IS there a way to turn off discriminator training?

[Suvi-dha]

[Suvi-dha]

I am getting this error when training to train just shape maps using your pretrained network. And I don't have labels in my training data.

`Traceback (most recent call last): File "/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/framework/ops.py", line 638, in set_shape c_api.TF_GraphSetTensorShape_wrapper( tensorflow.python.framework.errors_impl.InvalidArgumentError: Dimension 1 in both shapes must be equal, but are 0 and 7. Shapes are [?,0] and [?,7].

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "train.py", line 290, in tfutil.call_func_by_name(**config.train) File "/root/TBGAN/tfutil.py", line 234, in call_func_by_name return import_obj(func)(*args, **kwargs) File "/root/TBGAN/train.py", line 189, in train_progressive_gan G_loss = tfutil.call_func_by_name(G=G_gpu, D=D_gpu, opt=G_opt, training_set=training_set, minibatch_size=minibatch_split, **config.G_loss) File "/root/TBGAN/tfutil.py", line 234, in call_func_by_name return import_obj(func)(*args, **kwargs) File "/root/TBGAN/loss.py", line 30, in G_wgan_acgan fake_images_out = G.get_output_for(latents, labels, is_training=True) File "/root/TBGAN/tfutil.py", line 507, in get_output_for out_expr = self._build_func(*named_inputs, **all_kwargs) File "", line 175, in G_paper File "/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/framework/ops.py", line 645, in set_shape raise ValueError(str(e)) ValueError: Dimension 1 in both shapes must be equal, but are 0 and 7. Shapes are [?,0] and [?,7].`

[rohit7044]

Hey @Suvi-dha ! Can you tell me how you manged to get the training dataset ready?

[Suvi-dha]

Hi @rohit7044 I have few 3D models with their textures. I converted those 3D models to shape maps and normals.

[rohit7044]

But the input says modalities. What does modalities mean ? I am so confused. Given multi-view images as input I am trying to understand if TB-GAN can generate normal and UV map. Can you share your opinion?