When working with the title only, you can naively split your dataset into train-test randomly – after removing items with the same title.
確定! 回上一頁