pydbm.dbm.interface package

Submodules

pydbm.dbm.interface.dbm_builder module

class pydbm.dbm.interface.dbm_builder.DBMBuilder

Bases: object

The Builder interface in Builder Pattern, which generates the object of DBM.

attenuate_epoch

Attenuate the learning_rate by a factor of learning_attenuate_rate every attenuate_epoch.

feature_neuron_part

Build neurons for feature points in virtual visbile layer.

Build neurons in n layers.

For associating with n-1 layers, the object activates as neurons in hidden layer. On the other hand, for associating with n+1 layers, the object activates as neurons in virtual visible layer.

Parameters:
  • activation_function_list – The list of activation function.
  • neuron_count_list – The list of the number of neurons.
get_result

Return builded restricted boltzmann machines.

Returns:The list of restricted boltzmann machines.
graph_part

Build complete bipartite graph.

Parameters:
  • approximate_interface_list – The list of function approximation.
  • scale – Scale of parameters which will be ParamsInitializer.
  • params_initializer – is-a ParamsInitializer.
  • params_dict – dict of parameters other than size to be input to function ParamsInitializer.sample_f.
hidden_neuron_part

Build neurons in hidden layers.

Parameters:
  • activation_function – Activation function.
  • neuron_count – The number of neurons.
learning_attenuate_rate

Attenuate the learning_rate by a factor of this value every attenuate_epoch.

learning_rate

Learning rate.

visible_neuron_part

Build neurons in visible layer.

Parameters:
  • activation_function – Activating function.
  • neuron_count – The number of neurons.

pydbm.dbm.interface.rt_rbm_builder module

class pydbm.dbm.interface.rt_rbm_builder.RTRBMBuilder

Bases: object

The Builder interface in Builder Pattern, which generates the object of RTRBM.

attenuate_epoch

Attenuate the learning_rate by a factor of learning_attenuate_rate every attenuate_epoch.

get_result

Return builded restricted boltzmann machines.

Returns:The list of restricted boltzmann machines.
graph_part

Build RTRBM graph.

Parameters:
  • approximate_interface – The function approximation.
  • scale – Scale of parameters which will be ParamsInitializer.
  • params_initializer – is-a ParamsInitializer.
  • params_dict – dict of parameters other than size to be input to function ParamsInitializer.sample_f.
hidden_neuron_part

Build neurons in hidden layers.

Parameters:
  • activation_function – Activation function.
  • neuron_count – The number of neurons.
learning_attenuate_rate

Attenuate the learning_rate by a factor of this value every attenuate_epoch.

learning_rate

Learning rate.

rnn_neuron_part

Build neurons for RNN.

Parameters:rnn_activating_function – Activation function
visible_neuron_part

Build neurons in visible layer.

Parameters:
  • activation_function – Activating function.
  • neuron_count – The number of neurons.

Module contents