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
Feature request: segment_argmax #13460
Comments
Contributions are welcome! |
I can try to implement this. Per the example by OP, the segments are Is my understanding correct ? |
yes, my intention was to use these indices to slice the original tensor for all the max tensors |
I will go to work on this. |
Any update on this? I would also extend the feature request to support unsorted_segment_argmax, segment_argmax, unsorted_segment_argmin, and segment_argmin. |
Would also like to throw my vote on this as a feature ; this would be super useful. |
This issue is stale because it has been open for 180 days with no activity. It will be closed if no further activity occurs. Thank you. |
How is this progressing? |
Currently trying to return argmax from a tensor for selected slices (segments)
slices do not have the same length, so reshaping can't be used.
was looking for a function similar to tf.segment_max, only with indices as the return value.
for e.g
a = [1, 2, 3, 4, 5, 6]
seg = [0, 0, 0, 1, 1, 2]
tf.segment_argmax return value will be
[2, 4, 5]
The text was updated successfully, but these errors were encountered: