def build_embedding(dictionary,
embed_dim,
path=None,
num_embed_chunks=1):
assert embed_dim % num_embed_chunks == 0, (
f"Number of embedding chunks = {num_embed_chunks} should be " +
f"divisible by the embedding dimension = {embed_dim}")
assert path is None or num_embed_chunks == 1, (
"Loading embedding from a path with number of embedding chunks > 1"
+ " is not yet supported")
num_embeddings = len(dictionary)
padding_idx = dictionary.pad()
# if provided, load from preloaded dictionaries
if path:
emb = Embedding(num_embeddings, embed_dim, padding_idx)
embed_dict = utils.parse_embedding(path)
utils.load_embedding(embed_dict, dictionary, emb)
else:
embed_chunk_dim = embed_dim // num_embed_chunks
emb = nn.ModuleList()
for i in range(num_embed_chunks):
emb.append(
Embedding(num_embeddings, embed_chunk_dim,
padding_idx))
return emb
def load_pretrained_embedding_from_file(embed_path, dictionary,
embed_dim):
num_embeddings = len(dictionary)
padding_idx = dictionary.pad()
embed_tokens = Embedding(num_embeddings, embed_dim, padding_idx)
embed_dict = utils.parse_embedding(embed_path)
utils.print_embed_overlap(embed_dict, dictionary)
return utils.load_embedding(embed_dict, dictionary, embed_tokens)
def build_embedding(dictionary, embed_dim, path=None):
num_embeddings = len(dictionary)
padding_idx = dictionary.pad()
emb = Embedding(num_embeddings, embed_dim, padding_idx)
# if provided, load from preloaded dictionaries
if path:
embed_dict = utils.parse_embedding(path)
utils.load_embedding(embed_dict, dictionary, emb)
return emb
def build_model(cls, args, task):
"""Build a new model instance."""
# make sure that all args are properly defaulted (in case there are any new ones)
base_architecture(args)
encoder_embed_dict = None
if args.encoder_embed_path:
encoder_embed_dict = utils.parse_embedding(args.encoder_embed_path)
utils.print_embed_overlap(encoder_embed_dict,
task.source_dictionary)
decoder_embed_dict = None
if args.decoder_embed_path:
decoder_embed_dict = utils.parse_embedding(args.decoder_embed_path)
utils.print_embed_overlap(decoder_embed_dict,
task.target_dictionary)
encoder = FConvEncoder(
dictionary=task.source_dictionary,
embed_dim=args.encoder_embed_dim,
embed_dict=encoder_embed_dict,
convolutions=eval(args.encoder_layers),
dropout=args.dropout,
max_positions=args.max_source_positions,
)
decoder = FConvDecoder(
dictionary=task.target_dictionary,
embed_dim=args.decoder_embed_dim,
embed_dict=decoder_embed_dict,
convolutions=eval(args.decoder_layers),
out_embed_dim=args.decoder_out_embed_dim,
attention=eval(args.decoder_attention),
dropout=args.dropout,
max_positions=args.max_target_positions,
share_embed=args.share_input_output_embed,
)
return FConvModel(encoder, decoder)