datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
runpod_Lora_StyleGalMaster-style-lora-dataset
GalMaster Style LoRA Dataset
Overview
329 images (145 CG + 184 character illustrations)
Source: GalMaster (Galgame)
Purpose: Style LoRA training for Stable Diffusion
Structure
All files are flat in the repo root (no subdirectories):
329 image files (.png / .jpg)
329 caption files (.txt, same name as image)
tag_summary.csv — caption summary for all 329 images
Captioning
Tagger: WD ConvNeXt Tagger v3… See the full description on the dataset page: https://huggingface.co/datasets/huanx/GalMaster-style-lora-dataset.kn1febrush-flux2-style-lora-datasetautotrain-sdxl-vintage-face-style-loramarvel_zombies_style_lora_flux2_nf4tron_ares_style_lora_fluxmcl-style-loramarvel_zombies_style_lora_fluxtron_ares_style_lora_flux2_nf4datesetimg_Blue_Archive_style_lora01_Background
Dataset
This dataset contains files.
