Add xception model

[mattdangerw]

Just porting over from Keras applications.

[Copilot]

Pull Request Overview

The purpose of this PR is to add support for the Xception model in Keras Hub by porting over components from Keras Applications.

• Introduces a new checkpoint conversion script for the Xception model.
• Adds new presets, image converter, classifier, and backbone implementations along with their tests.
• Updates API exports to include the new Xception components.

Reviewed Changes

Copilot reviewed 11 out of 11 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
tools/checkpoint_conversion/convert_xception_checkpoints.py	Adds a script to convert Xception checkpoints with a few minor docstring/typo issues.
keras_hub/src/models/xception/xception_presets.py	Provides preset configuration for the Xception model.
keras_hub/src/models/xception/xception_image_converter.py	Exports the Xception image converter implementation.
keras_hub/src/models/xception/xception_image_classifier_test.py	Adds tests for the Xception image classifier and contains a variable naming typo.
keras_hub/src/models/xception/xception_image_classifier_preprocessor.py	Implements preprocessor using the new Xception image converter.
keras_hub/src/models/xception/xception_image_classifier.py	Introduces the Xception image classifier.
keras_hub/src/models/xception/xception_backbone_test.py	Provides tests for the Xception backbone.
keras_hub/src/models/xception/xception_backbone.py	Implements the Xception backbone, including an example snippet with a preset naming inconsistency.
keras_hub/src/models/xception/init.py	Registers the Xception backbone presets.
keras_hub/api/models/init.py	Updates API exports to include Xception components.
keras_hub/api/layers/init.py	Exports the Xception image converter via the API.

Comments suppressed due to low confidence (2)

tools/checkpoint_conversion/convert_xception_checkpoints.py:1

• The module-level docstring appears to have mismatched quotes. Please update it to use proper triple quotes (e.g., """ ... """).

'''Convert Xception model to Keras Hub."

keras_hub/src/models/xception/xception_backbone.py:44

• The example preset name 'exception_41_imagenet' appears inconsistent with the actual preset 'xception_41_imagenet'. Please update the example to reflect the correct preset name.

model = keras_hub.models.Backbone.from_preset("exception_41_imagenet")

[Copilot]

Pull Request Overview

This PR ports the Xception model from Keras applications to Keras Hub while adding the necessary conversion tools, presets, and tests.

• Add a checkpoint conversion script for the Xception model.
• Introduce Xception backbone, image converter, image classifier, preprocessor, and presets along with tests.
• Update API registration for the new Xception modules.

Reviewed Changes

Copilot reviewed 11 out of 11 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
tools/checkpoint_conversion/convert_xception_checkpoints.py	Adds a conversion script for mapping Keras Xception model weights.
keras_hub/src/models/xception/xception_presets.py	Defines presets for the new Xception model.
keras_hub/src/models/xception/xception_image_converter.py	Introduces an image converter for the Xception backbone.
keras_hub/src/models/xception/xception_image_classifier_test.py	Adds tests for the Xception image classifier functionality.
keras_hub/src/models/xception/xception_image_classifier_preprocessor.py	Implements the preprocessor for the Xception image classifier.
keras_hub/src/models/xception/xception_image_classifier.py	Registers the Xception image classifier.
keras_hub/src/models/xception/xception_backbone_test.py	Provides tests for the Xception backbone.
keras_hub/src/models/xception/xception_backbone.py	Implements the Xception backbone with custom configurable layers.
keras_hub/src/models/xception/init.py	Registers the Xception backbone and presets.
keras_hub/api/models/init.py	Updates API registration with Xception models.
keras_hub/api/layers/init.py	Updates API registration with the Xception image converter.

Comments suppressed due to low confidence (2)

tools/checkpoint_conversion/convert_xception_checkpoints.py:1

• The docstring on line 1 is not properly closed with triple quotes; please fix it to ensure proper string termination.

'''Convert Xception model to Keras Hub."

keras_hub/src/models/xception/xception_image_classifier_test.py:30

• [nitpick] The variable 'self.prepocessor' appears to be misspelled; consider renaming it to 'self.preprocessor' for clarity.

self.prepocessor = XceptionImageClassifierPreprocessor(

[divyashreepathihalli]

LGTM!