TypeError: cannot unpack non-iterable NoneType object

bcmi / F2GAN-Few-Shot-Image-Generation

Fusing-and-Filling GAN (F2GAN) for few-shot image generation, ACM MM2020

78 stars 11 forks source link

TypeError: cannot unpack non-iterable NoneType object #2

Closed jinzhenmu closed 4 years ago

jinzhenmu commented 4 years ago

when i use " python3 train_dagan_with_matchingclassifier.py --dataset emnist --image_width 28 --batch_size 20 --experiment_title MMF2GAN/emnist1way3shot --selected_classes 1 --support_number 3 --loss_G 1 --loss_D 1 --loss_CLA 1 --loss_recons_B 1 --loss_matching_G 0.01 --loss_matching_D 1 --loss_sim 1 "

i got the wrong message at "dagan_networks_wgan_with_matchingclassifier.py", line 147, in generate -> z_dim = self.z_dim) TypeError: cannot unpack non-iterable NoneType object

i can not to find what happen

hy-zpg commented 4 years ago

you can modify /utils/parser_utils.py by setting the default of '--matching ' as 1, or adding --matching 1 in the scripts.

jinzhenmu commented 4 years ago

When I modify /utils/parser_utils.py by setting the default of '--matching ' as 1
the "TypeError: cannot unpack non-iterable NoneType object" is missing.
but i got a new error at "dagan_architectures_with_matchingclassifier.py", line 711,
in call z_inputs_2_embedding, KL_loss_2 = self.VAE_kl_loss(conditional_input,
z_dim)
TypeError: VAE_kl_loss() missing 3 required positional arguments: 'training',
'dropout', and 'scope'
And what "dropout_rate " should i add?

hy-zpg commented 4 years ago

I have updated utils/parser_utils.py and dagan_architectures_with_matchingclassifier.py, you can download this two files and try again.