Update Storydiffusion_node.py
fix macOS load config
This commit is contained in:
@@ -921,7 +921,9 @@ def msdiffusion_main(pipe, image_1, image_2, prompts_dual, width, height, steps,
|
||||
return res
|
||||
if _model_type == "img2img" : # 图生图双角色目前只能先用原方法,2者encoder模型不同,没法相互调用
|
||||
del pipe
|
||||
|
||||
if sys.platform=="darwin":
|
||||
cur_path=os.getcwd()
|
||||
original_config_file=os.path.join(cur_path,'config', 'sd_xl_base.yaml')
|
||||
# load SDXL pipeline
|
||||
if dif_repo!="none":
|
||||
single_files=False
|
||||
|
||||
Reference in New Issue
Block a user