feat: add Ultra-FineWeb-L3 QA Synthetic pre-train export scripts (en/zh)

This commit is contained in:
2026-07-31 10:03:49 +08:00
parent 2b3bf442e9
commit cc451e5492
2 changed files with 40 additions and 0 deletions
@@ -0,0 +1,20 @@
from datasets import load_dataset
from pipeline import export_dataset
def process_func(input_dict: dict):
return {"text": input_dict["content"]}
if __name__ == "__main__":
dataset = load_dataset(
"openbmb/Ultra-FineWeb-L3",
"Ultra-FineWeb-L3-en-QA-Synthetic",
split="train",
)
export_dataset(
dataset=dataset,
output_dir="./dataset",
output_prefix="ultra-fineweb-l3-en-qa-synthetic-pretrain",
process_func=process_func,
)
@@ -0,0 +1,20 @@
from datasets import load_dataset
from pipeline import export_dataset
def process_func(input_dict: dict):
return {"text": input_dict["content"]}
if __name__ == "__main__":
dataset = load_dataset(
"openbmb/Ultra-FineWeb-L3",
"Ultra-FineWeb-L3-zh-QA-Synthetic",
split="train",
)
export_dataset(
dataset=dataset,
output_dir="./dataset",
output_prefix="ultra-fineweb-l3-zh-qa-synthetic-pretrain",
process_func=process_func,
)