refactor(data): 重构数据处理流程以适应新需求

This commit is contained in:
2025-07-22 10:03:06 +08:00
parent b588c30f09
commit 5ea1c6355d
3 changed files with 7 additions and 4 deletions
+2 -2
View File
@@ -16,7 +16,7 @@ def get_processor(tokenizer: BpeTokenizer):
masks = torch.zeros_like(tokens)
masks[:len(prefix_ids)] = 1
return tokens, masks
return {"sequence": tokens, "mask": masks}
return processor
@@ -34,4 +34,4 @@ if __name__ == "__main__":
processor = get_processor(tokenizer)
dump_pkl_files(files, base_out_dir,)
dump_pkl_files(files, base_out_dir,processor, ["sequence", "mask"])