Computes the product of x and y and returns 0 if the y is zero, even if x is NaN or infinite.
tf.math.multiply_no_nan(
    x, y, name=None
)
Example:
import numpy as np
x = tf.constant([5.0, np.inf, np.nan])
y = tf.constant([-6.0, 0, 0])
tf.math.multiply_no_nan(x, y) ==> [-30. ,0. ,0.]
import numpy as np
x = tf.constant([-2.0, -5.0, 0, 0])
y = tf.constant([-1.0, -6, np.inf, np.nan])
tf.math.multiply_no_nan(x, y) ==> [2. ,30. ,nan ,nan]
Note this is noncommutative: if y is NaN or infinite and x is 0, the result
will be NaN.
| Args | 
|---|
| x | A Tensor. Must be one of the following types:float32,float64,complex64,complex128. | 
| y | A Tensorwhose dtype is compatible withx. | 
| name | A name for the operation (optional). | 
| Returns | 
|---|
| The element-wise value of the x times y. |