search

Justin-Tan / generative-compression

TensorFlow Implementation of Generative Adversarial Networks for Extreme Learned Image Compression
MIT License
511 stars 108 forks source link

Problems with the data set " paths = Data.load_dataframe(directories.train) " #50

Open WenBingo opened 11 months ago

WenBingo commented 11 months ago

In train.py 31st " paths = Data.load_dataframe(directories.train) “。I get an address for the training data, but there is no image at this address, so much so that an error is reported later. I would like to ask if this should be a separate download for the image of the cityscapes data?

WenBingo commented 11 months ago

I down the data of cityscapes and put it into the "./data" , but when i debug to the "train_dataset = Data.load_dataset ( ) " in the model.py , donot have value of the image

WenBingo commented 11 months ago

train_dataset = Data.load_dataset(self.path_placeholder, config.batch_size, augment=False, training_dataset=dataset, use_conditional_GAN=config.use_conditional_GAN, semantic_map_paths=self.semantic_map_path_placeholder)

code of the model.py ,when run this code i get the train_dataset is a tensor of image_path . is right?

zhanpengA commented 10 months ago

I have this issue during the operation. Have you encountered it before. I have placed the cityscapes dataset in the data section, but there will still be this error message.

tensorflow.python.framework.errors_impl.NotFoundError:2 root error(s)found.Not found:data/leftImg8bit/test/munich/munich_000211_000019_leftImg8bit.png;No such file or directory

WenBingo commented 10 months ago

I have not solved this issue