The problem is that Pandas use float64 and int64 numpy dtypes by ... Furthermore, it is not even possible to change this default behaviour.
確定! 回上一頁