yingchen001 / wavefill Goto Github PK

Could you please provided the train.flist of the CelebA dataset, Thank you very much

I have test celeba pretrained model, and it looks nice. However I got poor test results for place2 pretrained models. Is something wrong?

When testing the celeba-HQ pre-training model, I don’t know how to set the parameters in test_wavelet.sh, Thank you very much！

The result of my current operation is like this, I don't know why. Thank you for your review！

Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
RuntimeError: Function 'L1LossBackward0' returned nan values in its 0th output.

Hello, I am very interested in your work, but when I test, an error occurred, and he I don't know how to solve.

In models.networks.generator, there should be a class whose name matches spadegenerator in lowercase without underscore(_)

Process finished with exit code 0

Thankyou!