xinyun/ComfyUI
1
0
Fork 0
mirror of https://git.datalinker.icu/comfyanonymous/ComfyUI synced 2026-11-13 05:27:00 +08:00
Code Issues Packages Projects Releases Wiki Activity

fixed the rename bug.

Browse Source
This commit is contained in:
zhangsongbo 2024-09-16 14:50:44 +08:00
parent 559a626ce3
commit c782dcc6ca
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
comfy_extras/nodes_photomaker_v2.py
View File

@ -209,9 +209,9 @@ class PhotoMakerEncodeV2:
CATEGORY = "_for_testing/photomaker" CATEGORY = "_for_testing/photomaker"
def apply_photomaker(self, photomaker, image, clip, text): def apply_photomaker(self, photomakerv2, image, clip, text):
special_token = "photomaker" special_token = "photomaker"
pixel_values = comfy.clip_vision.clip_preprocess(image.to(photomaker.load_device)).float() pixel_values = comfy.clip_vision.clip_preprocess(image.to(photomakerv2.load_device)).float()
try: try:
if special_token in text: if special_token in text:
text = text.replace(special_token, " " + special_token + " ") text = text.replace(special_token, " " + special_token + " ")
@ -238,11 +238,11 @@ class PhotoMakerEncodeV2:
token_index = index - 1 token_index = index - 1
num_id_images = image.shape[0] num_id_images = image.shape[0]
num_tokens_v2 = 2 num_tokens_v2 = 2
id_embeds = id_embeds.unsqueeze(0).to(device=photomaker.load_device) id_embeds = id_embeds.unsqueeze(0).to(device=photomakerv2.load_device)
class_tokens_mask = [True if token_index <= i < token_index+(num_id_images*num_tokens_v2) else False for i in range(cond.shape[1])] class_tokens_mask = [True if token_index <= i < token_index+(num_id_images*num_tokens_v2) else False for i in range(cond.shape[1])]
out = photomaker(id_pixel_values=pixel_values.unsqueeze(0), prompt_embeds=cond.to(photomaker.load_device), out = photomakerv2(id_pixel_values=pixel_values.unsqueeze(0), prompt_embeds=cond.to(photomakerv2.load_device),
class_tokens_mask=torch.tensor(class_tokens_mask, dtype=torch.bool, device=photomaker.load_device).unsqueeze(0),id_embeds=id_embeds.to(photomaker.load_device)) class_tokens_mask=torch.tensor(class_tokens_mask, dtype=torch.bool, device=photomakerv2.load_device).unsqueeze(0),id_embeds=id_embeds.to(photomakerv2.load_device))
else: else:
out = cond out = cond