def test_per_dim_mean_var_norm():
mean = np.asarray([2.], dtype=np.float32)
inv_stddev = np.asarray([0.5], dtype=np.float32)
x = C.input_variable((1, ))
func = C.per_dim_mean_variance_normalize(x, mean, inv_stddev)
result = func.eval({x: np.asarray([[3.], [1.]], dtype=np.float32)})
assert np.array_equal(result, [[.5], [-.5]])
def test_per_dim_mean_var_norm():
mean = np.asarray([2.], dtype=np.float32)
inv_stddev = np.asarray([0.5], dtype=np.float32)
x = C.input_variable((1,))
func = C.per_dim_mean_variance_normalize(x, mean, inv_stddev)
result = func.eval({x : np.asarray([[3.], [1.]], dtype=np.float32)})
assert np.array_equal(result, [[.5], [-.5]])
def per_dim_mean_variance_normalize(operand, mean, inv_stddev, name=''):
'''
Computes per dimension mean-variance normalization of the specified input operand.
Args:
operand: the variable to be normalized
mean: per dimension mean to use for the normalization
inv_stddev: per dimension standard deviation to use for the normalization
name (str): the name of the node in the network
Returns:
:class:`cntk.Function`
'''
from cntk import per_dim_mean_variance_normalize
return per_dim_mean_variance_normalize(operand, mean, inv_stddev, name)
def per_dim_mean_variance_normalize(operand, mean, inv_stddev, name=''):
'''
Computes per dimension mean-variance normalization of the specified input operand.
Args:
operand: the variable to be normalized
mean: per dimension mean to use for the normalization
inv_stddev: per dimension standard deviation to use for the normalization
name (str): the name of the node in the network
Returns:
:class:`cntk.Function`
'''
from cntk import per_dim_mean_variance_normalize
return per_dim_mean_variance_normalize(operand, mean, inv_stddev, name)
