tf.contrib.framework.convolutional_orthogonal_2d

Initializer that generates a 2D orthogonal kernel for ConvNets.

tf.contrib.framework.convolutional_orthogonal_2d(
    gain=1.0, seed=None, dtype=tf.dtypes.float32
)

The shape of the tensor must have length 4. The number of input filters must not exceed the number of output filters. The orthogonality(==isometry) is exact when the inputs are circular padded. There are finite-width effects with non-circular padding (e.g. zero padding). See algorithm 1 in (Xiao et al., 2018).

Args
gain	Multiplicative factor to apply to the orthogonal matrix. Default is 1. This has the effect of scaling the output 2-norm by a factor of gain.
seed	A Python integer. Used to create random seeds. See tf.compat.v1.set_random_seed for behavior.
dtype	Default data type, used if no dtype argument is provided when calling the initializer. Only floating point types are supported.

References:

Xiao et al., 2018 (pdf)

Methods

from_config

View source

@classmethod
from_config(
    config
)

Instantiates an initializer from a configuration dictionary.

Example:

initializer = RandomUniform(-1, 1)
config = initializer.get_config()
initializer = RandomUniform.from_config(config)

Args
config	A Python dictionary. It will typically be the output of get_config.

Returns
An Initializer instance.

get_config

View source

get_config()

Returns the configuration of the initializer as a JSON-serializable dict.

Returns
A JSON-serializable Python dict.

__call__

View source

__call__(
    shape, dtype=None, partition_info=None
)

Returns a tensor object initialized as specified by the initializer.

Args
shape	Shape of the tensor.
dtype	Optional dtype of the tensor. If not provided use the initializer dtype.
partition_info	Optional information about the possible partitioning of a tensor.