Files
DataPipeline/scripts/supervised_finetuning/sft_alpaca_gpt4.py
T
jiachun efba7a009f feat: add sft scripts for Magicoder-Evol-Instruct-110K and alpaca-gpt4-data
- sft_magicoder.py: export ise-uiuc/Magicoder-Evol-Instruct-110K
- sft_alpaca_gpt4.py: export llm-wizard/alpaca-gpt4-data with instruction+input
- remove deprecated SFT scripts
2026-06-01 15:50:13 +08:00

23 lines
599 B
Python

from datasets import load_dataset
from pipeline import export_dataset
def process_func(input_dict: dict):
instruction = input_dict["instruction"]
inp = input_dict.get("input", "")
if inp:
query = instruction + "\n" + inp
else:
query = instruction
return {"query": query, "response": input_dict["output"]}
if __name__ == "__main__":
dataset = load_dataset("llm-wizard/alpaca-gpt4-data")
export_dataset(
dataset=dataset["train"],
output_dir="./dataset",
output_prefix="alpaca-gpt4-data",
process_func=process_func,
)