First of a series of EVO brand cars LoRA models.All training suggestions are welcome.v0.741 first SDXL version, includes EVO 3 car, Cattedrale di Palermo square, Teatro Massimo &amp; Palermo streets

EVO 3

Trained on a dataset of 118 hi-res images (wiith BLIP captions manually edited) with Repeats = 20 and 2360 regularisation images, source model: stabilityai/stable-diffusion-2-1 with v2 and v_parameterization enabled.Relevant Kohya settings:<ul><li>Learning rate: 0.0001</li><li>LR Scheduler: cosine</li><li>LR warmup (% of steps): 10</li><li>Optimizer: AdamW8bit</li><li>Text Encoder learning rate: 0.00005</li><li>Unet learning rate: 0.0001</li><li>Network Rank (Dimension): 128</li><li>Network Alpha: 1</li><li>Max resolution: 768,768</li><li>Clip skip: 1</li><li>Min SNR gamma: 0 </li></ul>

evo-car-lora-sd21-768_v02.yaml

evo-car-lora-sd21-768_v02.safetensors

v0.2 (SD 2.1 768)

v0.85 Milazzo (SDXL)

v0.85 Ragusa (SDXL)

v0.83 Agrigento (SDXL)

v0.741 Palermo (SDXL)

v0.61 Poli+Lecce (SD1.5)

v0.55 Polignano (SD1.5)

v0.45 Lecce (SD2.1 768)

v0.44 (SD2.1 768)

v0.3 (SD2.1 768)

Files with SHA256: 239234f0b052be4732bacd545aaff3e1ad3ed63c63ff3c0e716422ea6d84bd69