Type | |
Stats | 729 8,307 |
Reviews | (127) |
Published | Oct 30, 2023 |
Base Model | |
Training | Steps: 7,272 Epochs: 7 |
Usage Tips | Clip Skip: 2 |
Trigger Words | Crystal Movies Style |
Hash | AutoV2 95E47BEED8 |
Oh we scored a ton of dataset material and our hard drive hates us for it.
This was trained via the CivitAI lora trainer.
It cost us buzz cries
THIS WORKS REALLY WELL and i don't know what we did right because we did wrong entirely on the colab version because we tried to mess with our training settings - but we're really bad at math, so we kinda really fudged up the numbers.
Evidently "EXTREMELY LARGE" doesn't count for under 1000 images, so 5e4 for main and 1e4 for text encoder was just fine - like the tutorial i wrote for semi large datasets? I just budgeted the number of repeats, epochs, batch size and cried in my soup when it cost us more than 500 buzz. (it was a lorge 800+ image dataset.)
We ARE DOING MORE Sailor Moon SEASONS, but trust me when i say this: 30 episodes and each episode has like 500-1000 screencaps - it gonna be awhile. Not only that but in order to retain THAT SEASONS's style you need to remove FILLER episodes. Yea, the ones where the animators went on break lol.
Recipe:
1 cup sass
3 cups WHOOPS.
5 cups WHERES THE. CHEETOS?
Bake at 450 for 24 hours and then call the nearest fire department.
Read a few memes while you're in the ER for severe burns
(KIDDING.)
Contact Us:
Our Discord: https://discord.gg/HhBSvM9gBY
Earth & Dusk Media https://discord.gg/5t2kYxt7An
Backups: https://huggingface.co/EarthnDusk
Send a Pizza: https://ko-fi.com/duskfallcrew/
WE ARE PROUDLY SUPPORTED BY: https://yodayo.com/ / https://moescape.ai/
JOIN OUR DA GROUP: https://www.deviantart.com/diffusionai
JOIN OUR SUBREDDIT: https://www.reddit.com/r/earthndusk/