File size: 323 Bytes
b663f38
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
from transformers import AutoTokenizer, AutoModel

# Arabic tokenizer from AraBERT v2
tokenizer = AutoTokenizer.from_pretrained("aubmindlab/bert-base-arabertv2")
base_model = AutoModel.from_pretrained("aubmindlab/bert-base-arabertv2")

def get_tokenizer():
    return tokenizer

def get_base_model():
    return base_model