Tensorflow: Getting Attention Activations to Visualize Attention in Seq2Seq

Created on 27 Mar 2016  ·  3Comments  ·  Source: tensorflow/tensorflow

All attention papers feature some visualization of the attention weights on some input. Has anyone been able to run a sample through the Seq2Seq Attention Decoder model in translate.py and get the attention activations to do such a visualization?

picture nicholaslocascio

Most helpful comment

The link is broken. What is the correct link?

picture jarheadfa on 27 Jun 2017
👍10

All 3 comments

The attention mask is available as a tensor here :
https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/ops/seq2seq.py#L522

It should be easy to fetch it out during a run call and visualize it. You can try posting this to StackOverflow to see if someone in the general community has done this visualization. I am closing this issue, since we have the required functionality in TensorFlow.

keveman picture keveman on 28 Mar 2016

The link is broken. What is the correct link?

jarheadfa picture jarheadfa on 27 Jun 2017
👍10

same problem

guotong1988 picture guotong1988 on 28 Aug 2017
Was this page helpful?
0 / 5 - 0 ratings

Related issues

jacobma-create picture jacobma-create  ·  3Comments
ppwwyyxx picture ppwwyyxx  ·  3Comments
waleedka picture waleedka  ·  3Comments
fobus42 picture fobus42  ·  3Comments
kepricon picture kepricon  ·  3Comments