Understanding the scope, drivers, and energy justice implications of solar adoption in the United States wastewater sector using applied computer vision and satellite imagery
While tagging, as expected, I see duplicate WWTP candidates, since we did decide to include "part-of-WWTP" images into the "Yes" bucket, as well as combining different datasets.
Which means two things:
our model correctly identifies them, given they have few differences. Cheer!
we need to deduplicate after we have completed our verification list. As we discussed, we can do a 1-km-wise comparison of different WWTP locations. After that, we want to integrate location/lat-long info and other details into the dataset.
Hi all,
While tagging, as expected, I see duplicate WWTP candidates, since we did decide to include "part-of-WWTP" images into the "Yes" bucket, as well as combining different datasets.
Which means two things: