Hi Harry Wang
Thank you for your attention.
I am writing because when I run
lora_add runwayml/stable-diffusion-v1-5 ./models/dreambooth/3d-avatar/final_lora.safetensors ./output_merged.ckpt 0.7 --mode upl-ckpt-v2
error occurs and says
###############################
Lora Add, mode upl-ckpt-v2
You will be using output_merged as the token in A1111 webui. Make sure output_merged is unique enough token.
text_config_dict is provided which will be used to initialize CLIPTextConfig. The value text_config["id2label"] will be overriden.
Traceback (most recent call last):
File "/home/rtcai3/anaconda3/envs/lora/bin/lora_add", line 8, in
sys.exit(main())
File "/home/rtcai3/anaconda3/envs/lora/lib/python3.9/site-packages/lora_diffusion/cli_lora_add.py", line 187, in main
fire.Fire(add)
File "/home/rtcai3/anaconda3/envs/lora/lib/python3.9/site-packages/fire/core.py", line 141, in Fire
component_trace = _Fire(component, args, parsed_flag_args, context, name)
File "/home/rtcai3/anaconda3/envs/lora/lib/python3.9/site-packages/fire/core.py", line 475, in _Fire
component, remaining_args = _CallAndUpdateTrace(
File "/home/rtcai3/anaconda3/envs/lora/lib/python3.9/site-packages/fire/core.py", line 691, in _CallAndUpdateTrace
component = fn(*varargs, **kwargs)
File "/home/rtcai3/anaconda3/envs/lora/lib/python3.9/site-packages/lora_diffusion/cli_lora_add.py", line 142, in add
tok_dict = patch_pipe(loaded_pipeline, path_2, patch_ti=False)
File "/home/rtcai3/anaconda3/envs/lora/lib/python3.9/site-packages/lora_diffusion/lora.py", line 1012, in patch_pipe
monkeypatch_or_replace_safeloras(pipe, safeloras)
File "/home/rtcai3/anaconda3/envs/lora/lib/python3.9/site-packages/lora_diffusion/lora.py", line 800, in monkeypatch_or_replace_safeloras
loras = parse_safeloras(safeloras)
File "/home/rtcai3/anaconda3/envs/lora/lib/python3.9/site-packages/lora_diffusion/lora.py", line 562, in parse_safeloras
info = metadata.get(name)
AttributeError: 'NoneType' object has no attribute 'get'
###############################
I've already prepared "stable-diffusion-v1-5" (downloaded from hugging face), as well as the safetensors (converted by convert-to-safetensors.py; converted from pkl or bin which was trained using your codes).
Could you please look into this issue and I am looking forward to hearing from you. Thanks again.
Best,
Picard314
harrywang
