-
Notifications
You must be signed in to change notification settings - Fork 332
Add visualization output via tensorboard to the clustering example #508
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add visualization output via tensorboard to the clustering example #508
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you. Just a couple of minor things.
tensorflow_model_optimization/python/core/clustering/keras/BUILD
Outdated
Show resolved
Hide resolved
tensorflow_model_optimization/python/core/clustering/keras/clustering_callbacks.py
Outdated
Show resolved
Hide resolved
tensorflow_model_optimization/python/core/clustering/keras/clustering_callbacks.py
Show resolved
Hide resolved
b105810
to
5d75774
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. Thanks.
@benkli01, please resolve merge conflict |
Note: Batch-wise summaries of the clustering can be slow and might lead to a warning like this: "WARNING:tensorflow:Method (on_train_batch_end) is slow compared to the batch update (0.102097). Check your callbacks."
5d75774
to
6e4ae44
Compare
Merge conflict resolved and commit message adapted, because the simplification of the example (#516) is already merged. |
-- 5d75774 by Benjamin Klimczak <benjamin.klimczak@arm.com>: Add visualization output via tensorboard to the clustering example. Note: - The current example seems to be overparameterized, so that there is not very much to see in the visualization. This should be addressed by another PR making the example leaner. - Writing the visualization summaries batch-wise leads to a warning message like this: "WARNING:tensorflow:Method (on_train_batch_end) is slow compared to the batch update (0.102097). Check your callbacks." It should be further investigated if this can be avoided. COPYBARA_INTEGRATE_REVIEW=#508 from benkli01:toupstream/clustering-visualization 5d75774 PiperOrigin-RevId: 330547925
The commit has been added to Github (see here) and attributed to Benjamin. Unfortunately, due to the hiccup, I don't think we can ever mark the PR itself as merged. The cause was that I went through the regular merge procedure, but there was an issue that showed up after merging, from an internal check that makes sure that there is the "COPYRIGHT" part at the top of every file. I'm changing things so that the check will run earlier during "import/copybara" and this issue won't arise again. Apologize for the trouble. |
Closing this given the above comment. The merged commit is here: 5603678. |
Thanks for taking care of this, @alanchiao ! |
Some of the latest changes of this PR seem to be missing in the commit to the master branch. I will have a closer look and, if needed, I will just create a new PR to amend this. |
Note: Batch-wise summaries of the clustering can be slow and might lead to a warning like this: