nodes/kijai-ComfyUI-WanVideoWrapper
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

vid2vid, cfg scheduling

Browse Source
This commit is contained in:
kijai
2025-02-26 01:13:13 +02:00
parent 44c5944f70
commit b81ea1d0f7
2 changed files with 909 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files
nodes.py
+28 -16
View File
@@ -697,13 +697,13 @@ class WanVideoSampler:
},
# "optional": {
# #"samples": ("LATENT", {"tooltip": "init Latents to use for video2video process"} ),
# #"image_cond_latents": ("LATENT", {"tooltip": "init Latents to use for image2video process"} ),
# #"denoise_strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.01}),
"optional": {
"samples": ("LATENT", {"tooltip": "init Latents to use for video2video process"} ),
#"image_cond_latents": ("LATENT", {"tooltip": "init Latents to use for image2video process"} ),
"denoise_strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.01}),
# #"riflex_freq_index": ("INT", {"default": 0, "min": 0, "max": 1000, "step": 1, "tooltip": "Frequency index for RIFLEX, disabled when 0, default 4. Allows for new frames to be generated after 129 without looping"}),
# }
#"riflex_freq_index": ("INT", {"default": 0, "min": 0, "max": 1000, "step": 1, "tooltip": "Frequency index for RIFLEX, disabled when 0, default 4. Allows for new frames to be generated after 129 without looping"}),
}
}
RETURN_TYPES = ("LATENT",)
@@ -711,7 +711,7 @@ class WanVideoSampler:
FUNCTION = "process"
CATEGORY = "WanVideoWrapper"
def process(self, model, text_embeds, image_embeds, shift, steps, cfg, seed, scheduler, riflex_freq_index, force_offload=True):
def process(self, model, text_embeds, image_embeds, shift, steps, cfg, seed, scheduler, riflex_freq_index, force_offload=True, samples=None, denoise_strength=1.0):
patcher = model
model = model.model
transformer = model.diffusion_model
@@ -762,6 +762,8 @@ class WanVideoSampler:
#for name, param in transformer.named_parameters():
# print(name, param.data.device)
steps = int(steps/denoise_strength)
if scheduler == 'unipc':
sample_scheduler = FlowUniPCMultistepScheduler(
@@ -789,6 +791,9 @@ class WanVideoSampler:
else:
raise NotImplementedError("Unsupported solver.")
if denoise_strength < 1.0:
steps = int(steps * denoise_strength)
timesteps = timesteps[-(steps + 1):]
seed_g = torch.Generator(device=torch.device("cpu"))
seed_g.manual_seed(seed)
@@ -814,6 +819,10 @@ class WanVideoSampler:
device=torch.device("cpu"),
generator=seed_g)
if samples is not None:
latent_timestep = timesteps[:1].to(noise)
noise = noise * latent_timestep / 1000 + (1 - latent_timestep / 1000) * samples["samples"].squeeze(0).to(noise)
latent = noise.to(device)
d = transformer.dim // transformer.num_heads
@@ -824,6 +833,10 @@ class WanVideoSampler:
],
dim=1)
if not isinstance(cfg, list):
cfg = [cfg] * (steps +1)
print(cfg)
base_args = {
'clip_fea': image_embeds.get('clip_context', None),
'seq_len': seq_len,
@@ -855,12 +868,14 @@ class WanVideoSampler:
noise_pred_cond = transformer(
latent_model_input, t=timestep, **arg_c)[0].to(offload_device)
noise_pred_uncond = transformer(
latent_model_input, t=timestep, **arg_null)[0].to(offload_device)
noise_pred = noise_pred_uncond + cfg * (
noise_pred_cond - noise_pred_uncond)
if cfg[i] >= 1.0:
noise_pred_uncond = transformer(
latent_model_input, t=timestep, **arg_null)[0].to(offload_device)
noise_pred = noise_pred_uncond + cfg[i] * (
noise_pred_cond - noise_pred_uncond)
else:
noise_pred = noise_pred_cond
latent = latent.to(offload_device)
@@ -875,9 +890,6 @@ class WanVideoSampler:
x0 = [latent.to(device)]
if callback is not None:
print(t)
print("latent_model_input", latent_model_input[0].shape)
print("noise_pred", noise_pred.shape)
callback_latent = (latent_model_input[0].cpu() - noise_pred * t.cpu() / 1000).detach().permute(1,0,2,3)
callback(i, callback_latent, None, steps)
else: