feature: with_feature_group [feature_store] #3658

patrickmcarlos · 2023-02-14T18:42:32Z

Issue #, if available:

Description of changes:
Adds three new parameters to the with_feature_group Feature Store dataset_builder class.

These three new parameters are:

        feature_name_in_target (str): A string representing the feature in the target feature group
            that will be compared to the target feature in the base feature group. If None is
            provided, the record identifier will be used in the join statement. (default: None).
        join_comparator (JoinComparatorEnum): A JoinComparatorEnum representing the comparator used
            when joining the target feature in the base feature group and the feature in the target
            feature group (default: None).
        join_type (JoinTypeEnum): A JoinTypeEnum representing the type of join between the base and
            target feature groups. (default: None).

feature_name_in_target

Allows users to provide the feature name in the target feature group that will be used as a join key alongside the currently existing target_feature_name_in_base feature in the SQL JOIN statement. Before, only the record identifier feature in the target feature group could be used when creating the query.

join_comparator

Allows user to provide a Join comparator, which will be used to compare the two join keys. For example, we can now join a row together if feature_1 (target_feature_name_in_base) in feature group A is greater than feature_1 (feature_name_in_target). Before, only = (Equals) was able to be used.

join_type

Allows user to provide a Join type, which will specify the type of Join in the SQL statement. Before, users could only specify an (INNER) JOIN.
Enum keys are based on the join types available in the Athena docs https://docs.aws.amazon.com/athena/latest/ug/select.html

Usage

Say we have two feature groups, feature_group_a, and feature_group_b.

feature_group_a has three features, rec-id-a, event-time-id-a and feature-a.
feature_group_b has three features, rec-id-b, event-time-id-b and feature-b.

To join these feature groups together, a user might write the code:

dataset_builder = feature_store.create_dataset(
    base=feature_group_a,
    output_path=f"s3://{s3_bucket_name}",
)

result_df, query = dataset_builder.with_feature_group(
    feature_group=feature_group_b,
    target_feature_name_in_base="feature-a",
).to_dataframe()

However, a user may want to join feature-a from feature_group_a (base) and feature-b from feature_group_b (target). Furthermore, the user wants to perform a RIGHT_JOIN instead of a regular JOIN (inner) and use the LESS_THAN comparator instead of equals

Our code could look like this:

result_df, query = dataset_builder.with_feature_group(
    feature_group=feature_group_b,
    target_feature_name_in_base="feature-a",
    feature_name_in_target="feature-b"
    join_comparator=JoinComparatorEnum.LESS_THAN,
    join_type=JoinTypeEnum.RIGHT_JOIN).to_dataframe()

or without named arguments

result_df, query = dataset_builder.with_feature_group(
    feature_group_b,
    "feature-a",
    None,
    "feature-b"
    JoinComparatorEnum.LESS_THAN,
    JoinTypeEnum.RIGHT_JOIN).to_dataframe()

Testing done:

Unit & integration test to ensure properties of dataset_builder class are correctly updated when the new parameters are used in with_feature_group
Unit & integration test for the SQL query generated when using the with_feature_group with the new parameters.
Currently existing unit tests are unchanged.

Merge Checklist

Put an x in the boxes that apply. You can also fill these out after creating the PR. If you're unsure about any of them, don't hesitate to ask. We're here to help! This is simply a reminder of what we are going to look for before merging your pull request.

General

I have read the CONTRIBUTING doc
I certify that the changes I am introducing will be backward compatible, and I have discussed concerns about this, if any, with the Python SDK team
I used the commit message format described in CONTRIBUTING
I have passed the region in to all S3 and STS clients that I've initialized as part of this change.
I have updated any necessary documentation, including READMEs and API docs (if appropriate)

Tests

I have added tests that prove my fix is effective or that my feature works (if appropriate)
I have added unit and/or integration tests as appropriate to ensure backward compatibility of the changes
I have checked that my tests are not configured for a specific region or account (if appropriate)
I have used unique_name_from_base to create resource names in integ tests (if appropriate)

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

trajanikant

/bot run pr, slow-tests

trajanikant

/bot run all

sagemaker-bot · 2023-02-16T01:38:51Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-unit-tests
Commit ID: 487322e
Result: SUCCEEDED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

codecov-commenter · 2023-02-16T01:40:33Z

Codecov Report

Merging #3658 (c5b97b2) into master (1ad9d9c) will decrease coverage by 0.78%.
The diff coverage is 95.65%.

@@            Coverage Diff             @@
##           master    #3658      +/-   ##
==========================================
- Coverage   89.75%   88.98%   -0.78%     
==========================================
  Files         968      227     -741     
  Lines       91168    22569   -68599     
==========================================
- Hits        81824    20082   -61742     
+ Misses       9344     2487    -6857

Impacted Files	Coverage Δ
src/sagemaker/feature_store/dataset_builder.py	`87.87% <95.65%> (ø)`
...maker/cli/compatibility/v2/modifiers/image_uris.py
...10/lib/python3.10/site-packages/sagemaker/model.py
...ython3.7/site-packages/sagemaker/debugger/utils.py
...hon3.7/site-packages/sagemaker/local/exceptions.py
...ite-packages/sagemaker/async_inference/__init__.py
...ython3.10/site-packages/sagemaker/pytorch/model.py
...s/sagemaker/workflow/pipeline_experiment_config.py
...310/lib/python3.10/site-packages/sagemaker/logs.py
...3.8/site-packages/sagemaker/workflow/model_step.py
... and 1186 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

sagemaker-bot · 2023-02-16T01:46:00Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-local-mode-tests
Commit ID: 487322e
Result: SUCCEEDED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

sagemaker-bot · 2023-02-16T02:10:38Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-notebook-tests
Commit ID: 487322e
Result: SUCCEEDED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

sagemaker-bot · 2023-02-16T03:27:52Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-pr
Commit ID: 487322e
Result: FAILED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

sagemaker-bot · 2023-02-16T03:49:32Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-slow-tests
Commit ID: 487322e
Result: FAILED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

trajanikant

/bot run pr, slow-tests

sagemaker-bot · 2023-02-17T00:30:54Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-slow-tests
Commit ID: 487322e
Result: FAILED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

sagemaker-bot · 2023-02-17T01:47:26Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-pr
Commit ID: 487322e
Result: FAILED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

trajanikant

/bot run pr

sagemaker-bot · 2023-02-17T22:08:06Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-pr
Commit ID: 487322e
Result: FAILED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

trajanikant

/bot run pr

sagemaker-bot · 2023-02-21T20:34:08Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-pr
Commit ID: 487322e
Result: FAILED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

trajanikant

/bot run all

sagemaker-bot · 2023-03-04T02:34:17Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-pr
Commit ID: c5b97b2
Result: SUCCEEDED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

sagemaker-bot · 2023-03-04T02:45:36Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-unit-tests
Commit ID: c5b97b2
Result: SUCCEEDED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

sagemaker-bot · 2023-03-04T02:52:56Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-local-mode-tests
Commit ID: c5b97b2
Result: SUCCEEDED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

sagemaker-bot · 2023-03-04T02:54:21Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-notebook-tests
Commit ID: c5b97b2
Result: SUCCEEDED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

sagemaker-bot · 2023-03-04T04:50:26Z

AWS CodeBuild CI Report

CodeBuild project: sagemaker-python-sdk-slow-tests
Commit ID: c5b97b2
Result: FAILED
Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

patrickmcarlos marked this pull request as ready for review February 15, 2023 00:20

patrickmcarlos requested a review from a team as a code owner February 15, 2023 00:20

patrickmcarlos requested review from trajanikant and removed request for a team February 15, 2023 00:20

