Christian Bartz

Hmm, I think I can not help with that then, since I only tested the code with the version of MXNet showed in the README. I suggest that you have...

Oh and regarding your last screenshot: What is the actual error? I can only see warnings and that there is an error, while compiling `cuddn_convolution`

Hmm, seems that your validation files can not be found... If you are using the code to train a FSNS model, I suggest that you use the new code we...

Well... you should try to run `make install` after you've followed all installation steps, provided in the warpctc readme. You can change the install directory by editing the values in...

where did you put libwarpctc.so? Is it in one of the default search locations, like `/usr/lib` or `/usr/local/lib` or is it a path defined by yourself? But even if you...

You should have a look at your groundtruth file. The path to your images is definitely wrong. I think you should not have `/train/0.png` but rather `train/0.png`

You already figured out which `csv` file to use for training. This csv file provides the path to all images that are to be used for training. You should have...

That's a good question. It works because Stylegan uses progressive growing. That means that the output image is generated in several steps and you can grab an image at each...

Thanks for your interest, bu I'm afraid that we will not be able to provide a Colab Notebook, but I would highly appreciate to review a PR that adds a...

Hi, tanks for your kind words! Yes, `--disable-update-for` is exactly the correct way to achieve what you want. If you say `--disable-update-for noise` you will train a model that should...