Containerized Python Components

Create Python components with more complex dependencies

The following assumes a basic familiarity with Lightweight Python Components.

Containerized Python Components extend Lightweight Python Components by relaxing the constraint that Lightweight Python Components be hermetic (i.e., fully self-contained). This means Containerized Python Component functions can depend on symbols defined outside of the function, imports outside of the function, code in adjacent Python modules, etc. To achieve this, the KFP SDK provides a convenient way to package your Python code into a container.

As a production software best practice, component authors should prefer Containerized Python Components to Lightweight Python Components when their component specifies packages_to_install, since the KFP SDK will install these dependencies into the component’s image when it is built, rather than at task runtime.

The following shows how to use Containerized Python Components by modifying the add component from the Lightweight Python Components example:

from kfp import dsl

def add(a: int, b: int) -> int:
    return a + b

1. Source code setup

Start by creating an empty src/ directory to contain your source code:


Next, add the following simple module, src/, with one helper function:

# src/
def add_numbers(num1, num2):
    return num1 + num2

Lastly, move your component to src/ and modify it to use the helper function:

# src/
from kfp import dsl
from math_utils import add_numbers

def add(a: int, b: int) -> int:
    return add_numbers(a, b)

src now looks like this:


2. Modify the dsl.component decorator

In this step, you’ll provide base_image and target_image arguments to the @dsl.component decorator of your component in src/

def add(a: int, b: int) -> int:
    return add_numbers(a, b)

Setting target_image both (a) specifies the tag for the image you’ll build in Step 3, and (b) instructs KFP to run the decorated Python function in a container that uses the image with that tag.

In a Containerized Python Component, base_image specifies the base image that KFP will use when building your new container image. Specifically, KFP uses the base_image argument for the FROM instruction in the Dockerfile used to build your image.

The previous example includes base_image for clarity, but this is not necessary as base_image will default to 'python:3.7' if omitted.

3. Build the component

Now that your code is in a standalone directory and you’ve specified a target image, you can conveniently build an image using the kfp component build CLI command:

kfp component build src/ --component-filepattern --no-push-image

If you have a configured Docker to use a private image registry, you can replace the --no-push-image flag with --push-image to automatically push the image after building.

4. Use the component in a pipeline

Finally, you can use the component in a pipeline:

from kfp import dsl

def addition_pipeline(x: int, y: int) -> int:
    task1 = add(a=x, b=y)
    task2 = add(a=task1.output, b=x)
    return task2.output

compiler.Compiler().compile(addition_pipeline, 'pipeline.yaml')

Since add’s target_image uses Google Cloud Artifact Registry (indicated by the URI), the pipeline shown here assumes you have pushed your image to Google Cloud Artifact Registry, you are running your pipeline on Google Cloud Vertex AI Pipelines, and you have configured IAM permissions so that Vertex AI Pipelines can pull images from Artifact Registry.


Was this page helpful?