How To Use Output_projection (weights, Biases) In Seq2seq?

I am trying to determine how to use the output_projection parameter in the embedding_rnn_decoder. The seq2seq tutorial code shows the generation of output_projection here -- am I a

Solution 1:

The syntax is unclear to me, are they currently "creating" the variables for the weights and biases or are these actually variables that they are getting?

When the graph is first initialized, those variables are created and for the consecutive batch processing those parameters are learned and reused.

use of output_projecttion

It is not exactly logits (it's projected output); the output projection variables are used to project the output of the previous cell and the projected output is combined with the current cell inputs.

projected_output = tf.matmul(outputs, w) + b