-
Notifications
You must be signed in to change notification settings - Fork 1k
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
Fix materialize for None #1481
Fix materialize for None #1481
Conversation
Signed-off-by: qooba <dev@qooba.net>
Hi @qooba. Thanks for your PR. I'm waiting for a feast-dev member to verify that this patch is reasonable to test. If it is, they should reply with Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
Signed-off-by: qooba <dev@qooba.net>
/ok-to-test |
/kind feature |
Hey @qooba. Thanks again for the PR. This change seems reasonable. FYI you can run |
Signed-off-by: qooba <dev@qooba.net>
@qooba would you mind updating one of our tests so that we explicitly test for this condition? |
Signed-off-by: qooba <dev@qooba.net>
Sure :) @woop I have added checks to existing tests.
I'm not sure if such conversion should be added to feast itself (on the other hand different offline stores can behave differently and it should be consistent). |
/retest |
Signed-off-by: qooba <dev@qooba.net>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hey @qooba, this is great. Good catch on the None to NaN conversion that happens with pandas->parquet->pandas. Since we want results from the online and offline stores to be consistent, I think we should have the get_historical_features
call just return Nones rather than converting NaNs to Nones within the test, i.e., we'll want to convert in sdk/python/feast/infra/offline_stores/file.py
. I think for the BigQuery offline store we're ok since BigQuery nulls will turn into pandas Nones. With that change this should be good to go, thanks for the contribution!
Hi, @jklegar . I have checked native pandas methods |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Cool it seems like that's the best thing to go with right now then. Thanks @qooba!
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: jklegar, qooba The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
/lgtm |
* fix materialize for None Signed-off-by: qooba <dev@qooba.net> * fix materialize for None Signed-off-by: qooba <dev@qooba.net> * fix materialize for None Signed-off-by: qooba <dev@qooba.net> * fix materialize for None - add test Signed-off-by: qooba <dev@qooba.net> * Fix materialize - correct lint errors Signed-off-by: qooba <dev@qooba.net>
* fix materialize for None Signed-off-by: qooba <dev@qooba.net> * fix materialize for None Signed-off-by: qooba <dev@qooba.net> * fix materialize for None Signed-off-by: qooba <dev@qooba.net> * fix materialize for None - add test Signed-off-by: qooba <dev@qooba.net> * Fix materialize - correct lint errors Signed-off-by: qooba <dev@qooba.net>
Signed-off-by: qooba dev@qooba.net
What this PR does / why we need it:
This pull request fixes the issue during materialize view.
If in input data there is a None value feast can't parse its type in python_type_to_feast_value_type method
Which issue(s) this PR fixes:
Fixes #1480
Does this PR introduce a user-facing change?: