File size: 820 Bytes
17ac2d5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
from datasets import load_dataset, concatenate_datasets
from tokenizers import trainers, Tokenizer, normalizers, ByteLevelBPETokenizer

model_dir = "./"  # ${MODEL_DIR}

# load dataset
dataset = load_dataset("json", data_files=["/mnt/disks/flaxdisk/corpus/norwegian_colossal_corpus_validation.json","/mnt/disks/flaxdisk/corpus/special_chars.json"], split='train')


# Instantiate tokenizer
tokenizer = ByteLevelBPETokenizer()

def batch_iterator(batch_size=1000):
    for i in range(0, len(dataset), batch_size):
        yield dataset[i: i + batch_size]["text"]

# Customized training
tokenizer.train_from_iterator(batch_iterator(), vocab_size=50265, min_frequency=2, special_tokens=[
    "<s>",
    "<pad>",
    "</s>",
    "<unk>",
    "<mask>",
])


# Save files to disk
tokenizer.save(f"{model_dir}/tokenizer.json")