237 lines
10 KiB
Python
237 lines
10 KiB
Python
import comfy
|
|
import torch
|
|
import numpy as np
|
|
import latent_preview
|
|
|
|
|
|
def prepare_noise(latent_image, seed, noise_inds=None, noise_device="cpu", incremental_seed_mode="comfy", variation_seed=None, variation_strength=None):
|
|
"""
|
|
creates random noise given a latent image and a seed.
|
|
optional arg skip can be used to skip and discard x number of noise generations for a given seed
|
|
"""
|
|
|
|
latent_size = latent_image.size()
|
|
latent_size_1batch = [1, latent_size[1], latent_size[2], latent_size[3]]
|
|
|
|
if variation_strength > 0 or incremental_seed_mode.startswith("variation str inc"):
|
|
if noise_device == "cpu":
|
|
variation_generator = torch.manual_seed(variation_seed)
|
|
else:
|
|
torch.cuda.manual_seed(variation_seed)
|
|
variation_generator = None
|
|
|
|
variation_latent = torch.randn(latent_size_1batch, dtype=latent_image.dtype, layout=latent_image.layout,
|
|
generator=variation_generator, device=noise_device)
|
|
else:
|
|
variation_latent = None
|
|
|
|
def apply_variation(input_latent, strength_up=None):
|
|
if variation_latent is None:
|
|
return input_latent
|
|
else:
|
|
strength = variation_strength
|
|
|
|
if strength_up is not None:
|
|
strength += strength_up
|
|
|
|
variation_noise = variation_latent.expand(input_latent.size()[0], -1, -1, -1)
|
|
result = (1 - strength) * input_latent + strength * variation_noise
|
|
return result
|
|
|
|
# method: incremental seed batch noise
|
|
if noise_inds is None and incremental_seed_mode == "incremental":
|
|
batch_cnt = latent_size[0]
|
|
|
|
latents = None
|
|
for i in range(batch_cnt):
|
|
if noise_device == "cpu":
|
|
generator = torch.manual_seed(seed+i)
|
|
else:
|
|
torch.cuda.manual_seed(seed+i)
|
|
generator = None
|
|
|
|
latent = torch.randn(latent_size_1batch, dtype=latent_image.dtype, layout=latent_image.layout,
|
|
generator=generator, device=noise_device)
|
|
|
|
latent = apply_variation(latent)
|
|
|
|
if latents is None:
|
|
latents = latent
|
|
else:
|
|
latents = torch.cat((latents, latent), dim=0)
|
|
|
|
return latents
|
|
|
|
# method: incremental variation batch noise
|
|
elif noise_inds is None and incremental_seed_mode.startswith("variation str inc"):
|
|
batch_cnt = latent_size[0]
|
|
|
|
latents = None
|
|
for i in range(batch_cnt):
|
|
if noise_device == "cpu":
|
|
generator = torch.manual_seed(seed)
|
|
else:
|
|
torch.cuda.manual_seed(seed)
|
|
generator = None
|
|
|
|
latent = torch.randn(latent_size_1batch, dtype=latent_image.dtype, layout=latent_image.layout,
|
|
generator=generator, device=noise_device)
|
|
|
|
step = float(incremental_seed_mode[16:])
|
|
latent = apply_variation(latent, step*i)
|
|
|
|
if latents is None:
|
|
latents = latent
|
|
else:
|
|
latents = torch.cat((latents, latent), dim=0)
|
|
|
|
return latents
|
|
|
|
# method: comfy batch noise
|
|
if noise_device == "cpu":
|
|
generator = torch.manual_seed(seed)
|
|
else:
|
|
torch.cuda.manual_seed(seed)
|
|
generator = None
|
|
|
|
if noise_inds is None:
|
|
latents = torch.randn(latent_image.size(), dtype=latent_image.dtype, layout=latent_image.layout,
|
|
generator=generator, device=noise_device)
|
|
latents = apply_variation(latents)
|
|
return latents
|
|
|
|
unique_inds, inverse = np.unique(noise_inds, return_inverse=True)
|
|
noises = []
|
|
for i in range(unique_inds[-1] + 1):
|
|
noise = torch.randn([1] + list(latent_image.size())[1:], dtype=latent_image.dtype, layout=latent_image.layout,
|
|
generator=generator, device=noise_device)
|
|
if i in unique_inds:
|
|
noises.append(noise)
|
|
noises = [noises[i] for i in inverse]
|
|
noises = torch.cat(noises, axis=0)
|
|
return noises
|
|
|
|
|
|
def common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent, denoise=1.0,
|
|
noise_mode="CPU", disable_noise=False, start_step=None, last_step=None, force_full_denoise=False,
|
|
incremental_seed_mode="comfy", variation_seed=None, variation_strength=None):
|
|
device = comfy.model_management.get_torch_device()
|
|
noise_device = "cpu" if noise_mode == "CPU" else device
|
|
latent_image = latent["samples"]
|
|
|
|
if disable_noise:
|
|
noise = torch.zeros(latent_image.size(), dtype=latent_image.dtype, layout=latent_image.layout, device=noise_device)
|
|
else:
|
|
batch_inds = latent["batch_index"] if "batch_index" in latent else None
|
|
noise = prepare_noise(latent_image, seed, batch_inds, noise_device, incremental_seed_mode, variation_seed=variation_seed, variation_strength=variation_strength)
|
|
|
|
noise_mask = None
|
|
if "noise_mask" in latent:
|
|
noise_mask = latent["noise_mask"]
|
|
|
|
preview_format = "JPEG"
