mukosame / aoda Goto Github PK

If the value of lambda_E is 0, does net R still work?
What is the appropriate value of lambda_E?

Hello, thank you for your excellent work !
When will you give the code ?

Please download the weights from [GoogleDrive], and put it into the weights/ folder.

Could you please attach a link to the GoogleDrive, thanks!
or would you be able to provide pretrained model ?

Thanks for the awesome work!

After we downloaded and extracted the data into ./datasets/Object/
There are 3 different folders (i.e. Edge , GT and Sketch)

My questions:

1. What are the steps required in order to train a cat images (i.e. class no 17) so that we can use it for "Photo-to-Sketch Synthesis"? It appears that it is required to create trainA , trainB folder? how could we restructure it from the original dataset?
2. What are the parameters to use for train.py ?

Hello! Would you be interested in hosting your model weights on the Hugging Face Hub? Hosting is free, and it would make your model visible to our users, who I know would love to work with this model 🤗

We have guides on uploading models, but I'm also more than happy to help you out if you'd need assistance! We've also found that models get a lot of traction when they have a demo, which we can host for free on Hugging Face Spaces. (Docs available here)

Just like with uploading the models, I'd love to help you out with getting a demo set up! Let me know if these are things you'd find interesting 😄

cc: @osanseviero

Thank you for the work.
I would like to train a model like Anime2Sketch.
Will Unet generator version be available?

2022-10-24 15:03:50.918 File "/usr/local/python/lib/python3.7/site-packages/torch/nn/functional.py", line 1847, in linear

2022-10-24 15:03:50.918 return torch._C._nn.linear(input, weight, bias)

2022-10-24 15:03:50.918 RuntimeError: CUDA error: CUBLAS_STATUS_NOT_INITIALIZED when calling cublasCreate(handle)