|
embedded = self.mlp(embedded) # (n, seq_len, word_vec_size) |
Hi, I find a mlp in the language encoder, which is used to project the original word embeddings into new representations. While it is not mentioned in the paper, so I am just curious about why you do this in your actual implementation? Is it a necessary operation to make the model work? How would it affect the model's performance?
Looking forward to hearing from you. Thank you so much.
MAttNet/lib/layers/lang_encoder.py
Line 57 in de6f41c
Hi, I find a mlp in the language encoder, which is used to project the original word embeddings into new representations. While it is not mentioned in the paper, so I am just curious about why you do this in your actual implementation? Is it a necessary operation to make the model work? How would it affect the model's performance?
Looking forward to hearing from you. Thank you so much.