No cache version.
Caching disabled. Default setting for this page:enabled (code LNG204)
If the display is too slow, you can disable the user mode to view the cached version.
Predefined Constants
The constants below are defined by this extension, and will only be available when the extension has either been compiled into PHP or dynamically loaded at runtime.
-
Training algorithms
-
FANN_TRAIN_INCREMENTAL
(integer) - Standard backpropagation algorithm, where the weights are updated after each training pattern. This means that the weights are updated many times during a single epoch. For this reason some problems, will train very fast with this algorithm, while other more advanced problems will not train very well.
-
FANN_TRAIN_BATCH
(integer) - Standard backpropagation algorithm, where the weights are updated after calculating the mean square error for the whole training set. This means that the weights are only updated once during a epoch. For this reason some problems, will train slower with this algorithm. But since the mean square error is calculated more correctly than in incremental training, some problems will reach a better solutions with this algorithm.
-
FANN_TRAIN_RPROP
(integer) - A more advanced batch training algorithm which achieves good results for many problems. The RPROP training algorithm is adaptive, and does therefore not use the learning_rate. Some other parameters can however be set to change the way the RPROP algorithm works, but it is only recommended for users with insight in how the RPROP training algorithm works. The RPROP training algorithm is described by [Riedmiller and Braun, 1993], but the actual learning algorithm used here is the iRPROP- training algorithm which is described by [Igel and Husken, 2000] which is an variety of the standard RPROP training algorithm.
-
FANN_TRAIN_QUICKPROP
(integer) - A more advanced batch training algorithm which achieves good results for many problems. The quickprop training algorithm uses the learning_rate parameter along with other more advanced parameters, but it is only recommended to change these advanced parameters, for users with insight in how the quickprop training algorithm works. The quickprop training algorithm is described by [Fahlman, 1988].
-
FANN_TRAIN_SARPROP
(integer) - Even more advance training algorithm. Only for version 2.2
-
Activation functions
-
FANN_LINEAR
(integer) - Linear activation function.
-
FANN_THRESHOLD
(integer) - Threshold activation function.
-
FANN_THRESHOLD_SYMMETRIC
(integer) - Threshold activation function.
-
FANN_SIGMOID
(integer) - Sigmoid activation function.
-
FANN_SIGMOID_STEPWISE
(integer) - Stepwise linear approximation to sigmoid.
-
FANN_SIGMOID_SYMMETRIC
(integer) - Symmetric sigmoid activation function, aka. tanh.
-
FANN_SIGMOID_SYMMETRIC_STEPWISE
(integer) - Stepwise linear approximation to symmetric sigmoid
-
FANN_GAUSSIAN
(integer) - Gaussian activation function.
-
FANN_GAUSSIAN_SYMMETRIC
(integer) - Symmetric gaussian activation function.
-
FANN_GAUSSIAN_STEPWISE
(integer) - Stepwise gaussian activation function.
-
FANN_ELLIOT
(integer) - Fast (sigmoid like) activation function defined by David Elliott.
-
FANN_ELLIOT_SYMMETRIC
(integer) - Fast (symmetric sigmoid like) activation function defined by David Elliott.
-
FANN_LINEAR_PIECE
(integer) - Bounded linear activation function.
-
FANN_LINEAR_PIECE_SYMMETRIC
(integer) - Bounded linear activation function.
-
FANN_SIN_SYMMETRIC
(integer) - Periodical sinus activation function.
-
FANN_COS_SYMMETRIC
(integer) - Periodical cosinus activation function.
-
FANN_SIN
(integer) - Periodical sinus activation function.
-
FANN_COS
(integer) - Periodical cosinus activation function.
-
Error function used during training
-
FANN_ERRORFUNC_LINEAR
(integer) - Standard linear error function.
-
FANN_ERRORFUNC_TANH
(integer) - Tanh error function, usually better but can require a lower learning rate. This error function agressively targets outputs that differ much from the desired, while not targetting outputs that only differ a little that much. This activation function is not recommended for cascade training and incremental training.
-
Stop criteria used during training
-
FANN_STOPFUNC_MSE
(integer) - Stop criteria is Mean Square Error (MSE) value.
-
FANN_STOPFUNC_BIT
(integer) - Stop criteria is number of bits that fail. The number of bits means the number of output neurons which differs more than the bit fail limit (see fann_get_bit_fail_limit, fann_set_bit_fail_limit). The bits are counted in all of the training data, so this number can be higher than the number of training data.
-
Definition of network types used by fann_get_network_type()
-
FANN_NETTYPE_LAYER
(integer) - Each layer only has connections to the next layer.
-
FANN_NETTYPE_SHORTCUT
(integer) - Each layer has connections to all following layers
-
Errors
-
FANN_E_NO_ERROR
(integer) - No error.
-
FANN_E_CANT_OPEN_CONFIG_R
(integer) - Unable to open configuration file for reading.
-
FANN_E_CANT_OPEN_CONFIG_W
(integer) - Unable to open configuration file for writing.
-
FANN_E_WRONG_CONFIG_VERSION
(integer) - Wrong version of configuration file.
-
FANN_E_CANT_READ_CONFIG
(integer) - Error reading info from configuration file.
-
FANN_E_CANT_READ_NEURON
(integer) - Error reading neuron info from configuration file.
-
FANN_E_CANT_READ_CONNECTIONS
(integer) - Error reading connections from configuration file.
-
FANN_E_WRONG_NUM_CONNECTIONS
(integer) - Number of connections not equal to the number expected.
-
FANN_E_CANT_OPEN_TD_W
(integer) - Unable to open train data file for writing.
-
FANN_E_CANT_OPEN_TD_R
(integer) - Unable to open train data file for reading.
-
FANN_E_CANT_READ_TD
(integer) - Error reading training data from file.
-
FANN_E_CANT_ALLOCATE_MEM
(integer) - Unable to allocate memory.
-
FANN_E_CANT_TRAIN_ACTIVATION
(integer) - Unable to train with the selected activation function.
-
FANN_E_CANT_USE_ACTIVATION
(integer) - Unable to use the selected activation function.
-
FANN_E_TRAIN_DATA_MISMATCH
(integer) - Irreconcilable differences between two struct fann_train_data structures.
-
FANN_E_CANT_USE_TRAIN_ALG
(integer) - Unable to use the selected training algorithm.
-
FANN_E_TRAIN_DATA_SUBSET
(integer) - Trying to take subset which is not within the training set.
-
FANN_E_INDEX_OUT_OF_BOUND
(integer) - Index is out of bound.
-
FANN_E_SCALE_NOT_PRESENT
(integer) - Scaling parameters not present.
-
FANN_E_INPUT_NO_MATCH
(integer) - The number of input neurons in the ann data do not match
-
FANN_E_OUTPUT_NO_MATCH
(integer) - The number of output neurons in the ann and data do not match.
English translation
You have asked to visit this site in English. For now, only the interface is translated, but not all the content yet.If you want to help me in translations, your contribution is welcome. All you need to do is register on the site, and send me a message asking me to add you to the group of translators, which will give you the opportunity to translate the pages you want. A link at the bottom of each translated page indicates that you are the translator, and has a link to your profile.
Thank you in advance.
Document created the 30/01/2003, last modified the 26/10/2018
Source of the printed document:https://www.gaudry.be/en/php-rf-fann.constants.html
The infobrol is a personal site whose content is my sole responsibility. The text is available under CreativeCommons license (BY-NC-SA). More info on the terms of use and the author.
References
These references and links indicate documents consulted during the writing of this page, or which may provide additional information, but the authors of these sources can not be held responsible for the content of this page.
The author This site is solely responsible for the way in which the various concepts, and the freedoms that are taken with the reference works, are presented here. Remember that you must cross multiple source information to reduce the risk of errors.