Skip to content

Feature/kedro 1.0#45

Open
dermatologist wants to merge 9 commits into
developfrom
feature/kedro-1.0
Open

Feature/kedro 1.0#45
dermatologist wants to merge 9 commits into
developfrom
feature/kedro-1.0

Conversation

@dermatologist

Copy link
Copy Markdown
Owner

No description provided.

@dermatologist
dermatologist requested a review from Copilot August 23, 2025 14:49

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Pull Request Overview

This pull request upgrades a Kedro project from version 0.18.x to Kedro 1.0, modernizing the codebase structure and dependencies. The upgrade involves significant changes to align with Kedro's new patterns and conventions.

  • Upgrades to Kedro 1.0 with new session-based architecture and OmegaConfigLoader
  • Updates TensorFlow/Keras imports to use standalone keras package instead of tensorflow.keras
  • Modernizes dataset classes with new AbstractVersionedDataset base class and type annotations

Reviewed Changes

Copilot reviewed 54 out of 64 changed files in this pull request and generated 6 comments.

Show a summary per file
File Description
tests/test_run.py Completely new test file with updated Kedro session API and modernized test patterns
tests/pipelines/preprocess/test_pipeline.py Updated test structure removing deprecated KedroContext pattern
src/kedro_tf_image/settings.py Migrated to Kedro 1.0 settings format with OmegaConfigLoader
src/kedro_tf_image/pipelines/preprocess/nodes.py Updated keras imports and improved type annotations
src/kedro_tf_image/extras/datasets/*.py Modernized dataset classes with new base classes and type hints
pyproject.toml Complete project configuration overhaul for Kedro 1.0

Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.

Comment on lines +38 to +39
tf.TensorSpec(shape=(None, imagedim, imagedim, 3), dtype=tf.float32, name=None), # type: ignore
tf.TensorSpec(shape=(None,), dtype=tf.int32, name=None), # type: ignore

Copilot AI Aug 23, 2025

Copy link

Choose a reason for hiding this comment

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

[nitpick] The # type: ignore comments suggest type checking issues. Consider addressing the underlying type issues rather than ignoring them, or add more specific type ignore codes.

Suggested change
tf.TensorSpec(shape=(None, imagedim, imagedim, 3), dtype=tf.float32, name=None), # type: ignore
tf.TensorSpec(shape=(None,), dtype=tf.int32, name=None), # type: ignore
tf.TensorSpec(shape=cast(tuple[int | None, ...], (None, imagedim, imagedim, 3)), dtype=tf.float32, name=None),
tf.TensorSpec(shape=cast(tuple[int | None, ...], (None,)), dtype=tf.int32, name=None),

Copilot uses AI. Check for mistakes.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants