-
Notifications
You must be signed in to change notification settings - Fork 58
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
Related to issue #112: Exploration failure for weekly data #113
Conversation
Hey guys, I requested this change because I was trying to call that function, and it didn't work. I have a quick question regarding the modification here. I was expecting something like this to change.
which is the frequency to get to the immediately higher time interval. Why did you do |
@papaemman |
@sayanchk I was running some experiments, and the Only if you give Kindly re-check the PR. I think we need to go with the solution I mentioned here, meaning that we need to give a starting point for the Thank you! |
@papaemman the logic behind Can you share more about the example that you tried for testing the |
Hey @sayanchk, Sorry for the late response. I run some experiments setting both First case: W=4The code snippet was
This gave me reasonable trends and changepoints for multiple time series. For example, Second case: W=NoneThe code snippet was
This didn't return any trends and changepoints at all for multiple time series (even for obvious settings). For example, ConclusionIn conclusion, I believe that this PR isn't correct, and we need to go with the W=4 solution. I create a new PR for this. You can check more in my repository here: https://github.com/papaemman/luminaire_demo Thanks, |
The current approach for trend turning was enabled only for daily and hourly time series. Added a quick fix to trigger computation of window sizes for other frequency types.