trajanikant reviewed Feb 15, 2023

View reviewed changes

trajanikant reviewed Feb 16, 2023

View reviewed changes

trajanikant reviewed Feb 17, 2023

View reviewed changes

knikure assigned trajanikant Feb 21, 2023

trajanikant reviewed Feb 21, 2023

View reviewed changes

patrickmcarlos force-pushed the patmcar-improve-join-functionality branch 3 times, most recently from 152d7cf to c5b97b2 Compare March 3, 2023 23:50

mizanfiu approved these changes Mar 3, 2023

View reviewed changes

trajanikant reviewed Mar 4, 2023

View reviewed changes

patrickmcarlos added 9 commits March 6, 2023 12:19

updates with_feature-group def and adds test

119190d

applies lint

53cda21

adds integration test

acd1867

adds additional unit test for dataset builder object

26e6ace

adds docstrings, lints

b26b725

runs lint for test_dataset_builder

34ef36c

additional docstring changes

fb7cdbd

use enum types for join type and comparator params

7ace2de

adds support for more athena supported types and comparators

1ced27a

trajanikant force-pushed the patmcar-improve-join-functionality branch from c5b97b2 to 1ced27a Compare March 6, 2023 20:19

trajanikant changed the title ~~feature: feature store with_feature_group functionality changes~~ feature: with_feature_group [feature_store] Mar 6, 2023

trajanikant merged commit 67dc885 into aws:master Mar 6, 2023

nmadan pushed a commit to nmadan/sagemaker-python-sdk that referenced this pull request Apr 18, 2023

feature: with_feature_group [feature_store] (aws#3658)

768888e

feature: with_feature_group [feature_store] #3658

feature: with_feature_group [feature_store] #3658

Uh oh!

Conversation

patrickmcarlos commented Feb 14, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Usage

Merge Checklist

General

Tests

Uh oh!

trajanikant left a comment

Choose a reason for hiding this comment

Uh oh!

trajanikant left a comment

Choose a reason for hiding this comment

Uh oh!

sagemaker-bot commented Feb 16, 2023

AWS CodeBuild CI Report

Uh oh!

codecov-commenter commented Feb 16, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

sagemaker-bot commented Feb 16, 2023

AWS CodeBuild CI Report

Uh oh!

sagemaker-bot commented Feb 16, 2023

AWS CodeBuild CI Report

Uh oh!

sagemaker-bot commented Feb 16, 2023

AWS CodeBuild CI Report

Uh oh!

sagemaker-bot commented Feb 16, 2023

AWS CodeBuild CI Report

Uh oh!

trajanikant left a comment

Choose a reason for hiding this comment

Uh oh!

sagemaker-bot commented Feb 17, 2023

AWS CodeBuild CI Report

Uh oh!

sagemaker-bot commented Feb 17, 2023

AWS CodeBuild CI Report

Uh oh!

trajanikant left a comment

Choose a reason for hiding this comment

Uh oh!

sagemaker-bot commented Feb 17, 2023

AWS CodeBuild CI Report

Uh oh!

trajanikant left a comment

Choose a reason for hiding this comment

Uh oh!

sagemaker-bot commented Feb 21, 2023

AWS CodeBuild CI Report

Uh oh!

trajanikant left a comment

Choose a reason for hiding this comment

Uh oh!

sagemaker-bot commented Mar 4, 2023

AWS CodeBuild CI Report

Uh oh!

sagemaker-bot commented Mar 4, 2023

AWS CodeBuild CI Report

Uh oh!

sagemaker-bot commented Mar 4, 2023

AWS CodeBuild CI Report

Uh oh!

sagemaker-bot commented Mar 4, 2023

AWS CodeBuild CI Report

Uh oh!

sagemaker-bot commented Mar 4, 2023

AWS CodeBuild CI Report

Uh oh!

Uh oh!

patrickmcarlos commented Feb 14, 2023 •

edited

Loading

codecov-commenter commented Feb 16, 2023 •

edited

Loading