0

I found tensorflow 1b_lm project: https://github.com/tensorflow/models/tree/master/lm_1b

I'm just quite confused about the forth example

Give a sentence, dump the embedding from the LSTM state.

However, the results of this example includes 7 '.npy' files. It seems like it just generates every word embedding for every word in the sentence?

henrycarteruk
  • 12,708
  • 2
  • 36
  • 40
Zachary
  • 11
  • 2

1 Answers1

0

It outputs the internal state of each step, where each step corresponds to the sentence fragment seen so far. The last vector holds the embeddings of the whole sentence

Roni
  • 1