Sup mate, glad to see AI tagging come along! I ran into an issue with the 540 version on Linux (been using Qt5 before that, 490 something): I didn't get to the file storage upgrades I wanted this week, so I'll give it another go. This stuff is just getting ready to seriously play with, so if you are interested, give it a go! These use the recent 'I give text, it draws an image' AI tech everyone has been messing around with to do the reverse-you give it an image, and it automatically tags it. I added links to (and, which it is based on) to the Client API help. Also, I've started a background job today to fix some lingering bad duplicate pixel data for pngs, so if you have recently done a duplicate filter with 'must not be pixel dupes' but got some anyway, I hope this will fix itself soon. I fixed a bunch of broken 'system:similar files' image bitmap pastes that were not producing the correct pixel hash. if you right-click->copy it) should now parse ok. I think that any existing filetype predicate string (e.g. You can type 'system:filetype is image' or 'system:filetype is static gif, png' and it should just work. System:filetype parsing now supports more human-friendly words.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |