tfm.uplift.layers.heads.two_tower_logits_head.LayeringConfig
Stay organized with collections
Save and categorize content based on your preferences.
Configuration for all layering methods.
tfm.uplift.layers.heads.two_tower_logits_head.LayeringConfig(
layering_method: LayeringMethod = LayeringMethod.NONE,
linear_layering_config: (LinearLayeringConfig | None) = None
)
Attributes |
layering_method
|
specifies what layering method to apply. Defaults to
LayeringMethod.NONE.
|
linear_layering_config
|
configuration for the linear layering method. Will
only be used if the layering_method is set to
LayeringMethod.LINEAR_LAYERING.
|
Methods
from_dict
View source
@classmethod
from_dict(
config: dict[str, Any]
) -> LayeringConfig
__eq__
__eq__(
other
)
Return self==value.
Class Variables |
|
layering_method
|
<LayeringMethod.NONE: 'none'>
|
|
linear_layering_config
|
None
|
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates. Some content is licensed under the numpy license.
Last updated 2026-11-06 UTC.
[null,null,["Last updated 2026-11-06 UTC."],[],[]]