refactor(pretrain): 更新预训练数据集输出目录名称

This commit is contained in:
2025-08-02 17:29:21 +08:00
parent 931403a713
commit 241e24c400
7 changed files with 8 additions and 8 deletions
+1 -1
View File
@@ -5,5 +5,5 @@ if __name__ == "__main__":
dataset = load_dataset("shjwudp/chinese-c4") dataset = load_dataset("shjwudp/chinese-c4")
process_dataset( process_dataset(
dataset_dict=dataset, dataset_dict=dataset,
output_subdir="chinese-c4" output_subdir="chinese-c4-pretrain"
) )
+1 -1
View File
@@ -12,7 +12,7 @@ if __name__ == "__main__":
process_dataset( process_dataset(
dataset_dict=dataset, dataset_dict=dataset,
output_subdir="chinese-wiki", output_subdir="chinese-wiki-pretrain",
max_chunk_num=max_chunk_num, max_chunk_num=max_chunk_num,
chunk_size=chunk_size, chunk_size=chunk_size,
) )
+1 -1
View File
@@ -5,5 +5,5 @@ if __name__ == "__main__":
dataset = load_dataset("HuggingFaceFW/fineweb", "sample-10BT") dataset = load_dataset("HuggingFaceFW/fineweb", "sample-10BT")
process_dataset( process_dataset(
dataset_dict=dataset, dataset_dict=dataset,
output_subdir="english-fineweb", output_subdir="english-fineweb-pretrain",
) )
+1 -1
View File
@@ -5,6 +5,6 @@ if __name__ == "__main__":
dataset = load_dataset("Blaze7451/enwiki_structured_content") dataset = load_dataset("Blaze7451/enwiki_structured_content")
process_dataset( process_dataset(
dataset_dict=dataset, dataset_dict=dataset,
output_subdir="english-wiki", output_subdir="english-wiki-pretrain",
max_chunk_size=5, max_chunk_size=5,
) )
@@ -36,6 +36,6 @@ if __name__ == "__main__":
process_dataset( process_dataset(
dataset_dict=DatasetDict({"train": combined_dataset}), dataset_dict=DatasetDict({"train": combined_dataset}),
output_subdir="chinese-instruct", output_subdir="chinese-instruct-sft",
process_func=process_func, process_func=process_func,
) )
+1 -1
View File
@@ -15,7 +15,7 @@ if __name__ == "__main__":
process_dataset( process_dataset(
dataset_dict=dataset, dataset_dict=dataset,
output_subdir="Ling-Coder-SFT", output_subdir="Ling-Coder-sft",
process_func=process_func process_func=process_func
) )
+2 -2
View File
@@ -24,7 +24,7 @@ if __name__ == "__main__":
dataset = load_dataset("HuggingFaceTB/Magpie-Pro-300K-Filtered-H4") dataset = load_dataset("HuggingFaceTB/Magpie-Pro-300K-Filtered-H4")
process_dataset( process_dataset(
dataset_dict=dataset, dataset_dict=dataset,
output_subdir="belle-sft", output_subdir="Magpie-Pro-300K-sft",
process_func=process_func, process_func=process_func,
split_name="train-sft" split_name="train_sft",
) )