File size: 765 Bytes
945de56
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
"""BERT for sentence-pair boundary classification."""

from transformers import (
    AutoTokenizer,
    BertForSequenceClassification,
    PreTrainedTokenizerFast,
)

from src.datasets.combined_pairs_dataset import NUM_LABELS, ID2LABEL, LABEL2ID

BASE_MODEL = "bert-base-uncased"


def load_bert(
    pretrained: str = BASE_MODEL,
) -> BertForSequenceClassification:
    """Instantiate BERT for 3-class sentence-pair classification."""
    return BertForSequenceClassification.from_pretrained(
        pretrained,
        num_labels=NUM_LABELS,
        id2label=ID2LABEL,
        label2id=LABEL2ID,
    )


def load_bert_tokenizer(
    pretrained: str = BASE_MODEL,
) -> PreTrainedTokenizerFast:
    return AutoTokenizer.from_pretrained(pretrained, use_fast=True)