forked from EleutherAI/gpt-neox
-
Notifications
You must be signed in to change notification settings - Fork 7
/
prepare_data.py
executable file
·82 lines (75 loc) · 2.5 KB
/
prepare_data.py
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
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
# Copyright (c) 2021, EleutherAI
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
Usage:
python prepare_data.py <dataset> [-t <tokenizer>] [-d <data_dir>] [-v <vocab_file>] [-m <merge_file>] [-f]
python prepare_data.py "books1" --tokenizer "HFTokenizer" --data-dir "/fsx/shared/books" --vocab-file "/fsx/pile/20B_tokenizer.json"
"""
from tools.corpora import prepare_dataset, DATA_DOWNLOADERS
import argparse
TOKENIZER_CHOICES = [
"HFGPT2Tokenizer",
"HFTokenizer",
"GPT2BPETokenizer",
"CharLevelTokenizer",
"TiktokenTokenizer",
]
DATASET_CHOICES = [i for i in DATA_DOWNLOADERS.keys() if i != "pass"]
def get_args():
parser = argparse.ArgumentParser(description="Download & preprocess neox datasets")
parser.add_argument(
"dataset",
nargs="?",
default="enwik8",
help="name of dataset to download.",
choices=DATASET_CHOICES,
)
parser.add_argument(
"-t",
"--tokenizer",
default="GPT2BPETokenizer",
choices=TOKENIZER_CHOICES,
help=f'Type of tokenizer to use - choose from {", ".join(TOKENIZER_CHOICES)}',
)
parser.add_argument(
"-d",
"--data-dir",
default=None,
help=f"Directory to which to download datasets / tokenizer "
f"files - defaults to ./data",
)
parser.add_argument(
"-v", "--vocab-file", default=None, help=f"Tokenizer vocab file (if required)"
)
parser.add_argument(
"-m", "--merge-file", default=None, help=f"Tokenizer merge file (if required)"
)
parser.add_argument(
"-f",
"--force-redownload",
dest="force_redownload",
default=False,
action="store_true",
)
return parser.parse_args()
if __name__ == "__main__":
args = get_args()
prepare_dataset(
dataset_name=args.dataset,
tokenizer_type=args.tokenizer,
data_dir=args.data_dir,
vocab_file=args.vocab_file,
merge_file=args.merge_file,
force_redownload=args.force_redownload,
)