flux-2-klein-9B-schematic-dataset

# FLUX.2 Klein Schematic LoRA Dataset This is the training dataset used for the FLUX.2 Klein Schematic LoRA project. The LoRA repository is available here: - LoRA: https://huggingface.co/nomadoor/flux-2-klein-9B-schematic-lora For details about the experiment and dataset construction, see the blog post: - Blog: https://comfyui.nomadoor.net/en/notes/flux2-klein-schematic-lora/ ## Structure ```text data/ depth_relative/ input/ target/ caption/ surface_normal/ input/ target/ caption/ pose_body/ input/ target/ caption/ pose_full/ input/ target/ caption/ segmentation_binary/ input/ target/ caption/ segmentation_amodal/ input/ target/ caption/ metadata/ manifest.jsonl summary.json ``` Each row is an image-editing pair: - `input/`: source image - `target/`: RGB schematic target image - `caption/`: instruction prompt used for training ## Counts | Task | Positive | Negative | Total | |---|---:|---:|---:| | `depth_relative` | 300 | 0 | 300 | | `surface_normal` | 300 | 0 | 300 | | `pose_body` | 300 | 30 | 330 | | `pose_full` | 300 | 30 | 330 | | `segmentation_binary` | 300 | 30 | 330 | | `segmentation_amodal` | 300 | 30 | 330 | Total: `1920` image pairs. ## Notes - This dataset was prepared for LoRA training on `black-forest-labs/FLUX.2-klein-base-9B`. - Some pose and segmentation tasks include all-black negative targets. - The dataset is small and experimental.