bob.machine.MultipliedHyperbolicTangentActivation

class bob.machine.MultipliedHyperbolicTangentActivation((object)self[, (float)C=1.0[, (float)M=1.0]]) → None :

Bases: bob.machine._machine.Activation

Computes \(f(z) = C \cdot \tanh(Mz)\) as activation function

Builds a new hyperbolic tangent activation fucntion with a given constant for the inner and outter products. Don’t use this if you just want to set the constants to the default values (1.0). In such a case, prefer to use the more efficient bob.machine.HyperbolicTangentActivation.

__init__((object)self[, (float)C=1.0[, (float)M=1.0]]) → None :

Builds a new hyperbolic tangent activation fucntion with a given constant for the inner and outter products. Don’t use this if you just want to set the constants to the default values (1.0). In such a case, prefer to use the more efficient bob.machine.HyperbolicTangentActivation.

Methods

__init__((object)self [, (float)C=1.0 [, …)	Builds a new hyperbolic tangent activation fucntion with a given constant for the inner and outter products.
f((Activation)self, (object)z, (object)res)	Computes the activated value, given an input array z, placing results in res (and returning it)
f_prime((Activation)self, (object)z, (object)res)	Computes the derivative of the activated value, placing results in res (and returning it)
f_prime_from_f((Activation)self, (object)a, …)	Computes the derivative of the activated value, given the activated value a, placing results in res (and returning it)
load((Activation)self, (HDF5File)h5f)	Loads itself from a bob.io.HDF5File
save((Activation)self, (HDF5File)h5f)	Saves itself to a bob.io.HDF5File
unique_identifier((Activation)self)	Returns a unique identifier, used by this class in connection to the Activation registry.

Attributes

C	The outside multiplication factor for the hyperbolic tangent function
M	The inner multiplication factor for the argument

C

The outside multiplication factor for the hyperbolic tangent function

M

The inner multiplication factor for the argument

__call__((Activation)self, (object)z, (object)res) → object :

Computes the activated value, given an input array z, placing results in res (and returning it)

__call__( (Activation)self, (object)z) -> object :

Computes the activated value, given an input array z. Returns a newly allocated array with the same size as z

__call__( (Activation)self, (float)z) -> float :

Computes the activated value, given an input z

f((Activation)self, (object)z, (object)res) → object :

Computes the activated value, given an input array z, placing results in res (and returning it)

f( (Activation)self, (object)z) -> object :

Computes the activated value, given an input array z. Returns a newly allocated array with the answers

f( (Activation)self, (float)z) -> float :

Computes the activated value, given an input z

f_prime((Activation)self, (object)z, (object)res) → object :

Computes the derivative of the activated value, placing results in res (and returning it)

f_prime( (Activation)self, (object)z) -> object :

Computes the derivative of the activated value, given an input array z. Returns a newly allocated array with the same size as z

f_prime( (Activation)self, (float)z) -> float :

Computes the derivative of the activated value.

f_prime_from_f((Activation)self, (object)a, (object)res) → object :

Computes the derivative of the activated value, given the activated value a, placing results in res (and returning it)

f_prime_from_f( (Activation)self, (object)z) -> object :

Computes the derivative of the activated value, given the activated value a. Returns a newly allocated array with the same size as a with the answer.

f_prime_from_f( (Activation)self, (float)a) -> float :

Computes the derivative of the activation value, given the activated value a.

load((Activation)self, (HDF5File)h5f) → None :

Loads itself from a bob.io.HDF5File

save((Activation)self, (HDF5File)h5f) → None :

Saves itself to a bob.io.HDF5File

unique_identifier((Activation)self) → str :

Returns a unique identifier, used by this class in connection to the Activation registry.