feat(dataset): 更新数据集处理和保存逻辑
This commit is contained in:
@@ -38,7 +38,7 @@ def dump_pkl_files(
|
||||
):
|
||||
def process_line(line: str) -> Tensor:
|
||||
line = json.loads(line)[key]
|
||||
processed_line = encoder(line)
|
||||
processed_line = encoder(line) if encoder else line
|
||||
ids = tokenizer.encode(processed_line)
|
||||
arrow = torch.tensor(ids, dtype=torch.int32)
|
||||
return arrow
|
||||
|
||||
Reference in New Issue
Block a user