refactor: 从data 模块分离tokenizer

This commit is contained in:
2026-04-04 16:12:58 +08:00
parent b531232a9b
commit bd9741dc5f
16 changed files with 108 additions and 92 deletions
+2 -1
View File
@@ -6,7 +6,8 @@ from astrai.config import (
TrainConfig,
)
from astrai.factory import BaseFactory
from astrai.data import BpeTokenizer, DatasetFactory
from astrai.data import DatasetFactory
from astrai.tokenizer import BpeTokenizer
from astrai.inference.generator import (
BatchGenerator,
EmbeddingEncoder,