Changepoint Blog and FlaSH IJCAI Paper to Publications #904
Conversation
Ananya-Joshi
requested review from
rlunde,
melange396 and
carlynvandyke
as code owners
|
I am not sure how to debug this issue, but is it a problem with Hugo? https://discourse.gohugo.io/t/nil-pointer-evaluating-resource-resource-resize-in-version-0-57-0/20208/4 |
✅ Deploy Preview for cmu-delphi-main ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
|
Images are still not loading |
|
@rlunde @carlynvandyke @melange396 Happy new year everyone ! Would love to get your suggestions and comments on this blog so that it is ready to push. |
|
Two tiny nitpicks in the last section: |
Thank you for those fixes! I just changed them and hopefully made the leading/lagging sentence a bit more clear (but please let me know what you think!) |
There was a problem hiding this comment.
ive got a number of suggestions that you can take or leave, but it looks great!
also, this is probably not a big deal, but it might not be completely future-proof if our stylesheets change: in the blog article, image3.png and image5.png do not have transparent backgrounds, but images 1, 2, and 4 do (and none of those outside the article have transparency).
| - **On-time**: The major indicator changepoints appear within a week of the ground truth changepoints | ||
| - **Late**: The major indicator changepoints lag behind the ground truth changepoints by more than a week. | ||
|
|
||
| As we expected each wave of the pandemic to last around four weeks, we necessitate that any major indicator changepoints found with a distance greater than 28 days to the nearest ground truth changepoint have no relationship with the ground truth changepoint. |
There was a problem hiding this comment.
im not sure i understand this... what do you mean by "necessitate"? are you saying that any indicator that exhibits this behavior gets classified into "inconsistent/variable" or "undetermined" categories that you talk about later?
There was a problem hiding this comment.
I just added some clarification:
As we expect COVID dynamics (like waves) to last around four weeks, if any major indicator changepoints is more than 28 days away from the nearest ground truth changepoint, it is likely a changepoint from an unrelated phenomena. Therefore, we only consider changepoints to have a leading or lagging relationship if the major indicator changepoint is within 28 days from a ground truth changepoint.
What do you think?
There was a problem hiding this comment.
much better! id also change changepoints is to changepoint is
Co-authored-by: melange396 <george.haff@gmail.com>
Co-authored-by: melange396 <george.haff@gmail.com>
Co-authored-by: melange396 <george.haff@gmail.com>
Co-authored-by: melange396 <george.haff@gmail.com>
Co-authored-by: melange396 <george.haff@gmail.com>
Co-authored-by: melange396 <george.haff@gmail.com>
Co-authored-by: melange396 <george.haff@gmail.com>
Co-authored-by: melange396 <george.haff@gmail.com>
|
in the preview, you can see that now |
|
Oops yes - should I revert to the other images - Tara generated them so I'm not sure if there's an easy way to add back the background without the source without screen capture. What do you think would be best? |
im not sure... you can either go back to the transparent ones or redo the screencaps, or i think its possible to export the png without the alpha channel. let me try that and get back to you. EDITED TO ADD: that worked! ill send them to you in slack (because im lazy and thats easier than doing a PR into your fork!) |
|
They look great - thank you George ! |
Adding a blog post with Tara and updating the IJCAI publication for FlaSH