refactor: 修改项目结构

This commit is contained in:
2026-03-21 21:25:59 +08:00
parent 5b47ab1dea
commit 909f1b400c
32 changed files with 535 additions and 703 deletions
+14
View File
@@ -0,0 +1,14 @@
from datasets import load_dataset
from pipeline import export_dataset
if __name__ == "__main__":
dataset = load_dataset(
"opencsg/chinese-cosmopedia",
data_files={"train": [f"data/000{i:02d}.parquet" for i in range(25)]}
)
export_dataset(
dataset=dataset["train"],
output_dir="./dataset",
output_prefix="chinese-cosmopedia-pretrain",
chunk_size=1_000_000,
)