-
Notifications
You must be signed in to change notification settings - Fork 65
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
Add fix for genea native dst support #211
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @angerhang I just had one comment (see review)
Have you tried with a file with DST crossover in the other direction?
@angerhang Actually before we merge this, I would like to test it with files where DST crossover actually happened (usually April and October). Can you do this? Or PM me with two such files for me to try. |
This is a good idea but unfortunately, we don't have access to any files that have DST crossover. https://zenodo.org/record/1160410#.YzODmC8w1qs Perhaps you are aware that there are some data files sitting around for Geneactive. |
Yes I have done the crossover validation in the other direction with GGIR both in the winter and the summer. |
@angerhang Do you want to test your git-fu and rebase the latest changes in master onto your PR? |
I've rebased master into the branch. What's git-fu? |
nvm! it's kung-fu but for git 😆 |
FYI you've merged master into your branch, not rebased your branch onto master (hence the extra merge commit). No big deal, but a rebase would've been the 'clean' way. But Shing can now 'fix' it by doing a squash merge 🥳 |
You read my mind :D I was just doing that + squashing both commits |
oh I guess I am not to supposed to merge the PR? Or clicked the wrong button? lolol |
Anyhow, thanks for reviewing the PR :D Now I am unblocked. |
@angerhang |
Currently GENEActive processing doesn't account for timezone or DST. This will create issues when the data was not recored in UTC.
This PR address the issue above by
timeZone
to the processing pipelinenewValue
function whose input is the true UNIX timeTest plan:
Test file is from a sample Newcastle file
/well/doherty/projects/newcastle_psg/acc/1.bin
Original acc plot
New acc plot with time zone and DST
The introduced change correctly move the time series forward by one hour in the summer. The the class prediction might be different possibly due to that the original plot was using an older version of bbaa. Based on the PSG data, this subject was awake around 07:30 which is consistent with the new processing. R stands for REM sleep and W stands for wake.
Disclaimer: Couldn't do a PR from my fork because my fork master has some important changes with the upstream doesn't have.
Resolves #203
Please could you review ASAP because I am blocked by the Geneactive data processing to retrain the model. Thank you :D