search

sjtuplayer / anomalydiffusion

[AAAI 2024] AnomalyDiffusion: Few-Shot Anomaly Image Generation with Diffusion Model
MIT License
114 stars 14 forks source link

Hello, thank you for your work. I would like to ask a code question. #14

Open lsxwtt opened 4 months ago

lsxwtt commented 4 months ago

When I try to run the training code, I get the following error. What's the reason for this?

20240312171920
sjtuplayer commented 4 months ago

Maybe you have finished the training process. Since the code is borrowed from textual inversion, when it comes to the test stage, it means the training is finished.

lsxwtt commented 4 months ago

Maybe you have finished the training process. Since the code is borrowed from textual inversion, when it comes to the test stage, it means the training is finished.

Thank you very much for your reply!

May I ask if trainer.test(model, data) refers to the upcoming release of the exception generation inference code?

If the training is completed, how can I view the log during the training process?

sjtuplayer commented 4 months ago

Maybe you have finished the training process. Since the code is borrowed from textual inversion, when it comes to the test stage, it means the training is finished.

Thank you very much for your reply!

May I ask if trainer.test(model, data) refers to the upcoming release of the exception generation inference code?

If the training is completed, how can I view the log during the training process?

Actually, the upcoming inference code is not for trainer.test. When you encounter the problem of trainer.test, you can then run the inference code. And the training log is saved in ./logs automatically.