Experiencing under- and over-fitting in Sector 10

[barentsen]

I appear to experience severe under- and over-fitting of the background signals when applying tess-backdrop on Sector 10 data (using backdrop files obtained from the zip file sector10_jul6.zip).

The screenshot below shows an example for one star, but I encountered the issue with the every Sector 10 target I tried. For example:

• TIC 150428135 (underfit)
• TIC 178819686 (underfit)
• TIC 260004324 (overfit)
• TIC 280206394 (overfit)
• TIC 339672028 (overfit)

[christinahedges]

I think the answer here -might- be that the driver files are for a different sector, but I need to do some digging!

update: I think this is the right data, but the early images from this sector are really blown out e.g. see

https://www.youtube.com/watch?v=hbGg8oU1qNM&ab_channel=EthanKruse

So it may be that I need to think more about how to deal with those frames...