Update code and remove extra parts #64
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
According to the given structure in the paper, there is an extra convolutional layer in green block (inp_res). Also, before the final layer there is another extra layer called Input_Dec_Output. It is probably because the article is a little vague, but I checked this architecture with the paper of the winner of BraTS 2019. The winner of BraTS 2019 has used the same architecture, but they have explained it much clear. This is their paper : https://link.springer.com/chapter/10.1007/978-3-030-46640-4_22
Tensorflow has implemented group normalization. First install tensorflow addons using !pip install -q -U tensorflow-addons==0.9.1, then use it.
In order to support tensorflow 2, I've replaced some functions of keras with their equivalents in tensorflow.
As I checked the code with these versions and it works, I recommend to use these versions: tensorflow(2.1.0), tensorflow addons (0.9.1) and keras (2.2.4)
As we have upgraded green block in previous commits, it causes some errors. I compared the architecture with that of the winner of 2019, so I understood the number of filters in some layers should be changed.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There are three extra layers in this model (inp_res in green block and two blue blocks). As the winner of BraTS 2019 used the same architecture and explained it more clear, I have removed extra layers and updated the number of convolution filters. Also, I've upgraded the code and now it supports tensorflow version 2.