Hi everyone, is there any suggestions to run task2 until task1 finished when they have no dependency?
Copy code
with Flow('feature-factory') as flow:
a, b = pre_task()
task1.map(a)
task2.map(b)
Zac Chien
08/25/2021, 9:54 AM
It seems like an easy question. Currently my workaround solution for this is to add a meaningless dependency for them. Is there any better way to achieve this?
Copy code
with Flow('feature-factory') as flow:
a, b = pre_task()
state = task1.map(a)
task2.map(b, state)
r
Robert Hales
08/25/2021, 11:00 AM
Think
upstream_tasks
kwarg is what you’re looking for
z
Zac Chien
08/25/2021, 11:16 AM
Copy code
b.map(task2, upstream_tasks=[task1.map(a)])
I don’t know why task2 only run once (should be 9 runs), in my case, feature_generator=task1, dataset_generator=task2
Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.