Very cool tool! It's generally accurate, but, there are spots showing shadows where there are no trees or anything at all to cast a shadow, and missing several spots where there are actually trees.
It's missing a 40' tall ash, a 40' tall maple, and all four 30' tall blue spruce. So they're not exactly small trees. Our septic field is showing as in shadow all day even though there are no trees near it. That shadow doesn't move at all like the rest so I'm not sure what it's detecting.
Sorry about that. The current tree estimate is done via computer vision but I am working on migrating to LiDAR backed data which should eliminate false positives.
Just waiting for the TBs and TBs of data to download and then this Macbook M4 is going to make some heat...
Yea, we have several applications on heroku. Only failures are the ones that had platform initiated restarts after 5pm MST. Likely tried to get a new cert on restart and then fails. Don't deploy or do manual restarts right now.
I know this gets asked all the time, but what is your preferred workflow when using local models? I was pretty deep into it early on, with Tabby and Continue.dev, but once I started using Claude Code with Opus it was hard to go back. I do the same as you, and still use them on flights and whatnot, but I think my implementation could be improved.
It's missing a 40' tall ash, a 40' tall maple, and all four 30' tall blue spruce. So they're not exactly small trees. Our septic field is showing as in shadow all day even though there are no trees near it. That shadow doesn't move at all like the rest so I'm not sure what it's detecting.