Open
Description
bug描述 Describe the Bug
paddle <= 2.5.1
import paddle
import tensorflow as tf
x = paddle.to_tensor([-7.450581e-09],dtype='float32')
y = paddle.to_tensor([-1.1175871e-08],dtype='float32')
result1 = paddle.equal(x, y)
print(result1)
x = tf.constant([-7.450581e-09],dtype='float32')
y = tf.constant([-1.1175871e-08],dtype='float32')
result1 = tf.equal(x, y)
print(result1)
paddle output - Tensor(shape=[1], dtype=bool, place=Place(cpu), stop_gradient=True,
[True]);
tensorflow output - tf.Tensor([False], shape=(1,), dtype=bool);
其他补充信息 Additional Supplementary Information
i have tried with all the dtypes ' float32, float64 and etc that are supported looking forward for more discussion.