You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I try to use the gpu in the one-pixel attack for imagenet, because the model run on imagenet is too slow, but I use os.environ["CUDA_VISIBLE_DEVICES"] seems did not work, can you give some help. Thanks so much !
The text was updated successfully, but these errors were encountered:
Did you run the code in the notebook to check if the GPU is enabled? You can also try this StackOverflow post. If it's enabled for TensorFlow, it should be enabled for Keras.
Also, the repo now supports Google Colab, so you can also try out a free GPU cloud instance that way.
I try to use the gpu in the one-pixel attack for imagenet, because the model run on imagenet is too slow, but I use os.environ["CUDA_VISIBLE_DEVICES"] seems did not work, can you give some help. Thanks so much !
The text was updated successfully, but these errors were encountered: