-
Notifications
You must be signed in to change notification settings - Fork 171
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
clean_flicker_noise skipped when called as a standalone step #9258
Comments
I think that's known behavior- I've seen this with some other off-by-default steps as well. Not sure if there's an easy way to handle this other than setting skip=False when running the step standalone? Tagging Tyler Pauly |
Comment by Tyler Pauly on JIRA: This is the designed behavior - because the step has been added to the detector1 pipeline, its default behavior is that skip is set to True. This is so that parameter references can be delivered by instrument teams to enable it on a case-by-case basis. The alternative, to have skip default to False, would require parameter reference file updates to set it to True for every mode until testing demonstrates a positive outcome. Once testing is complete, we could request that those parameter reference updates are made - the ideal scenario would be for skip to have a default value of False to avoid this confusing situation. But as it is still considered to be RTT rather than science-ready, having it skip by default is not entirely unreasonable. |
Comment by Tyler Pauly on JIRA: I think we'd need to test to be sure, but my understanding is that any parameter-specific behavior we'd insert into the pipeline code would overwrite fetched values from CRDS. This why pipelines use .run() in the internals - the classes have already been instantiated and configured according to the params fetched, then overridden if necessary by user prompts. |
Right, I think the aim would be if the pipeline could tell that a single step is being called directly (whether through strun or a python .call() approach) it seems generally clear that the user wants to run that step regardless of what the regular skip settings are. It may not be worth pursuing this, but worth asking how complicated that might be (and if it might end up biting us in some unexpected way). |
Issue JP-3912 was created on JIRA by Bryan Hilbert:
While running some pipeline steps on some dark current exposures today, I noticed some odd behavior with the clean_flicker_noise step. When I called the individual step, it still ended up being skipped unless I explicitly set skip=False.
Here is my call:
f = CleanFlickerNoiseStep.call(filename,
n_sigma=15,
fit_by_channel=True,
background_method='model',
save_background=True,
save_mask=True,
save_results=True
)
And the resulting log messages are below. Note that in the list of input parameter values, skip is set to True, and indeed, the step is skipped. Only if I add skip=False to my call above is the step executed.
2025-03-07 15:46:29,428 - stpipe.CleanFlickerNoiseStep - INFO - CleanFlickerNoiseStep instance created.
2025-03-07 15:46:29,546 - stpipe.CleanFlickerNoiseStep - INFO - Step CleanFlickerNoiseStep running with args ('../jw06626001001_02201_00001_nrcalong_no_donotuse_cropped20groups_jump.fi
ts',).
2025-03-07 15:46:29,547 - stpipe.CleanFlickerNoiseStep - INFO - Step CleanFlickerNoiseStep parameters are:
pre_hooks: []
post_hooks: []
output_file: None
output_dir: None
output_ext: .fits
output_use_model: False
output_use_index: True
save_results: True
skip: True
suffix: None
search_output_file: True
input_dir: ''
fit_method: median
fit_by_channel: True
background_method: model
background_box_size: None
mask_science_regions: False
n_sigma: 15
fit_histogram: False
single_mask: True
user_mask: None
save_mask: True
save_background: True
save_noise: False
2025-03-07 15:46:29,549 - stpipe.CleanFlickerNoiseStep - INFO - Step skipped.
The text was updated successfully, but these errors were encountered: