epiCFluX-Photo LORA
Back

epiCFluX-Photo LORA
4 0 6

Characterization

epiCFluX-Photo LORA

Lets sum it up, style training on Photography with a bigger database via Loras is a pain, chose the version you like the most. I won't spent more time on this :D
In summary, using a larger database via Loras for photography style training is a painful process. Choose the version you like the most. I won't spend more time on this :D

In general training styles work best on 10 20 images max, singlecaption with mixed no caption, or no caption at all seems also very good, but yeah small datasets everything else should be finetuned
In general, training styles work best on a maximum of 10 to 20 images. Single captions mixed with no captions, or no captions at all, also seem very good. But yes,  small datasets everything else should be finetuned.

This model is sourced from an external transfer (transfer address: https://civitai.com/models/651187/epicflux-photo-lora ）,if the original author has objections to this transfer, you can click，
Appeal
We will, within 24 hours, edit, delete, or transfer the model to the original author according to the original author's request

先行芝士

先行芝士

Characterization

Model Information

Frozen
Model Type：
LoRA
Basic Model：
F1基础 D
Trigger Words：
无
Resource Name：
models/loras/35mm_joycap.safetensors
MD5：
ca563c71b837dfd7dbe208b9756301cd

Lets sum it up, style training on Photography with a bigger database via Loras is a pain, chose the version you like the most. I won't spent more time on this :D
In summary, using a larger database via Loras for photography style training is a painful process. Choose the version you like the most. I won't spend more time on this :D

In general training styles work best on 10 20 images max, singlecaption with mixed no caption, or no caption at all seems also very good, but yeah small datasets everything else should be finetuned
In general, training styles work best on a maximum of 10 to 20 images. Single captions mixed with no captions, or no captions at all, also seem very good. But yes,  small datasets everything else should be finetuned.

This model is sourced from an external transfer (transfer address: https://civitai.com/models/651187/epicflux-photo-lora ）,if the original author has objections to this transfer, you can click，
Appeal
We will, within 24 hours, edit, delete, or transfer the model to the original author according to the original author's request