在Tensorflow中可视化注意力激活

Question

在TensorFlow的seq2seq模型中，有没有办法可视化某些输入的注意权重，如上面链接中的图（来自Bahdanau等，2014）？我已经找到了TensorFlow's github issue这个问题，但是我无法在会话期间找到如何获取注意力掩码。

Answer 1

我还希望将Tensorflow seq2seq ops的注意力量可视化为我的文本摘要任务。我认为临时解决方案是使用session.run（）来评估上面提到的注意掩码张量。有趣的是，原始的seq2seq.py操作被认为是遗留版本，并且无法在github中轻松找到，因此我只使用了0.12.0滚轮分发中的seq2seq.py文件并对其进行了修改。为了绘制热图，我使用了'Matplotlib'包，非常方便。

新闻标题文本的关注可视化的最终输出如下所示：

我修改了以下代码： https://github.com/rockingdingo/deepnlp/tree/master/deepnlp/textsum#attention-visualization

seq2seq_attn.py

# Find the attention mask tensor in function attention_decoder()-> attention()
# Add the attention mask tensor to ‘return’ statement of all the function that calls the attention_decoder(), 
# all the way up to model_with_buckets() function, which is the final function I use for bucket training.

def attention(query):
  """Put attention masks on hidden using hidden_features and query."""
  ds = []  # Results of attention reads will be stored here.

  # some code

  for a in xrange(num_heads):
    with variable_scope.variable_scope("Attention_%d" % a):
      # some code

      s = math_ops.reduce_sum(v[a] * math_ops.tanh(hidden_features[a] + y),
                              [2, 3])
      # This is the attention mask tensor we want to extract
      a = nn_ops.softmax(s)

      # some code

  # add 'a' to return function
  return ds, a

seq2seq_model_attn.py

# modified model.step() function and return masks tensor
self.outputs, self.losses, self.attn_masks = seq2seq_attn.model_with_buckets(…)

# use session.run() to evaluate attn masks
attn_out = session.run(self.attn_masks[bucket_id], input_feed)
attn_matrix = ...

predict_attn.py和eval.py

# Use the plot_attention function in eval.py to visual the 2D ndarray during prediction.

eval.plot_attention(attn_matrix[0:ty_cut, 0:tx_cut], X_label = X_label, Y_label = Y_label)

可能在未来，tensorflow将有更好的方法来提取和可视化注意力量图。有什么想法吗？