koichiro11 / residual-attention-network Goto Github PK

Hello, could you please provide download url for pretrained model on ImageNet? Thanks a lot!

the validation accuracy and the test accuracy was only 80%, I did not change the code just added the test set code, who can help me？

Hey,when I run this code
there is an error
File "/home/user/anaconda3/envs/tf1.4/lib/python3.6/site-packages/tensorflow/python/layers/base.py", line 1224, in __init__ layer.outbound_nodes.append(self) AttributeError: 'UpSampling2D' object has no attribute 'outbound_nodes'
Do you know how to fix it?
Thank you .

hi, @koichiro11
i am so appreciated with your great job.
have you test the model in cifar-10, what is the result?

hello, @koichiro11 . Thanks for your code. I run this code with Cifar10, but the accurrency only about 81%. It lower than the result of ResNet . How about your results? Whether i have some wrong ? thank you. I am looking forward your reply.

Did anyone meet this error?
tensorflow.python.framework.errors_impl.InvalidArgumentError: buffer_size must be greater than zero.
[[Node: ShuffleDataset_1 = ShuffleDataset[output_shapes=[[]], output_types=[DT_STRING], reshuffle_each_iteration=true](ShuffleDataset, ShuffleDataset_1/Cast, ShuffleDataset/seed, ShuffleDataset/seed)]]
[[Node: OneShotIterator = OneShotIteratorcontainer="", dataset_factory=_make_dataset_6fgP5f2Haq0[], output_shapes=[[?,32,32,3], [?,5]], output_types=[DT_FLOAT, DT_INT32], shared_name="", _device="/job:localhost/replica:0/task:0/device:CPU:0"]]

hello,koichiro11,when i use your code to train on cifar-10,i get the 0.82 accuracy.But i don't know how to visualize the result images using residual attention network and how to test it , would you mind telling me how to do it,thanks!