Last active
June 30, 2023 10:08
-
-
Save shirayu/6e09123b365ff904c0cf7c4b8d78c97e to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python3 | |
from transformers import AutoTokenizer | |
def main() -> None: | |
txt: str = "鯯を食べたい" | |
m1 = "rinna/japanese-gpt-neox-3.6b-instruction-ppo" | |
t1 = AutoTokenizer.from_pretrained( | |
m1, | |
use_fast=False, | |
) | |
x1 = t1(txt) | |
print(m1, x1, t1.decode(x1["input_ids"]), "\n") | |
m2 = "cyberagent/open-calm-7b" | |
t2 = AutoTokenizer.from_pretrained(m2) | |
x2 = t2(txt) | |
print(m2, x2, t2.decode(x2["input_ids"]), "\n") | |
m3 = "retrieva-jp/t5-xl" | |
t3 = AutoTokenizer.from_pretrained( | |
m3, | |
use_fast=False, | |
) | |
x3 = t3(txt) | |
print(m3, x3, t3.decode(x3["input_ids"]), "\n") | |
if __name__ == "__main__": | |
main() |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
rinna/japanese-gpt-neox-3.6b-instruction-ppo {'input_ids': [240, 182, 182, 18998, 1292, 3], 'attention_mask': [1, 1, 1, 1, 1, 1]} 鯯を食べたい</s> | |
cyberagent/open-calm-7b {'input_ids': [8862, 109, 9714, 1026], 'attention_mask': [1, 1, 1, 1]} 鯯を食べたい | |
retrieva-jp/t5-xl {'input_ids': [8, 2, 14271, 605, 1], 'attention_mask': [1, 1, 1, 1, 1]} <unk> を食べたい</s> |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment