该文观点仅代表作者本人， 转载请联系作者并注明出处：http://www.betaflare.com/web/1745168756a2645843.html， 本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌抄袭侵权/违法违规的内容，一经查实，本站将立刻删除。