Creative-Adversarial-Networks
Creative-Adversarial-Networks copied to clipboard
(WIP) Implementation of Creative Adversarial Networks https://arxiv.org/pdf/1706.07068.pdf
Hi all, I wonder if anyone encountered gradient explosion after training for a few epochs? I am trying to use the CAN losses on a sequence generation model. But, the...
Which version of python and other packages I should use?
Resolving issue #29, corrected README with updated links for WikiArt dataset.
I have been having difficulties training the model from scratch. After running 'bash experiments/train_can_paper.sh' from the CAN folder, I get this error which I am not sure how to fix:...
The new link is for the readme of the dataset is: https://github.com/cs-chan/ArtGAN/tree/master/WikiArt%20Dataset And the download link is: http://web.fsktm.um.edu.my/~cschan/source/ICIP2017/wikiart.zip
Hi there, Have you guys considered using the TF Dataset API? https://www.tensorflow.org/programmers_guide/datasets#using_high-level_apis One thing that makes it really appealing is that you can easily set up input pipelines to remotely...
I am trying to run this on windows 10. Python 3.6 64bit tensorflow-gpu 1.4.0 With wikiart in ./data/wikiart When runnning: `python main.py --epoch 25 --learning_rate .0001 --beta 0.5 --batch_size 16...
Besides acquiring the pretrained models, I was aiming to train a CAN model for myself to see the progress in learning but sadly, I've encountered mode-collapse issue at a fairly...
Hi, As I run 'bash experiments/train_can_paper.sh' and 'bash experiments/eval_can_paper.sh' with the provided code, I always get the same error below: ResourceExhaustedError (see above for traceback): OOM when allocating tensor of...
Hello, I was trying to get a pretrained model for vanilla can via google drive but the folder was empty when I last checked like an hour ago. Could you...