Skip to content

Conversation

workingloong
Copy link
Collaborator

Fix #1844

Arguments:
ignore_value: Entries in inputs equal to this value will be
absent from the output `SparseTensor`. If `None`, default value of
inputs dtype will be used ('' for `str`, -1 for `int`).
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should we expose this?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

-1 for int seems dangerous as this is application specific

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

-1 for int is the default value and the layer will ignore it during transformation.

Copy link
Collaborator

@brightcoder01 brightcoder01 Mar 25, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

-1 for int seems dangerous as this is application specific

Use -1 as the default ignore_value for int type is also the implementation inside feature column. Please check the code snippet.

@brightcoder01 brightcoder01 merged commit 29a8060 into sql-machine-learning:develop Mar 26, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Support convert categorical features to SparseTensor for embedding in Keras

4 participants