Convert class conditional models(stylegan2-ada pytorch) to genforce format?

[waleedrazakhan92]

I'm trying to convert stylegan2-ada pytorch version into genforce format. The ones that are not conditional are converted without any error but when there is a conditional model, i get this error assert tf_var_name in official_tf_to_pth_var_mapping AssertionError. Which i believe tells that some of the mappings are not found in the model. Can you please tell if the class conditional models are supported or not. Or what else could be the problem. I'm using cifar10 model file for now. And I've also tested with one of the conditional model on faces that i've trained. Both give the same error.

[loopback-kr]

You can use converter

[bridenmj]

Just to follow up. I had the same issue. In stylegan2ada_pth_converter.py, I tracked the issue to:

    G = build_model(gan_type=GAN_TPYE,
                    module='generator',
                    resolution=resolution,
                    z_space_dim=z_space_dim,
                    w_space_dim=w_space_dim,
                    label_size=label_size,
                    repeat_w=repeat_w,
                    image_channels=image_channels)
    G_state_dict = G.state_dict()
  

I had to pass more parameters to build_model to correctly build the StyleGAN2Generator, the code for which is in stylegan2_generator.py. In my case, my mapping network had 2 layers but the default is 8.