doantientai
doantientai
I tried 128, and it gave me the same error.
Thank you @damlasena, I switched to numpy method and it works. I also think that the line 306 in inception_utils.py should be: ` FID = numpy_calculate_frechet_distance(mu, sigma, data_mu, data_sigma)` instead...
@kanxx030 No you don't have to, because the program will resize the images for you. Use the argument `img_size` to specify your expected size. More info: https://github.com/taki0112/BigGAN-Tensorflow/blob/d8c4367a7c0a7e90ada258e343d23ba6a3ead342/main.py#L50
A simple way is to write a short program to reshape / crops your images to the requested size, then move them to the input folder. That's what I do.
The differences between 2 papers seem not enough to split them into 2 papers. I think they were submitted at quite the same time so they both got published. However,...
Wonderful work. Good luck!
Hi @gitter-me , as a newbie, I have a silly question. In your solution, it is likely that I need to define onChange(value) somewhere before using it. What do I...
Thank you @MakeDirtyCode. I also found out that. I fix it by applying a max_pooling layer.
Hi @phamnam95 I'm not the author but glad if we can support each other. About the input size format, I'm pretty sure it is batch_size, num_channels, height, width (I'm not...
@phamnam95 not yet. I guess the author assumed that the image size is always 64x64x3 already. It would be tough to make the code to be generic without the author's...