You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi
Thank you for the amazing work you have done, hopefully I can use it for my thesis. I am looking forward to try it out myself, for that, I have so far 4 questions:
1 - how long do I need to get each embedding for one concept?
2- I will be using faces photos to training on, do you encourage me to lower lr to 1e-6 and use higher max_train_steps, i.e. 5000. other recommended settings?
3- instant_prompt shall only include the identifier as per the example or should I use the usual 'a photo of rzk person'?
4 - the json example file you provided 'guidance_config_example.json' file is not working. can you check it out please? I am receive the following error message.
JSONDecodeError: Expecting value: line 7 column 1 (char 6)
when running
import json
# Opening JSON file
with open('/mnt/c/Users/noobw/PycharmProjects/pythonProject/thesis/cones_v2/embeddings/guidance_config_example.json', 'r') as openfile:
# json_object = json.loads(openfile.read())
# Reading from json file
json_object = json.load(openfile)
print(json_object)
print(type(json_object))
Hi
Thank you for the amazing work you have done, hopefully I can use it for my thesis. I am looking forward to try it out myself, for that, I have so far 4 questions:
1 - how long do I need to get each embedding for one concept?
2- I will be using faces photos to training on, do you encourage me to lower lr to 1e-6 and use higher max_train_steps, i.e. 5000. other recommended settings?
3- instant_prompt shall only include the identifier as per the example or should I use the usual 'a photo of rzk person'?
4 - the json example file you provided 'guidance_config_example.json' file is not working. can you check it out please? I am receive the following error message.
when running
@Johanan528
The text was updated successfully, but these errors were encountered: