"""
pretrained_bert_model = nemo_nlp.huggingface.BERT(
pretrained_model_name=args.pretrained_bert_model, factory=nf)
tokenizer = BertTokenizer.from_pretrained(args.pretrained_bert_model)
hidden_size = pretrained_bert_model.local_parameters["hidden_size"]
data_desc = JointIntentSlotDataDesc(args.data_dir, args.do_lower_case,
args.dataset_name)
query = args.query
if args.do_lower_case:
query = query.lower()
data_layer = nemo_nlp.BertJointIntentSlotInferDataLayer(
queries=[query],
tokenizer=tokenizer,
max_seq_length=args.max_seq_length,
batch_size=1)
# Create sentence classification loss on top
classifier = nemo_nlp.JointIntentSlotClassifier(
hidden_size=hidden_size,
num_intents=data_desc.num_intents,
num_slots=data_desc.num_slots,
dropout=args.fc_dropout)
ids, type_ids, input_mask, loss_mask, subtokens_mask = data_layer()
hidden_states = pretrained_bert_model(input_ids=ids,
token_type_ids=type_ids,
attention_mask=input_mask)
tokenizer = BertTokenizer.from_pretrained(args.pretrained_bert_model)
hidden_size = pretrained_bert_model.local_parameters["hidden_size"]
data_desc = JointIntentSlotDataDesc(
args.dataset_name, args.data_dir, args.do_lower_case)
query = args.query
if args.do_lower_case:
query = query.lower()
dataset = nemo_nlp.BertJointIntentSlotInferDataset(
queries=[query],
tokenizer=tokenizer,
max_seq_length=args.max_seq_length)
data_layer = nemo_nlp.BertJointIntentSlotInferDataLayer(dataset,
batch_size=1)
# Create sentence classification loss on top
classifier = nemo_nlp.JointIntentSlotClassifier(
hidden_size=hidden_size,
num_intents=data_desc.num_intents,
num_slots=data_desc.num_slots,
dropout=args.fc_dropout)
ids, type_ids, input_mask, slot_mask = data_layer()
hidden_states = pretrained_bert_model(input_ids=ids,
token_type_ids=type_ids,
attention_mask=input_mask)