|
|
if preview_format not in ["JPEG", "PNG"]:
|
|
preview_format = "JPEG"
|
|
|
|
previewer = latent_preview.get_previewer(device, model.model.latent_format)
|
|
|
|
pbar = comfy.utils.ProgressBar(steps)
|
|
def callback(step, x0, x, total_steps):
|
|
preview_bytes = None
|
|
if previewer:
|
|
preview_bytes = previewer.decode_latent_to_preview_image(preview_format, x0)
|
|
pbar.update_absolute(step + 1, total_steps, preview_bytes)
|
|
|
|
samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image,
|
|
denoise=denoise, disable_noise=disable_noise, start_step=start_step, last_step=last_step,
|
|
force_full_denoise=force_full_denoise, noise_mask=noise_mask, callback=callback, seed=seed)
|
|
out = latent.copy()
|
|
out["samples"] = samples
|
|
return (out, )
|
|
|
|
|
|
class KSampler_inspire:
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {"required":
|
|
{"model": ("MODEL",),
|
|
"seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
|
|
"steps": ("INT", {"default": 20, "min": 1, "max": 10000}),
|
|
"cfg": ("FLOAT", {"default": 8.0, "min": 0.0, "max": 100.0}),
|
|
"sampler_name": (comfy.samplers.KSampler.SAMPLERS, ),
|
|
"scheduler": (comfy.samplers.KSampler.SCHEDULERS, ),
|
|
"positive": ("CONDITIONING", ),
|
|
"negative": ("CONDITIONING", ),
|
|
"latent_image": ("LATENT", ),
|
|
"denoise": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.01}),
|
|
"noise_mode": (["GPU(=A1111)", "CPU"],),
|
|
"batch_seed_mode": (["incremental", "comfy", "variation str inc:0.01", "variation str inc:0.05"],),
|
|
"variation_seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
|
|
"variation_strength": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.01}),
|
|
}
|
|
}
|
|
|
|
RETURN_TYPES = ("LATENT",)
|
|
FUNCTION = "sample"
|
|
|
|
CATEGORY = "InspirePack/a1111_compat"
|
|
|
|
def sample(self, model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise, noise_mode, batch_seed_mode="comfy", variation_seed=None, variation_strength=None):
|
|
return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise, noise_mode, incremental_seed_mode=batch_seed_mode, variation_seed=variation_seed, variation_strength=variation_strength)
|
|
|
|
|
|
class KSamplerAdvanced_inspire:
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {"required":
|
|
{"model": ("MODEL",),
|
|
"add_noise": ("BOOLEAN", {"default": True, "label_on": "enable", "label_off": "disable"}),
|
|
"noise_seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
|
|
"steps": ("INT", {"default": 20, "min": 1, "max": 10000}),
|
|
"cfg": ("FLOAT", {"default": 8.0, "min": 0.0, "max": 100.0, "step":0.5, "round": 0.01}),
|
|
"sampler_name": (comfy.samplers.KSampler.SAMPLERS, ),
|
|
"scheduler": (comfy.samplers.KSampler.SCHEDULERS, ),
|
|
"positive": ("CONDITIONING", ),
|
|
"negative": ("CONDITIONING", ),
|
|
"latent_image": ("LATENT", ),
|
|
"start_at_step": ("INT", {"default": 0, "min": 0, "max": 10000}),
|
|
"end_at_step": ("INT", {"default": 10000, "min": 0, "max": 10000}),
|
|
"noise_mode": (["GPU(=A1111)", "CPU"],),
|
|
"return_with_leftover_noise": ("BOOLEAN", {"default": False, "label_on": "enable", "label_off": "disable"}),
|
|
"batch_seed_mode": (["incremental", "comfy", "variation str inc:0.01", "variation str inc:0.05"],),
|
|
"variation_seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
|
|
"variation_strength": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.01}),
|
|
}
|
|
}
|
|
|
|
RETURN_TYPES = ("LATENT",)
|
|
FUNCTION = "sample"
|
|
|
|
CATEGORY = "InspirePack/a1111_compat"
|
|
|
|
def sample(self, model, add_noise, noise_seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, start_at_step, end_at_step, noise_mode, return_with_leftover_noise, denoise=1.0, batch_seed_mode="comfy", variation_seed=None, variation_strength=None):
|
|
force_full_denoise = True
|
|
|
|
if return_with_leftover_noise:
|
|
force_full_denoise = False
|
|
|
|
disable_noise = False
|
|
|
|
if not add_noise:
|
|
disable_noise = True
|
|
|
|
return common_ksampler(model, noise_seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image,
|
|
denoise=denoise, disable_noise=disable_noise, start_step=start_at_step, last_step=end_at_step,
|
|
force_full_denoise=force_full_denoise, noise_mode=noise_mode, incremental_seed_mode=batch_seed_mode,
|
|
variation_seed=variation_seed, variation_strength=variation_strength)
|
|
|
|
|
|
NODE_CLASS_MAPPINGS = {
|
|
"KSampler //Inspire": KSampler_inspire,
|
|
"KSamplerAdvanced //Inspire": KSamplerAdvanced_inspire,
|
|
}
|
|
NODE_DISPLAY_NAME_MAPPINGS = {
|
|
"KSampler //Inspire": "KSampler (inspire)",
|
|
"KSamplerAdvanced //Inspire": "KSamplerAdvanced (inspire)"
|
|
}
|