paddle的element-wise加减乘除不支持浮点类型与整型tensor之间运算

[liuqi45]

bug描述 Describe the Bug

import paddle
a = paddle.zeros((2,2),dtype = paddle.float64)
b = paddle.zeros((2,2),dtype = paddle.int64)
c = a * b
print(c)

错误信息:
TypeError: (InvalidType) Type promotion only support calculations between floating-point numbers and between complex and real numbers. But got different data type x: float64, y: int64. (at ..\paddle/phi/common/type_promotion.h:233)

其他补充信息 Additional Supplementary Information

pytorch支持同样运算

import torch
a = torch.zeros((2,2),dtype = torch.float64)
b = torch.zeros((2,2),dtype = torch.int64)
c = a * b
print(c)

[FeixLiu]

感谢您的建议，我们会对此需求进行内部评估。

[fangfangssj]

https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/guides/advanced/auto_type_promotion_cn.html ，飞桨的隐式类型提升为了更方便用户能够快速排查由于类型导致的问题，Tensor 之间的自动类型提升将仅支持浮点型之间的计算
好像这个是飞桨的特性