Egg Fusion - LoRa Merge

512dim lora와 chilloutmix_NiPrunedFp32Fix + deliberate_v2 기반의 완전히 새롭게 개선된 체크포인트입니다.

학습 데이터: 512 DIM LORA
/model/41893/lora-eggeaster-fusion

다음은 모델 위에 lora를 적용했을 때의 결과입니다:

lora 가중치: TEnc Weight 0.2 / UNet Weight 1

512dim lora를 chillout과 deliberate에 병합하고, --ratios 0.99를 사용하여 내 융합 lora에 가중치를 적용함

C는 CHillout / d는 deliberate

C와 D를 다음과 같이 병합:

Base_alpha=0.53
Weight_values=0,0.157576195987654,0.28491512345679,0.384765625,0.459876543209877,0.512996720679012,0.546875,0.564260223765432,0.567901234567901,0.560546875,0.544945987654321,0.523847415123457,0.5,0.476152584876543,0.455054012345679,0.439453125,0.432098765432099,0.435739776234568,0.453125,0.487003279320987,0.540123456790124,0.615234375,0.71508487654321,0.842423804012347,1

그리고
1,0.842423804012346,0.71508487654321,0.615234375,0.540123456790123,0.487003279320988,0.453125,0.435739776234568,0.432098765432099,0.439453125,0.455054012345679,0.476152584876543,0.5,0.523847415123457,0.544945987654321,0.560546875,0.567901234567901,0.564260223765432,0.546875,0.512996720679013,0.459876543209876,0.384765625,0.28491512345679,0.157576195987653,0

두 결과를 합성함

Weight_values=
1,0.842423804012346,0.71508487654321,0.615234375,0.540123456790123,0.487003279320988,0.453125,0.435739776234568,0.432098765432099,0.439453125,0.455054012345679,0.476152584876543,0.5,0.523847415123457,0.544945987654321,0.560546875,0.567901234567901,0.564260223765432,0.546875,0.512996720679013,0.459876543209876,0.384765625,0.28491512345679,0.157576195987653,0

그리고 마침내..
512dim으로 학습된 lora를 적용한 Egg_fusion 체크포인트입니다.

달걀을 즐기세요 🤟 🥃