Hi @<1523703107031142400:profile|FlatOctopus65> , can you please elaborate on what exactly happens and when? Do you have a snippet to play with ?

hi, @<1523701070390366208:profile|CostlyOstrich36>

sure:

from clearml.automation.controller import PipelineDecorator


@PipelineDecorator.component(return_values=['img'])
def get_image(image_path):
    import cv2
    img = cv2.imread(image_path)
    img = img[..., ::-1]
    return img


@PipelineDecorator.pipeline(
    project='test',
    name='test pipeline',
    version='0.0.1',
    return_value='result',
)
def executing_pipeline(params):
    print(params)
    result = get_image(params['image_path'])
    return result


if __name__ == '__main__':
    params = {'image_path': ''}
    PipelineDecorator.run_locally()
    executing_pipeline(params)

python test.py will results in:

Launching step [get_image]
  File "/tmp/tmp7drwjuy0.py", line 8
    img = img[(..., ::(- 1))]
                    ^
SyntaxError: invalid syntax

@<1523701070390366208:profile|CostlyOstrich36> sorry for the typo

Thank you!

Hi @<1523703107031142400:profile|FlatOctopus65> ! python3.9 introduced a breaking change for codebases that parse code containing slices. You can read more about it here: None . Notable:

* The code that produces a Python code from AST will need to handle indexing with tuples specially (see Tools/parser/unparse.py) because d[(a, b)] is valid syntax (although parenthesis are redundant), but d[(a, b:c)] is not.

What you could do is downgrade to Python 3.8 or use the slice function:
img = img[..., slice(None, None, -1)]
Hope this workaround works for you until we fix this