search

ZHO-ZHO-ZHO / ComfyUI-InstantID

Unofficial implementation of InstantID for ComfyUI
1.21k stars 67 forks source link

Error after InstantID Generation node #6

Open razvanab opened 5 months ago

razvanab commented 5 months ago

I get this error after the InstantID Generation node:

ERROR:root:!!! Exception during processing !!! ERROR:root:Traceback (most recent call last): File "H:\Stable-diffusion\ComfyUI\execution.py", line 155, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "H:\Stable-diffusion\ComfyUI\execution.py", line 85, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "H:\Stable-diffusion\ComfyUI\custom_nodes\ComfyUI-0246\utils.py", line 363, in new_func res_value = old_func(*final_args, **kwargs) File "H:\Stable-diffusion\ComfyUI\execution.py", line 78, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) File "H:\Stable-diffusion\ComfyUI\nodes.py", line 1415, in save_images full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path(filename_prefix, self.output_dir, images[0].shape[1], images[0].shape[0]) AttributeError: 'str' object has no attribute 'shape'

Semonxue commented 5 months ago

Same as I.

VJDemien commented 5 months ago

Same

pointave commented 5 months ago

It's probably the face you are transferring is too small. Add padding or find another photo

johndpope commented 5 months ago

check the boot up logs when you start comfy. you maybe missing modules. https://github.com/ZHO-ZHO-ZHO/ComfyUI-InstantID/pull/17/files

ludogemma commented 5 months ago

me too same error but after some tries i got it to work .. but ... only with some images.. and i couldn't figure yet what is .. i mean some 512x512 works.. some don't .. i thought the file size no! the file type neither... some .webp works .. but also some png... most of image that works are less than 512px... anyway is almost impossibile right now to predict and understand the images that could works..