def l2_normalize(incoming, dim, epsilon=1e-12, name="l2_normalize"):
""" L2 Normalization.
Normalizes along dimension `dim` using an L2 norm.
For a 1-D tensor with `dim = 0`, computes
output = x / sqrt(max(sum(x**2), epsilon))
```
For `x` with more dimensions, independently normalizes each 1-D slice along
dimension `dim`.
Arguments:
incoming: `Tensor`. Incoming Tensor.
dim: `int`. Dimension along which to normalize.
epsilon: `float`. A lower bound value for the norm. Will use
`sqrt(epsilon)` as the divisor if `norm < sqrt(epsilon)`.
name: `str`. A name for this layer (optional).
Returns:
A `Tensor` with the same shape as `x`.
"""
with tf.name_scope(name) as name:
x = tf.convert_to_tensor(incoming, name="x")
square_sum = tf.reduce_sum(tf.square(x), [dim], keep_dims=True)
x_inv_norm = tf.rsqrt(tf.maximum(square_sum, epsilon))
return tf.multiply(x, x_inv_norm, name=name)```
