filter component commit

tensorflow / tfx-addons

Developers helping developers. TFX-Addons is a collection of community projects to build new components, examples, libraries, and tools for TFX. The projects are organized under the auspices of the special interest group, SIG TFX-Addons. Join the group at http://goo.gle/tfx-addons-group

Apache License 2.0

125 stars 64 forks source link

filter component commit #158

Closed vulkomilev closed 1 year ago

vulkomilev commented 2 years ago

Fixes #

It's a good idea to open an issue first for discussion.

[ ] Tests pass
[ ] Appropriate changes to README are included in PR

github-actions[bot] commented 2 years ago

Thanks for the PR! :rocket:

Instructions: Approve using /lgtm and mark for automatic merge by using /merge.

review-notebook-app[bot] commented 2 years ago

Check out this pull request on

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

casassg commented 2 years ago

Also generally, why not use a Beam pipeline to do so vs this numpy filter?

rcrowe-google commented 2 years ago

I don't understand. Iris_example_colab.ipynb is the example notebook from the feature selection component. Why is it included here?

vulkomilev commented 2 years ago

I don't understand. Iris_example_colab.ipynb is the example notebook from the feature selection component. Why is it included here?

it will be removed.

vulkomilev commented 1 year ago

made a new push @casassg

hanneshapke commented 1 year ago

@vulkomilev Here are the steps to finish the component PR:

Update copy right in all files of your component
Update tests, make sure they run and remove anything Pandas Transform related
run the linter and make sure that the linter is green on Github!
update the example notebook with an example csv file, load the csv data, and apply the filter component, and print the content of the TFRecords (the first 4 samples)
resolve all conversations above which you have addressed, leave the ones open for which you have questions. That helps to quickly review the PR.

Thank you!