C5_W4_A1_Transformer_Subclass_v1 wrong expected size in comment?

Course Q&A Deep Learning Specialization Sequence Models
1

Hello. I’m wondering how to interpret comments in exercise UNQ_C5 / Exercise 5 - Encoder.

Ex. line of code, lines 48-49:

# Pass input through the Embedding layer
x = None # (batch_size, input_seq_len, fully_connected_dim)

Assuming I do understand it well and this describes expected shape after performing the operation - shouldn’t comment mention embedding_dim instead fully_connected_dim?

2

Hi wst,

I also believe this should be embedding_dim instead of fully_connected_dim. I’ll report this at the backend. Thanks!

3

Thanks. BTW I found it in couple of other places, too. Maybe it would be good to revisit all those comments.