Description
This LoRA trained using kohya_ss on 60 images 768×768 for 2 epochs 1560 steps each with manually enhanced blip2 and/or llava captions and without regularization images.
Source model: ‘v1-5-pruned-emaonly’ checkpoint (SD1.5), so this LoRA should works well with most of Stable Diffusion v1.5 forks and mixes.
Learning rate: 2e-05 or 0.00001.
Precision: fp16.
LR Scheduler: constant_with_warmup.
LR warmup: 30%.
Preview images were rendered on checkpoints: pirsusEpicRealism_v31Pruned (hash: 582daed225) and realisticVisionV30_v30VAE (hash: c52892e92a)
If you are going to remake preview images you should have:
checkpoints: pirsusEpicRealism_v31Pruned and realisticVisionV30_v30VAE
lora: AIDA_LoRA_EmSam
embeddings: kkw-ph1
Download link to AIDA_LoRA_EmSam.safetensors will be available for you after purchase. For your comfort I packed AIDA_LoRA_EmSam.safetensors into .zip archive with AIDA_LoRA_EmSam.png and AIDA_LoRA_EmSam.txt , preview and description are compatible with AUTOMATIC1111 extra networks menu.
Also you will get .zip archive with all preview images and generation data for them.
Purchasing this asset you will achieve rights for selling images generated with AIDA_LoRA_EmSam.safetensors LoRA. If you want say special thanks to me, you can add additional amount to your payment.
Please contact me if you need some extended rights for sharing or selling AIDA_LoRA_EmSam, merging AIDA_LoRA_EmSam with other models, using AIDA_LoRA_EmSam in generation services and so on.
Reviews
There are no reviews yet.