Closed rafstahelin closed 4 months ago
It can't handle non-numeric weight syntax, which loractl supports. It's trying to interpret the complex weight syntax as a float and chokes because it's unable to do so. It's likely just not functioning.
Just disabled composable lora. working nowraf stahelinOn 12 Apr 2024, at 16:18, Chris Heald @.***> wrote: It can't handle non-numeric weight syntax, which loractl supports. It's trying to interpret the complex weight syntax as a float and chokes because it's unable to do so. It's likely just not functioning.
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: @.***>
Just wondering if composable lora clashing with lora CTL?
I am getting this error in console when inferring:
Traceback (most recent call last): File "E:\auto1111\modules\scripts.py", line 784, in process script.process(p, *script_args) File "E:\auto1111\extensions\stable-diffusion-webui-composable-lora\scripts\composable_lora_script.py", line 133, in process composable_lora.load_prompt_loras(prompt) File "E:\auto1111\extensions\stable-diffusion-webui-composable-lora\composable_lora.py", line 111, in load_prompt_loras multiplier = float(params.items[1]) if len(params.items) > 1 else 1.0 ValueError: could not convert string to float: '0@0,1@0.4'
The CTL inference syntax does still give me an image but wondered if it was necessary to remove the composable lora extension?