This structure is subject to change at any time. If you need to use the values contained herein, please see the Options functions. If these functions do not fulfill your needs, please open a feature request on our SourceForge project page.
Properties
The type of error that last occurred.
Where to log error messages.
A string representation of the last error.
The learning rate of the network.
The connection rate of the network. Between 0 and 1, 1 meaning fully connected.
Is 1 if shortcut connections are used in the ann otherwise 0 Shortcut connections are connections that skip layers. A fully connected ann with shortcut connections is an ann where neurons have connections to all neurons in all later layers.
ANNs with shortcut connections are created by fann_create_shortcut.
Pointer to the first layer (input layer) in an array of all the layers, including the input and output layer.
Pointer to the layer past the last layer in an array of all the layers, including the input and output layer.
Total number of neurons. Very useful, because the actual neurons are allocated in one long array.
Number of input neurons (not calculating bias)
Number of output neurons (not calculating bias)
Used to contain the error deltas used during training Is allocated during first training session, which means that if we do not train, it is never allocated.
Used to choose which activation function to use in the output layer.
Used to choose which activation function to use in the hidden layers.
Parameters for the activation function in the hidden layers.
Parameters for the activation function in the output layer.
Training algorithm used when calling fann_train_on_... and fann_train_epoch.
Fixed point only. The decimal point, used for shifting the fix point in fixed point integer operations.
Fixed point only. The multiplier, used for multiplying the fix point in fixed point integer operations. Only used in special cases, since the decimal_point is much faster.
An array of six members used by some activation functions to hold results for the hidden layer(s).
An array of six members used by some activation functions to hold values for the hidden layer(s).
An array of six members used by some activation functions to hold results for the output layer.
An array of six members used by some activation functions to hold values for the output layer.
Total number of connections. Very useful, because the actual connections are allocated in one long array.
Used to store outputs in.
The number of data used to calculate the mean square error.
The total error value. The real mean square error is MSE_value/num_MSE.
When using this, training is usually faster. Makes the error used for calculating the slopes higher when the difference is higher.
Decay is used to make the weights not go so high.
Mu is a factor used to increase and decrease the step-size.
Tells how much the step-size should increase during learning.
Tells how much the step-size should decrease during learning.
The minimum step-size.
The maximum step-size.
Used to contain the slope errors used during batch training Is allocated during first training session, which means that if we do not train, it is never allocated.
The previous step taken by the quickprop/rprop procedures. Not allocated if not used.
The slope values used by the quickprop/rprop procedures. Not allocated if not used.