Closed mgcyung closed 1 year ago
Hello mgcyung,
Thanks for getting in touch. Stable Diffusion access through HuggingFace requires you to be logged in to the HuggingFace CLI in the terminal where you ran the python script. You might find this quickstart guide helpful:
https://huggingface.co/docs/huggingface_hub/quick-start
In addition, based on this line of the textual inversion script, class names are converted into a special token like <Chrysomelida_hindwing>
.
Let me know if these fix the issue!
Best, Brandon
Hello brandontrabucco,
Thank you for reaching out to me. I am glad to inform you that I have
solved the problem by replacing the use_auth_token=True
with
use_auth_token=False
in the from_pretrained function. I appreciate
your suggestion and the quickstart guide you provided. It was very
helpful in understanding the Stable Diffusion access through
HuggingFace.
I also took note of the line you mentioned in the textual inversion
script, where class names Chrysomelida hindwing
are converted into a
special token <Chrysomelida_hindwing>
but not <Chrysomelida hindwing>
.
Thank you once again for your assistance.
Best mgcyung
Hello brandontrabucco,
Thank you for reaching out to me. I am glad to inform you that I have solved the problem by replacing the
use_auth_token=True
withuse_auth_token=False
in the from_pretrained function. I appreciate your suggestion and the quickstart guide you provided. It was very helpful in understanding the Stable Diffusion access through HuggingFace.I also took note of the line you mentioned in the textual inversion script, where class names
Chrysomelida hindwing
are converted into a special token<Chrysomelida_hindwing>
but not<Chrysomelida hindwing>
.Thank you once again for your assistance.
Best mgcyung
Hello, can you tell us how to solve it?
A more detailed answer to this is in issue https://github.com/brandontrabucco/da-fusion/issues/9
Hi, I have ran the fine_tune.py with a custom dataset with only one class called "Chrysomelida hindwing" and the learned_embeddings are generated. However, when I try to run generate_images.py, an error said "OSError: Token is required (
token=True
), but no token found. "The command is as follows:
And the full backtrace is as follows: