fix: SRCNN impl and tests

[Tohrusky]

Summary by Sourcery

Fix the SRCNN implementation by correcting the RGB to YUV conversion condition and add tests for utility functions to ensure proper error handling and functionality.

Bug Fixes:

• Fix the condition in the SRCNN architecture to correctly handle RGB to YUV conversion only when the input tensor has 3 channels.

Tests:

• Add new tests for utility functions including device configuration and color conversion functions, ensuring they handle errors and perform conversions correctly.

[sourcery-ai]

Reviewer's Guide by Sourcery

This PR fixes the SRCNN implementation by adjusting the color space conversion logic and adds comprehensive tests for color conversion utilities. The main changes include a condition update in the SRCNN forward pass and new test coverage for RGB-YUV conversion.

Class diagram for new test utilities

classDiagram
    class TestUtilities {
        + test_device(): None
        + test_color(): None
    }
    class ColorConversion {
        + rgb_to_yuv(input)
        + yuv_to_rgb(input)
    }
    TestUtilities --> ColorConversion : uses
    note for TestUtilities "New tests for color conversion utilities"

Loading

File-Level Changes

Change	Details	Files
Modified SRCNN forward pass color space conversion logic	Updated condition to check both number of channels and input tensor channels Added additional check for x.size(1) == 3 to ensure proper RGB input handling	ccrestoration/arch/srcnn_arch.py
Added comprehensive test coverage for color conversion utilities	Added type checking tests for rgb_to_yuv and yuv_to_rgb functions Added tensor dimension validation tests Implemented end-to-end color conversion test with image similarity check	tests/test_util.py
Updated coverage configuration	Added device utility module to coverage omit list	pyproject.toml

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time. You can also use
  this command to specify where the summary should be inserted.

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[sourcery-ai]

Hey @Tohrusky - I've reviewed your changes and they look great!

Here's what I looked at during the review

• 🟢 General issues: all looks good
• 🟢 Security: all looks good
• 🟡 Testing: 2 issues found
• 🟢 Complexity: all looks good
• 🟢 Documentation: all looks good

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

[sourcery-ai]

issue (testing): Test function test_device only prints the device without any assertions

This test doesn't verify any behavior. Consider adding assertions to check if DEFAULT_DEVICE is set correctly or has expected properties.

[sourcery-ai]

suggestion (testing): Test function test_color could benefit from parametrization and more specific assertions

Consider splitting this test into multiple test cases: one for error conditions, one for RGB->YUV conversion, and one for YUV->RGB conversion. Also, consider adding assertions about the specific values or ranges in the YUV color space to ensure the conversion is correct.

@pytest.mark.parametrize("invalid_input", [1, "string", [1, 2, 3]])
def test_color_type_errors(invalid_input) -> None:
    with pytest.raises(TypeError):
        rgb_to_yuv(invalid_input)
    with pytest.raises(TypeError):
        yuv_to_rgb(invalid_input)

@pytest.mark.parametrize("invalid_shape", [torch.zeros(1, 1), torch.zeros(2, 3, 4)])
def test_color_value_errors(invalid_shape) -> None:
    with pytest.raises(ValueError):
        rgb_to_yuv(invalid_shape)
    with pytest.raises(ValueError):
        yuv_to_rgb(invalid_shape)

def test_color_conversion() -> None:
    img = load_image()
    img_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
    img_tensor = transforms.ToTensor()(img_rgb).unsqueeze(0).to("cpu")

    yuv = rgb_to_yuv(img_tensor)
    assert yuv.shape == img_tensor.shape
    assert (yuv[:, 0] >= -1).all() and (yuv[:, 0] <= 1).all()  # Y channel bounds

    rgb = yuv_to_rgb(yuv)
    rgb_np = (rgb.squeeze(0).permute(1, 2, 0).cpu().numpy() * 255).clip(0, 255).astype("uint8")
    rgb_np = cv2.cvtColor(rgb_np, cv2.COLOR_RGB2BGR)

    assert calculate_image_similarity(rgb_np, img)