refactor(utils): 重构 process_files 函数并添加新功能

This commit is contained in:
2025-07-14 15:18:54 +08:00
parent 4cf9681ebb
commit 55e0e06dc6
2 changed files with 34 additions and 30 deletions
+2 -2
View File
@@ -1,4 +1,4 @@
from utils import process_files, fetch_files
from utils import dump_pkl_files, fetch_files
from tokenizer import BpeTokenizer
import os
@@ -15,4 +15,4 @@ if __name__ == "__main__":
for dir_path in base_dir:
files.extend(fetch_files(dir_path))
process_files(tokenizer, files, base_out_dir)
dump_pkl_files(tokenizer, files, base_out_dir)