#Settings similar to the default ones on the Disco Diffusion Colab Notebook (https://colab.research.google.com/github/alembics/disco-diffusion/blob/main/Disco_Diffusion.ipynb) [model_list] model_list = ['openimages'] [basic_settings] #Perceptor things width = 512 height = 512 clip_guidance_scale = 5000 step = 100 aesthetic_loss_scale = 100 augment_cuts=True #Init image settings starting_timestep = 0.9 init_scale = 1000 mask_scale = 0 [advanced_settings] #Add CLIP Guidance and all the flavors or just run normal Latent Diffusion use_secondary_model=False use_original_as_clip_in=0 lerp=True #Cut settings cut_overview = [12]*400+[4]*600 cut_innercut = [4]*400+[12]*600 cut_ic_pow = 1 cut_icgray_p = [0.2]*100 + [0]*1200 cutn_batches = 4 sat_index = [0]*1040 range_index = [1500000]*100 + [0]*1000 eta_index = [0.8]*100 active_function = "softsign" tv_scales = [150]*4 tv_scale_2 = [] n_batches = 1 unified_cutouts = True ns_cutn = 10 step_enhance=True mid_point = 0.6 steps_pow = 1 #cfg_scale only for cc12m_cfg cfg_scale = 3 #If you uncomment this line you can schedule the CLIP guidance across the steps. Otherwise the clip_guidance_scale will be used clip_guidance_schedule = [2400]*1000 #Apply symmetric loss (force simmetry to your results) symmetric_loss_scale = 0 #Grad and mag advanced settings grad_center = False #Lower value result in more coherent and detailed result, higher value makes it focus on more dominent concept grad_scale=0.1 mag_mul = 10 clamp_start_=0 clamp_index = [0.05]*200 + [0.04]*800 #More settings RGB_min = -0.95 RGB_max = 0.95 #How to pad the image with cut_overview padargs = {'mode': 'constant', 'value': -1} flip_aug=False cc = 6 #Internal upscaler settings activate_upscaler = False upscale_model = "yfcc_2" multiply_image_size_by = 2