[ZIT] Angkor Wat

Generates backgrounds styled after the ruins of Angkor Wat, a massive, nearly nine hundred year old, temple complex in what is now Cambodia. I wanted to practice writing natural language captions, so I picked a dataset I had where I could show my work and get feedback, and a PD+CC based dataset seemed good to welcome a new year of public domain stuff. May train a tag based version for ZIT next week for comparison of the final results. Trained a tag version using same dataset as Illustrious version, minus the one picture I dropped in the natural language version

Natural Language:

Trigger word: Ankohl (e.g., a photo inside Ankohl)
Used words: carved stone, raised doorway, broken statue, stone head, spire

Tag:

Trigger: angkol

All training data was public domain or Creative Commons, so I have released the training data (or tried to: forgot to click share dataset in training. I think deleting then reuploading data fixes this, will have to figure out something if it doesn't) and made LoRA permissions as open as possible (do note that you'll need to obey CC-By if you reuse the dataset). Below are the credits for the CC-by images. I dropped one image from the previous versions because it had too many tourists.

CC-By credits:
Modification: These were obviously used to train the resource. I don't think they were knowingly modified beyond resizing and compression.
by H. Grobe
by Christine Zenino from Chicago, US
by Photo Dharma from Penang, Malaysia
by Olaf Tausch
by Jorge Láscar from Australia
by Jorge Láscar from Australia
by cloud.shepherd
by Allie Caulfield
by Allie Caulfield
by Sasha India
by Sasha India
by Sasha India
by Sasha India
by Sasha India
by Sasha India
by Sasha India
by Sasha India
by Zz Lighting