Week3 - Trigger word detection - Why do we need 2 GRU layers

Course Q&A Deep Learning Specialization Sequence Models
1

I find it interesting in the NN architecture that there are 2 GRU layers.

Question 1: why do we need 2 GRU layers? Why is one ‘gate’ not sufficient?

Question 2: why are the 2 GRU layers implemented slightly differently? i.e. why a dropout is added after batch normalization in the second GRU layer?

image
image966×262 10.3 KB

Thank you!

2

NN architecture is usually constrained by memory and computational resources. To achieve the desired accuracy, the notebook author might have found this architecture to be effective.

Feel free to try a custom NN and observe how results change.