Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[sdk] Regression in KFP 2.11.0: Dynamic CPU-limit placeholders fail to compile (worked in 2.10.0) #11646

Open
zackgow opened this issue Feb 17, 2025 · 0 comments

Comments

@zackgow
Copy link

zackgow commented Feb 17, 2025

Environment

  • KFP SDK version: 2.11.0
  • All dependencies version:
    • kfp: 2.11.0
    • kfp-pipeline-spec: 0.6.0
    • kfp-server-api: 2.3.0

Steps to reproduce

#11097 adds support dynamic machine type parameters in pipeline task setters, but this appears broken in 2.11.0.

import kfp
from kfp import dsl


@dsl.component
def cpu_limit() -> str:
    return "4000m"


@dsl.component
def sum_numbers(a: int, b: int) -> int:
    return a + b


@dsl.pipeline
def pipeline():
    sum_numbers_task = sum_numbers(a=1, b=2)
    sum_numbers_task.set_cpu_limit(cpu_limit().output)


kfp.compiler.Compiler().compile(
    pipeline_func=pipeline,
    package_path="test.json",
)

Expected result

A successfully compiled pipeline.

Materials and Reference

Downgrading allows a successful compilation: pip install --upgrade --force-reinstall "kfp==2.10.0"


Impacted by this bug? Give it a 👍.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant