The script will copy the files, ignoring duplicates, and format them as follows. Files are renamed to their date of creation (when the photo/video was taken/shot), e.g. 2019.01.07_08.28.34.jpg, and then stored in folders sorted by year.
ETA is 40 hours for the remaining 150GB. I suspect that the bottleneck is my network (running the script on my PC, but the photos are available through NFS). I assume it would go much faster if photos are stored on the same computer as the script is being run on.
Anyway, I thought this might be relevant for anyone who wants to get a grip of their photos :)
PS: would greatly appreciate style/performance/correctness tips on the code AND recommendations of any other naming strategy you deem better than my approach!