Assign to slice in Tensorflow2

Question

Suppose I have the following variable x and I wish to assign 1 to all the zeros.

x = tf.random.poisson((3,5), 1)
idx = x == 0
y = tf.Variable(tf.zeros_like(x, dtype=tf.float32))
y[idx] = 1.0

When I try y[idx].assign(1.0) I get the error: AttributeError: 'tensorflow.python.framework.ops.EagerTensor' object has no attribute 'assign' .

I also tried other variations of this like below with no luck:

y[idx].assign(tf.ones_like(y[idx]))
y[idx] = tf.ones_like(y[idx], dtype=tf.float32)

Optional

In my case what I really have to assign is -infinity. I'm assuming if I can do above I can assign that, but please do let me know if this is more complicated for some reason.

Answer 1

Without further context, I don't know why you want to update a Variable . But when you create one with a value (under tensorflow 2.0 eager execution), you can preset the matrix first:

>>> x
<tf.Tensor: shape=(3, 5), dtype=float32, numpy=
array([[2., 2., 2., 0., 0.],
       [1., 1., 1., 2., 0.],
       [1., 0., 2., 0., 1.]], dtype=float32)>
>>> idx
<tf.Tensor: shape=(3, 5), dtype=bool, numpy=
array([[False, False, False,  True,  True],
       [False, False, False, False,  True],
       [False,  True, False,  True, False]])>
>>> tf.where(idx, np.inf, tf.zeros_like(x, dtype=tf.float32))
<tf.Tensor: shape=(3, 5), dtype=float32, numpy=
array([[ 0.,  0.,  0., inf, inf],
       [ 0.,  0.,  0.,  0., inf],
       [ 0., inf,  0., inf,  0.]], dtype=float32)>
>>> y = tf.Variable(tf.where(idx, np.inf, tf.zeros_like(x, dtype=tf.float32)), dtype=tf.float32)
>>> y
<tf.Variable 'Variable:0' shape=(3, 5) dtype=float32, numpy=
array([[ 0.,  0.,  0., inf, inf],
       [ 0.,  0.,  0.,  0., inf],
       [ 0., inf,  0., inf,  0.]], dtype=float32)>