Code analyses still coded as 'CHECK'

[egouldo]

Hi Hannah or @parkerth,

Could you please check the master_data file (at "data-raw/anonymised_data/master_data.csv") and code as necessary or mark as NA if can't resolve? there are a few analyses remaining with values 'CHECK':

response_id	submission_id	analysis_id	split_id	test_variable	Bayesian	linear_model	model_subclass	exclusions_effect_analysis	Conclusion	data_cleaning_preprocessing_tool	data_cleaning_preprocessing_version	data_analysis_tool	data_analysis_version
R_11787O3NmejXKAH	1	2	2	net_rearing_manipulation	0	generalised	standard	exclude_all	CHECK	R	NA	R	NA
R_11787O3NmejXKAH	1	2	3	net_rearing_manipulation	0	generalised	standard	exclude_all	CHECK	R	NA	R	NA
R_11787O3NmejXKAH	1	2	1	net_rearing_manipulation	0	generalised	standard	exclude_all	CHECK	R	NA	R	NA
R_1eXlFKlQdiD2F59	2	2	1	rear_Cs_at_start_of_rearing	0	linear	standard	retain	CHECK	CHECK	CHECK	CHECK	CHECK
R_1GJlffAgZv6SY4y	1	1	1	NA	CHECK	CHECK	standard	exclude_all	none_c	R	4.0.0	R	4.0.0
R_1M0cMZL2IPYWHoi	1	1	1	NA	CHECK	CHECK	CHECK	exclude_all	CHECK	CHECK	CHECK	CHECK	CHECK
R_1M0cMZL2IPYWHoi	1	1	1	NA	1	generalised	CHECK	exclude_all	NA	R	3.6.3	R	3.6.3
R_1QlnXdW5tKuUQIr	1	1	1	CHECK	CHECK	CHECK	CHECK	CHECK	neg_q	R	3.6.1	R	3.6.1
R_21gmMa0uclrNoTP	2	1	1	NA	CHECK	linear	CHECK	exclude_all	CHECK	CHECK	CHECK	CHECK	CHECK
R_2Pjoz1X4q5XRClO	2	1	1	NA	CHECK	CHECK	standard	exclude_all	CHECK	CHECK	CHECK	CHECK	CHECK
R_2zNKAmJcWbM4QtY	1	1	1	NA	0	CHECK	standard	retain	none_q	R	3.6.1	R	3.6.1
R_3EbbZxcQ3gctVZu	1	1	1	NA	0	CHECK	hurdle	exclude_all	neg_q	R	3.6.1	R	3.6.1
R_3Kvy0h01LXHWniT	2	2	1	NA	0	generalised	standard	retain	CHECK	R	NA	R	NA
R_3nBCE4hMLh5s3qt	3	1	2	NA	CHECK	CHECK	standard	NA	CHECK	CHECK	CHECK	CHECK	CHECK
R_3nBCE4hMLh5s3qt	3	1	1	NA	CHECK	CHECK	standard	NA	CHECK	CHECK	CHECK	CHECK	CHECK
R_AzL6RdNTHtPjxzX	1	1	2	NA	CHECK	CHECK	standard	exclude_all	CHECK	CHECK	CHECK	CHECK	CHECK
R_es2jrrN9CTGwl5D	1	1	1	NA	CHECK	CHECK	CHECK	exclude_all	neg_q	R	4.0.0	R	4.0.0

[parkerth]

@egouldo
not sure what to do here. The rows with "CHECK" in the Bayesian and/or linear_model column are all 'exclude' (or in one case, 'check.mixed.team').
I guess I will go ahead and change them to NA, but it seems like there might be a bigger problem with the code if this matters. Is the 'exclude' code working?

[parkerth]

... also true for 'model_subclass' variable (all checks are also exclusions)

[parkerth]

... also true for these columns:

Conclusion | software | data_cleaning_preprocessing_tool | data_cleaning_preprocessing_version | data_analysis_tool | data_analysis_version

[parkerth]

I'm uploading a version of the master_data.csv file with all "CHECK" replaced by "NA". Just let me know if, instead, we need to revert to the former version with the "CHECK" values

[parkerth]

That file is now on branch(?) with a pull request(?) (if my use of jargon is correct)

[hannahsfraser]

Awesome stuff Tim. You're fast like a fox!

[egouldo]

Thanks Tim, all while I was sleeping. So that’s not the source of the bug I haven’t yet solved, but I was doing some processing on the full dataset before exclusion. That being said, I will check for instances where the exclusions haven’t been applied where they should be. Elliot.

…

Sent from my iPhone On 18 Jun 2024, at 12:42 AM, Hannah Fraser ***@***.***> wrote:  Awesome stuff Tim. You're fast like a fox! — Reply to this email directly, view it on GitHub<#43 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACAC6KRY7COLLO6R2OH2PWLZH5X65AVCNFSM6AAAAABJMUWHY6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNZUGYZTCOBZGU>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[parkerth]

That makes sense. Where do we stand at this point?

[egouldo]

Note that Capella-1-1-1 was marked as CHECK.mixed.team for exclusions_all. which was replaced with NA.mixed.team, and then I just did a replacement with NA across multiple columns for this entry in prev. commit, b00d31f. But that probably means we should check whether we want to include it at all @parkerth ? Note that I did also notice that the response_variable_name was coded incorrectly as "euc_sdlgs>2m" in the course of checking #80 and have replaced it. So I think this entry probably needs double checking anyways.

[egouldo]

Looks like has been fixed, row has been marked as 'exclude' and 'exclude_all' under relevant exclusion columns.