/usr/share/doc/libfann-doc/html/files/fann_data-h.html is in libfann-doc 2.1.0~beta~dfsg-8ubuntu1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 | <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0//EN" "http://www.w3.org/TR/REC-html40/strict.dtd">
<html><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><title>FANN Datatypes</title><link rel="stylesheet" type="text/css" href="../styles/main.css"><script language=JavaScript src="../javascript/main.js"></script><script language=JavaScript src="../javascript/prettify.js"></script><script language=JavaScript src="../javascript/searchdata.js"></script></head><body class="ContentPage" onLoad="NDOnLoad();prettyPrint();"><script language=JavaScript><!--
if (browserType) {document.write("<div class=" + browserType + ">");if (browserVer) {document.write("<div class=" + browserVer + ">"); }}// --></script>
<!-- Generated by Natural Docs, version 1.51 -->
<!-- http://www.naturaldocs.org -->
<!-- saved from url=(0026)http://www.naturaldocs.org -->
<div id=Content><div class="CSection"><div class=CTopic id=MainTopic><h1 class=CTitle><a name="FANN_Datatypes"></a>FANN Datatypes</h1><div class=CBody><p>The two main datatypes used in the fann library is <a href="#struct_fann" class=LClass id=link14 onMouseOver="ShowTip(event, 'tt1', 'link14')" onMouseOut="HideTip('tt1')">struct fann</a>, which represents an artificial neural network, and <a href="fann_train-h.html#struct_fann_train_data" class=LClass id=link15 onMouseOver="ShowTip(event, 'tt2', 'link15')" onMouseOut="HideTip('tt2')">struct fann_train_data</a>, which represent training data.</p><!--START_ND_SUMMARY--><div class=Summary><div class=STitle>Summary</div><div class=SBorder><table border=0 cellspacing=0 cellpadding=0 class=STable><tr class="SMain"><td class=SEntry><a href="#FANN_Datatypes" >FANN Datatypes</a></td><td class=SDescription>The two main datatypes used in the fann library is <a href="#struct_fann" class=LClass id=link1 onMouseOver="ShowTip(event, 'tt1', 'link1')" onMouseOut="HideTip('tt1')">struct fann</a>, which represents an artificial neural network, and <a href="fann_train-h.html#struct_fann_train_data" class=LClass id=link2 onMouseOver="ShowTip(event, 'tt2', 'link2')" onMouseOut="HideTip('tt2')">struct fann_train_data</a>, which represent training data.</td></tr><tr class="SGroup SIndent1"><td class=SEntry><a href="#Types" >Types</a></td><td class=SDescription></td></tr><tr class="SType SIndent2 SMarked"><td class=SEntry><a href="#fann_type" >fann_type</a></td><td class=SDescription>fann_type is the type used for the weights, inputs and outputs of the neural network.</td></tr><tr class="SGroup SIndent1"><td class=SEntry><a href="#Enumerations_and_Constants" >Enumerations and Constants</a></td><td class=SDescription></td></tr><tr class="SEnumeration SIndent2 SMarked"><td class=SEntry><a href="#fann_train_enum" >fann_train_enum</a></td><td class=SDescription>The Training algorithms used when training on <a href="fann_train-h.html#struct_fann_train_data" class=LClass id=link3 onMouseOver="ShowTip(event, 'tt2', 'link3')" onMouseOut="HideTip('tt2')">struct fann_train_data</a> with functions like <a href="fann_train-h.html#fann_train_on_data" class=LFunction id=link4 onMouseOver="ShowTip(event, 'tt3', 'link4')" onMouseOut="HideTip('tt3')">fann_train_on_data</a> or <a href="fann_train-h.html#fann_train_on_file" class=LFunction id=link5 onMouseOver="ShowTip(event, 'tt4', 'link5')" onMouseOut="HideTip('tt4')">fann_train_on_file</a>. </td></tr><tr class="SConstant SIndent2"><td class=SEntry><a href="#FANN_TRAIN_NAMES" >FANN_TRAIN_NAMES</a></td><td class=SDescription>Constant array consisting of the names for the training algorithms, so that the name of an training function can be received by:</td></tr><tr class="SEnumeration SIndent2 SMarked"><td class=SEntry><a href="#fann_activationfunc_enum" >fann_activationfunc_enum</a></td><td class=SDescription>The activation functions used for the neurons during training. </td></tr><tr class="SConstant SIndent2"><td class=SEntry><a href="#FANN_ACTIVATIONFUNC_NAMES" >FANN_ACTIVATIONFUNC_NAMES</a></td><td class=SDescription>Constant array consisting of the names for the activation function, so that the name of an activation function can be received by:</td></tr><tr class="SEnumeration SIndent2 SMarked"><td class=SEntry><a href="#fann_errorfunc_enum" >fann_errorfunc_enum</a></td><td class=SDescription>Error function used during training.</td></tr><tr class="SConstant SIndent2"><td class=SEntry><a href="#FANN_ERRORFUNC_NAMES" >FANN_ERRORFUNC_NAMES</a></td><td class=SDescription>Constant array consisting of the names for the training error functions, so that the name of an error function can be received by:</td></tr><tr class="SEnumeration SIndent2 SMarked"><td class=SEntry><a href="#fann_stopfunc_enum" >fann_stopfunc_enum</a></td><td class=SDescription>Stop criteria used during training.</td></tr><tr class="SConstant SIndent2"><td class=SEntry><a href="#FANN_STOPFUNC_NAMES" >FANN_STOPFUNC_NAMES</a></td><td class=SDescription>Constant array consisting of the names for the training stop functions, so that the name of a stop function can be received by:</td></tr><tr class="SEnumeration SIndent2 SMarked"><td class=SEntry><a href="#fann_network_type_enum" >fann_network_type_enum</a></td><td class=SDescription>Definition of network types used by <a href="fann-h.html#fann_get_network_type" class=LFunction id=link6 onMouseOver="ShowTip(event, 'tt5', 'link6')" onMouseOut="HideTip('tt5')">fann_get_network_type</a></td></tr><tr class="SConstant SIndent2"><td class=SEntry><a href="#FANN_NETWORK_TYPE_NAMES" >FANN_NETWORK_TYPE_NAMES</a></td><td class=SDescription>Constant array consisting of the names for the network types, so that the name of an network type can be received by:</td></tr><tr class="SGroup SIndent1"><td class=SEntry><a href="#Types" >Types</a></td><td class=SDescription></td></tr><tr class="SType SIndent2 SMarked"><td class=SEntry><a href="#fann_callback_type" >fann_callback_type</a></td><td class=SDescription>This callback function can be called during training when using <a href="fann_train-h.html#fann_train_on_data" class=LFunction id=link7 onMouseOver="ShowTip(event, 'tt3', 'link7')" onMouseOut="HideTip('tt3')">fann_train_on_data</a>, <a href="fann_train-h.html#fann_train_on_file" class=LFunction id=link8 onMouseOver="ShowTip(event, 'tt4', 'link8')" onMouseOut="HideTip('tt4')">fann_train_on_file</a> or <a href="fann_cascade-h.html#fann_cascadetrain_on_data" class=LFunction id=link9 onMouseOver="ShowTip(event, 'tt6', 'link9')" onMouseOut="HideTip('tt6')">fann_cascadetrain_on_data</a>.</td></tr><tr class="SClass"><td class=SEntry><a href="#struct_fann_error" id=link10 onMouseOver="ShowTip(event, 'tt7', 'link10')" onMouseOut="HideTip('tt7')">struct fann_error</a></td><td class=SDescription>Structure used to store error-related information, both <a href="#struct_fann" class=LClass id=link11 onMouseOver="ShowTip(event, 'tt1', 'link11')" onMouseOut="HideTip('tt1')">struct fann</a> and <a href="fann_train-h.html#struct_fann_train_data" class=LClass id=link12 onMouseOver="ShowTip(event, 'tt2', 'link12')" onMouseOut="HideTip('tt2')">struct fann_train_data</a> can be casted to this type.</td></tr><tr class="SClass"><td class=SEntry><a href="#struct_fann" id=link13 onMouseOver="ShowTip(event, 'tt1', 'link13')" onMouseOut="HideTip('tt1')">struct fann</a></td><td class=SDescription>The fast artificial neural network(fann) structure.</td></tr><tr class="SGroup SIndent1"><td class=SEntry><a href="#struct_fann.Types" >Types</a></td><td class=SDescription></td></tr><tr class="SType SIndent2 SMarked"><td class=SEntry><a href="#struct_fann.fann_connection" >fann_connection</a></td><td class=SDescription>Describes a connection between two neurons and its weight</td></tr></table></div></div><!--END_ND_SUMMARY--></div></div></div>
<div class="CGroup"><div class=CTopic><h3 class=CTitle><a name="Types"></a>Types</h3></div></div>
<div class="CType"><div class=CTopic><h3 class=CTitle><a name="fann_type"></a>fann_type</h3><div class=CBody><p>fann_type is the type used for the weights, inputs and outputs of the neural network.</p><h4 class=CHeading>fann_type is defined as a</h4><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry>float</td><td class=CDLDescription>if you include fann.h or floatfann.h</td></tr><tr><td class=CDLEntry>double</td><td class=CDLDescription>if you include doublefann.h</td></tr><tr><td class=CDLEntry>int</td><td class=CDLDescription>if you include fixedfann.h (please be aware that fixed point usage is only to be used during execution, and not during training).</td></tr></table></div></div></div>
<div class="CGroup"><div class=CTopic><h3 class=CTitle><a name="Enumerations_and_Constants"></a>Enumerations and Constants</h3></div></div>
<div class="CEnumeration"><div class=CTopic><h3 class=CTitle><a name="fann_train_enum"></a>fann_train_enum</h3><div class=CBody><p>The Training algorithms used when training on <a href="fann_train-h.html#struct_fann_train_data" class=LClass id=link16 onMouseOver="ShowTip(event, 'tt2', 'link16')" onMouseOut="HideTip('tt2')">struct fann_train_data</a> with functions like <a href="fann_train-h.html#fann_train_on_data" class=LFunction id=link17 onMouseOver="ShowTip(event, 'tt3', 'link17')" onMouseOut="HideTip('tt3')">fann_train_on_data</a> or <a href="fann_train-h.html#fann_train_on_file" class=LFunction id=link18 onMouseOver="ShowTip(event, 'tt4', 'link18')" onMouseOut="HideTip('tt4')">fann_train_on_file</a>. The incremental training looks alters the weights after each time it is presented an input pattern, while batch only alters the weights once after it has been presented to all the patterns.</p><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_TRAIN_INCREMENTAL"></a>FANN_TRAIN_INCREMENTAL</td><td class=CDLDescription>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.</td></tr><tr><td class=CDLEntry><a name="FANN_TRAIN_BATCH"></a>FANN_TRAIN_BATCH</td><td class=CDLDescription>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.</td></tr><tr><td class=CDLEntry><a name="FANN_TRAIN_RPROP"></a>FANN_TRAIN_RPROP</td><td class=CDLDescription>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.</td></tr><tr><td class=CDLEntry><a name="FANN_TRAIN_QUICKPROP"></a>FANN_TRAIN_QUICKPROP</td><td class=CDLDescription>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].</td></tr></table><h4 class=CHeading>See also</h4><p><a href="fann_train-h.html#fann_set_training_algorithm" class=LFunction id=link19 onMouseOver="ShowTip(event, 'tt8', 'link19')" onMouseOut="HideTip('tt8')">fann_set_training_algorithm</a>, <a href="fann_train-h.html#fann_get_training_algorithm" class=LFunction id=link20 onMouseOver="ShowTip(event, 'tt9', 'link20')" onMouseOut="HideTip('tt9')">fann_get_training_algorithm</a></p></div></div></div>
<div class="CConstant"><div class=CTopic><h3 class=CTitle><a name="FANN_TRAIN_NAMES"></a>FANN_TRAIN_NAMES</h3><div class=CBody><p>Constant array consisting of the names for the training algorithms, so that the name of an training function can be received by:</p><blockquote><pre class="prettyprint">char *name = FANN_TRAIN_NAMES[train_function];</pre></blockquote><h4 class=CHeading>See Also</h4><p><a href="#fann_train_enum" class=LType id=link21 onMouseOver="ShowTip(event, 'tt10', 'link21')" onMouseOut="HideTip('tt10')">fann_train_enum</a></p></div></div></div>
<div class="CEnumeration"><div class=CTopic><h3 class=CTitle><a name="fann_activationfunc_enum"></a>fann_activationfunc_enum</h3><div class=CBody><p>The activation functions used for the neurons during training. The activation functions can either be defined for a group of neurons by <a href="fann_train-h.html#fann_set_activation_function_hidden" class=LFunction id=link22 onMouseOver="ShowTip(event, 'tt11', 'link22')" onMouseOut="HideTip('tt11')">fann_set_activation_function_hidden</a> and <a href="fann_train-h.html#fann_set_activation_function_output" class=LFunction id=link23 onMouseOver="ShowTip(event, 'tt12', 'link23')" onMouseOut="HideTip('tt12')">fann_set_activation_function_output</a> or it can be defined for a single neuron by <a href="fann_train-h.html#fann_set_activation_function" class=LFunction id=link24 onMouseOver="ShowTip(event, 'tt13', 'link24')" onMouseOut="HideTip('tt13')">fann_set_activation_function</a>.</p><p>The steepness of an activation function is defined in the same way by <a href="fann_train-h.html#fann_set_activation_steepness_hidden" class=LFunction id=link25 onMouseOver="ShowTip(event, 'tt14', 'link25')" onMouseOut="HideTip('tt14')">fann_set_activation_steepness_hidden</a>, <a href="fann_train-h.html#fann_set_activation_steepness_output" class=LFunction id=link26 onMouseOver="ShowTip(event, 'tt15', 'link26')" onMouseOut="HideTip('tt15')">fann_set_activation_steepness_output</a> and <a href="fann_train-h.html#fann_set_activation_steepness" class=LFunction id=link27 onMouseOver="ShowTip(event, 'tt16', 'link27')" onMouseOut="HideTip('tt16')">fann_set_activation_steepness</a>.</p><h4 class=CHeading>The functions are described with functions where</h4><ul><li>x is the input to the activation function,</li><li>y is the output,</li><li>s is the steepness and</li><li>d is the derivation.</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_LINEAR"></a>FANN_LINEAR</td><td class=CDLDescription>Linear activation function.</td></tr></table><ul><li>span: -inf < y < inf</li><li>y = x*s, d = 1*s</li><li>Can NOT be used in fixed point.</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_THRESHOLD"></a>FANN_THRESHOLD</td><td class=CDLDescription>Threshold activation function.</td></tr></table><ul><li>x < 0 -> y = 0, x >= 0 -> y = 1</li><li>Can NOT be used during training.</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_THRESHOLD_SYMMETRIC"></a>FANN_THRESHOLD_SYMMETRIC</td><td class=CDLDescription>Threshold activation function.</td></tr></table><ul><li>x < 0 -> y = 0, x >= 0 -> y = 1</li><li>Can NOT be used during training.</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_SIGMOID"></a>FANN_SIGMOID</td><td class=CDLDescription>Sigmoid activation function.</td></tr></table><ul><li>One of the most used activation functions.</li><li>span: 0 < y < 1</li><li>y = 1/(1 + exp(-2*s*x))</li><li>d = 2*s*y*(1 - y)</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_SIGMOID_STEPWISE"></a>FANN_SIGMOID_STEPWISE</td><td class=CDLDescription>Stepwise linear approximation to sigmoid.</td></tr></table><ul><li>Faster than sigmoid but a bit less precise.</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_SIGMOID_SYMMETRIC"></a>FANN_SIGMOID_SYMMETRIC</td><td class=CDLDescription>Symmetric sigmoid activation function, aka. tanh.</td></tr></table><ul><li>One of the most used activation functions.</li><li>span: -1 < y < 1</li><li>y = tanh(s*x) = 2/(1 + exp(-2*s*x)) - 1</li><li>d = s*(1-(y*y))</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_SIGMOID_SYMMETRIC"></a>FANN_SIGMOID_SYMMETRIC</td><td class=CDLDescription>Stepwise linear approximation to symmetric sigmoid.</td></tr></table><ul><li>Faster than symmetric sigmoid but a bit less precise.</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_GAUSSIAN"></a>FANN_GAUSSIAN</td><td class=CDLDescription>Gaussian activation function.</td></tr></table><ul><li>0 when x = -inf, 1 when x = 0 and 0 when x = inf</li><li>span: 0 < y < 1</li><li>y = exp(-x*s*x*s)</li><li>d = -2*x*s*y*s</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_GAUSSIAN_SYMMETRIC"></a>FANN_GAUSSIAN_SYMMETRIC</td><td class=CDLDescription>Symmetric gaussian activation function.</td></tr></table><ul><li>-1 when x = -inf, 1 when x = 0 and 0 when x = inf</li><li>span: -1 < y < 1</li><li>y = exp(-x*s*x*s)*2-1</li><li>d = -2*x*s*(y+1)*s</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_ELLIOT"></a>FANN_ELLIOT</td><td class=CDLDescription>Fast (sigmoid like) activation function defined by David Elliott</td></tr></table><ul><li>span: 0 < y < 1</li><li>y = ((x*s) / 2) / (1 + |x*s|) + 0.5</li><li>d = s*1/(2*(1+|x*s|)*(1+|x*s|))</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_ELLIOT_SYMMETRIC"></a>FANN_ELLIOT_SYMMETRIC</td><td class=CDLDescription>Fast (symmetric sigmoid like) activation function defined by David Elliott</td></tr></table><ul><li>span: -1 < y < 1</li><li>y = (x*s) / (1 + |x*s|)</li><li>d = s*1/((1+|x*s|)*(1+|x*s|))</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_LINEAR_PIECE"></a>FANN_LINEAR_PIECE</td><td class=CDLDescription>Bounded linear activation function.</td></tr></table><ul><li>span: 0 <= y <= 1</li><li>y = x*s, d = 1*s</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_LINEAR_PIECE_SYMMETRIC"></a>FANN_LINEAR_PIECE_SYMMETRIC</td><td class=CDLDescription>Bounded linear activation function.</td></tr></table><ul><li>span: -1 <= y <= 1</li><li>y = x*s, d = 1*s</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_SIN_SYMMETRIC"></a>FANN_SIN_SYMMETRIC</td><td class=CDLDescription>Periodical sinus activation function.</td></tr></table><ul><li>span: -1 <= y <= 1</li><li>y = sin(x*s)</li><li>d = s*cos(x*s)</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_COS_SYMMETRIC"></a>FANN_COS_SYMMETRIC</td><td class=CDLDescription>Periodical cosinus activation function.</td></tr></table><ul><li>span: -1 <= y <= 1</li><li>y = cos(x*s)</li><li>d = s*-sin(x*s)</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_SIN"></a>FANN_SIN</td><td class=CDLDescription>Periodical sinus activation function.</td></tr></table><ul><li>span: 0 <= y <= 1</li><li>y = sin(x*s)/2+0.5</li><li>d = s*cos(x*s)/2</li></ul><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_COS"></a>FANN_COS</td><td class=CDLDescription>Periodical cosinus activation function.</td></tr></table><ul><li>span: 0 <= y <= 1</li><li>y = cos(x*s)/2+0.5</li><li>d = s*-sin(x*s)/2</li></ul><h4 class=CHeading>See also</h4><p><a href="fann_train-h.html#fann_set_activation_function_layer" class=LFunction id=link28 onMouseOver="ShowTip(event, 'tt17', 'link28')" onMouseOut="HideTip('tt17')">fann_set_activation_function_layer</a>, <a href="fann_train-h.html#fann_set_activation_function_hidden" class=LFunction id=link29 onMouseOver="ShowTip(event, 'tt11', 'link29')" onMouseOut="HideTip('tt11')">fann_set_activation_function_hidden</a>, <a href="fann_train-h.html#fann_set_activation_function_output" class=LFunction id=link30 onMouseOver="ShowTip(event, 'tt12', 'link30')" onMouseOut="HideTip('tt12')">fann_set_activation_function_output</a>, <a href="fann_train-h.html#fann_set_activation_steepness" class=LFunction id=link31 onMouseOver="ShowTip(event, 'tt16', 'link31')" onMouseOut="HideTip('tt16')">fann_set_activation_steepness</a>, <a href="fann_train-h.html#fann_set_activation_function" class=LFunction id=link32 onMouseOver="ShowTip(event, 'tt13', 'link32')" onMouseOut="HideTip('tt13')">fann_set_activation_function</a></p></div></div></div>
<div class="CConstant"><div class=CTopic><h3 class=CTitle><a name="FANN_ACTIVATIONFUNC_NAMES"></a>FANN_ACTIVATIONFUNC_NAMES</h3><div class=CBody><p>Constant array consisting of the names for the activation function, so that the name of an activation function can be received by:</p><blockquote><pre class="prettyprint">char *name = FANN_ACTIVATIONFUNC_NAMES[activation_function];</pre></blockquote><h4 class=CHeading>See Also</h4><p><a href="#fann_activationfunc_enum" class=LType id=link33 onMouseOver="ShowTip(event, 'tt18', 'link33')" onMouseOut="HideTip('tt18')">fann_activationfunc_enum</a></p></div></div></div>
<div class="CEnumeration"><div class=CTopic><h3 class=CTitle><a name="fann_errorfunc_enum"></a>fann_errorfunc_enum</h3><div class=CBody><p>Error function used during training.</p><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_ERRORFUNC_LINEAR"></a>FANN_ERRORFUNC_LINEAR</td><td class=CDLDescription>Standard linear error function.</td></tr><tr><td class=CDLEntry><a name="FANN_ERRORFUNC_TANH"></a>FANN_ERRORFUNC_TANH</td><td class=CDLDescription>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.</td></tr></table><h4 class=CHeading>See also</h4><p><a href="fann_train-h.html#fann_set_train_error_function" class=LFunction id=link34 onMouseOver="ShowTip(event, 'tt19', 'link34')" onMouseOut="HideTip('tt19')">fann_set_train_error_function</a>, <a href="fann_train-h.html#fann_get_train_error_function" class=LFunction id=link35 onMouseOver="ShowTip(event, 'tt20', 'link35')" onMouseOut="HideTip('tt20')">fann_get_train_error_function</a></p></div></div></div>
<div class="CConstant"><div class=CTopic><h3 class=CTitle><a name="FANN_ERRORFUNC_NAMES"></a>FANN_ERRORFUNC_NAMES</h3><div class=CBody><p>Constant array consisting of the names for the training error functions, so that the name of an error function can be received by:</p><blockquote><pre class="prettyprint">char *name = FANN_ERRORFUNC_NAMES[error_function];</pre></blockquote><h4 class=CHeading>See Also</h4><p><a href="#fann_errorfunc_enum" class=LType id=link36 onMouseOver="ShowTip(event, 'tt21', 'link36')" onMouseOut="HideTip('tt21')">fann_errorfunc_enum</a></p></div></div></div>
<div class="CEnumeration"><div class=CTopic><h3 class=CTitle><a name="fann_stopfunc_enum"></a>fann_stopfunc_enum</h3><div class=CBody><p>Stop criteria used during training.</p><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_STOPFUNC_MSE"></a>FANN_STOPFUNC_MSE</td><td class=CDLDescription>Stop criteria is Mean Square Error (MSE) value.</td></tr><tr><td class=CDLEntry><a name="FANN_STOPFUNC_BIT"></a>FANN_STOPFUNC_BIT</td><td class=CDLDescription>Stop criteria is number of bits that fail. The number of bits; means the number of output neurons which differ more than the bit fail limit (see <a href="fann_train-h.html#fann_get_bit_fail_limit" class=LFunction id=link37 onMouseOver="ShowTip(event, 'tt22', 'link37')" onMouseOut="HideTip('tt22')">fann_get_bit_fail_limit</a>, <a href="fann_train-h.html#fann_set_bit_fail_limit" class=LFunction id=link38 onMouseOver="ShowTip(event, 'tt23', 'link38')" onMouseOut="HideTip('tt23')">fann_set_bit_fail_limit</a>). The bits are counted in all of the training data, so this number can be higher than the number of training data.</td></tr></table><h4 class=CHeading>See also</h4><p><a href="fann_train-h.html#fann_set_train_stop_function" class=LFunction id=link39 onMouseOver="ShowTip(event, 'tt24', 'link39')" onMouseOut="HideTip('tt24')">fann_set_train_stop_function</a>, <a href="fann_train-h.html#fann_get_train_stop_function" class=LFunction id=link40 onMouseOver="ShowTip(event, 'tt25', 'link40')" onMouseOut="HideTip('tt25')">fann_get_train_stop_function</a></p></div></div></div>
<div class="CConstant"><div class=CTopic><h3 class=CTitle><a name="FANN_STOPFUNC_NAMES"></a>FANN_STOPFUNC_NAMES</h3><div class=CBody><p>Constant array consisting of the names for the training stop functions, so that the name of a stop function can be received by:</p><blockquote><pre class="prettyprint">char *name = FANN_STOPFUNC_NAMES[stop_function];</pre></blockquote><h4 class=CHeading>See Also</h4><p><a href="#fann_stopfunc_enum" class=LType id=link41 onMouseOver="ShowTip(event, 'tt26', 'link41')" onMouseOut="HideTip('tt26')">fann_stopfunc_enum</a></p></div></div></div>
<div class="CEnumeration"><div class=CTopic><h3 class=CTitle><a name="fann_network_type_enum"></a>fann_network_type_enum</h3><div class=CBody><p>Definition of network types used by <a href="fann-h.html#fann_get_network_type" class=LFunction id=link42 onMouseOver="ShowTip(event, 'tt5', 'link42')" onMouseOut="HideTip('tt5')">fann_get_network_type</a></p><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry><a name="FANN_NETTYPE_LAYER"></a>FANN_NETTYPE_LAYER</td><td class=CDLDescription>Each layer only has connections to the next layer</td></tr><tr><td class=CDLEntry><a name="FANN_NETTYPE_SHORTCUT"></a>FANN_NETTYPE_SHORTCUT</td><td class=CDLDescription>Each layer has connections to all following layers</td></tr></table><h4 class=CHeading>See Also</h4><p><a href="fann-h.html#fann_get_network_type" class=LFunction id=link43 onMouseOver="ShowTip(event, 'tt5', 'link43')" onMouseOut="HideTip('tt5')">fann_get_network_type</a></p><p>This enumeration appears in FANN >= 2.1.0</p></div></div></div>
<div class="CConstant"><div class=CTopic><h3 class=CTitle><a name="FANN_NETWORK_TYPE_NAMES"></a>FANN_NETWORK_TYPE_NAMES</h3><div class=CBody><p>Constant array consisting of the names for the network types, so that the name of an network type can be received by:</p><blockquote><pre class="prettyprint">char *network_type_name = FANN_NETWORK_TYPE_NAMES[fann_get_network_type(ann)];</pre></blockquote><h4 class=CHeading>See Also</h4><p><a href="fann-h.html#fann_get_network_type" class=LFunction id=link44 onMouseOver="ShowTip(event, 'tt5', 'link44')" onMouseOut="HideTip('tt5')">fann_get_network_type</a></p><p>This constant appears in FANN >= 2.1.0</p></div></div></div>
<div class="CGroup"><div class=CTopic><h3 class=CTitle><a name="Types"></a>Types</h3></div></div>
<div class="CType"><div class=CTopic><h3 class=CTitle><a name="fann_callback_type"></a>fann_callback_type</h3><div class=CBody><p>This callback function can be called during training when using <a href="fann_train-h.html#fann_train_on_data" class=LFunction id=link45 onMouseOver="ShowTip(event, 'tt3', 'link45')" onMouseOut="HideTip('tt3')">fann_train_on_data</a>, <a href="fann_train-h.html#fann_train_on_file" class=LFunction id=link46 onMouseOver="ShowTip(event, 'tt4', 'link46')" onMouseOut="HideTip('tt4')">fann_train_on_file</a> or <a href="fann_cascade-h.html#fann_cascadetrain_on_data" class=LFunction id=link47 onMouseOver="ShowTip(event, 'tt6', 'link47')" onMouseOut="HideTip('tt6')">fann_cascadetrain_on_data</a>.</p><blockquote><pre>typedef int (FANN_API * fann_callback_type) (struct fann *ann, struct fann_train_data *train,
unsigned int max_epochs,
unsigned int epochs_between_reports,
float desired_error, unsigned int epochs);</pre></blockquote><p>The callback can be set by using <a href="fann_train-h.html#fann_set_callback" class=LFunction id=link48 onMouseOver="ShowTip(event, 'tt27', 'link48')" onMouseOut="HideTip('tt27')">fann_set_callback</a> and is very usefull for doing custom things during training. It is recommended to use this function when implementing custom training procedures, or when visualizing the training in a GUI etc. The parameters which the callback function takes is the parameters given to the <a href="fann_train-h.html#fann_train_on_data" class=LFunction id=link49 onMouseOver="ShowTip(event, 'tt3', 'link49')" onMouseOut="HideTip('tt3')">fann_train_on_data</a>, plus an epochs parameter which tells how many epochs the training have taken so far.</p><p>The callback function should return an integer, if the callback function returns -1, the training will terminate.</p><h4 class=CHeading>Example of a callback function</h4><blockquote><pre>int FANN_API test_callback(struct fann *ann, struct fann_train_data *train,
unsigned int max_epochs, unsigned int epochs_between_reports,
float desired_error, unsigned int epochs)
{
printf("Epochs %8d. MSE: %.5f. Desired-MSE: %.5f\n", epochs, fann_get_MSE(ann), desired_error);
return 0;
}</pre></blockquote><h4 class=CHeading>See also</h4><p><a href="fann_train-h.html#fann_set_callback" class=LFunction id=link50 onMouseOver="ShowTip(event, 'tt27', 'link50')" onMouseOut="HideTip('tt27')">fann_set_callback</a>, <a href="fann_train-h.html#fann_train_on_data" class=LFunction id=link51 onMouseOver="ShowTip(event, 'tt3', 'link51')" onMouseOut="HideTip('tt3')">fann_train_on_data</a></p></div></div></div>
<div class="CClass"><div class=CTopic><h2 class=CTitle><a name="struct_fann_error"></a>struct fann_error</h2><div class=CBody><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td class="prettyprint">struct fann_error</td></tr></table></blockquote><p>Structure used to store error-related information, both <a href="#struct_fann" class=LClass id=link52 onMouseOver="ShowTip(event, 'tt1', 'link52')" onMouseOut="HideTip('tt1')">struct fann</a> and <a href="fann_train-h.html#struct_fann_train_data" class=LClass id=link53 onMouseOver="ShowTip(event, 'tt2', 'link53')" onMouseOut="HideTip('tt2')">struct fann_train_data</a> can be casted to this type.</p><h4 class=CHeading>See also</h4><p><a href="fann_error-h.html#fann_set_error_log" class=LFunction id=link54 onMouseOver="ShowTip(event, 'tt28', 'link54')" onMouseOut="HideTip('tt28')">fann_set_error_log</a>, <a href="fann_error-h.html#fann_get_errno" class=LFunction id=link55 onMouseOver="ShowTip(event, 'tt29', 'link55')" onMouseOut="HideTip('tt29')">fann_get_errno</a></p></div></div></div>
<div class="CClass"><div class=CTopic><h2 class=CTitle><a name="struct_fann"></a>struct fann</h2><div class=CBody><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td class="prettyprint">struct fann</td></tr></table></blockquote><p>The fast artificial neural network(fann) structure.</p><p>Data within this structure should never be accessed directly, but only by using the <b>fann_get_...</b> and <b>fann_set_...</b> functions.</p><p>The fann structure is created using one of the <b>fann_create_...</b> functions and each of the functions which operates on the structure takes <b>struct fann * ann</b> as the first parameter.</p><h4 class=CHeading>See also</h4><p><a href="fann-h.html#fann_create_standard" class=LFunction id=link56 onMouseOver="ShowTip(event, 'tt30', 'link56')" onMouseOut="HideTip('tt30')">fann_create_standard</a>, <a href="fann-h.html#fann_destroy" class=LFunction id=link57 onMouseOver="ShowTip(event, 'tt31', 'link57')" onMouseOut="HideTip('tt31')">fann_destroy</a></p><!--START_ND_SUMMARY--><div class=Summary><div class=STitle>Summary</div><div class=SBorder><table border=0 cellspacing=0 cellpadding=0 class=STable><tr class="SGroup"><td class=SEntry><a href="#struct_fann.Types" >Types</a></td><td class=SDescription></td></tr><tr class="SType SIndent1 SMarked"><td class=SEntry><a href="#struct_fann.fann_connection" >fann_connection</a></td><td class=SDescription>Describes a connection between two neurons and its weight</td></tr></table></div></div><!--END_ND_SUMMARY--></div></div></div>
<div class="CGroup"><div class=CTopic><h3 class=CTitle><a name="struct_fann.Types"></a>Types</h3></div></div>
<div class="CType"><div class=CTopic><h3 class=CTitle><a name="struct_fann.fann_connection"></a>fann_connection</h3><div class=CBody><p>Describes a connection between two neurons and its weight</p><table border=0 cellspacing=0 cellpadding=0 class=CDescriptionList><tr><td class=CDLEntry>from_neuron</td><td class=CDLDescription>Unique number used to identify source neuron</td></tr><tr><td class=CDLEntry>to_neuron</td><td class=CDLDescription>Unique number used to identify destination neuron</td></tr><tr><td class=CDLEntry>weight</td><td class=CDLDescription>The numerical value of the weight</td></tr></table><h4 class=CHeading>See Also</h4><p><a href="fann-h.html#fann_get_connection_array" class=LFunction id=link58 onMouseOver="ShowTip(event, 'tt32', 'link58')" onMouseOut="HideTip('tt32')">fann_get_connection_array</a>, <a href="fann-h.html#fann_set_weight_array" class=LFunction id=link59 onMouseOver="ShowTip(event, 'tt33', 'link59')" onMouseOut="HideTip('tt33')">fann_set_weight_array</a></p><p>This structure appears in FANN >= 2.1.0</p></div></div></div>
</div><!--Content-->
<div id=Footer><a href="http://www.naturaldocs.org">Generated by Natural Docs</a></div><!--Footer-->
<div id=Menu><div class=MEntry><div class=MFile><a href="fann_cascade-h.html">FANN Cascade Training</a></div></div><div class=MEntry><div class=MFile><a href="fann-h.html">FANN Creation/<wbr>Execution</a></div></div><div class=MEntry><div class=MFile id=MSelected>FANN Datatypes</div></div><div class=MEntry><div class=MFile><a href="fann_error-h.html">FANN Error Handling</a></div></div><div class=MEntry><div class=MFile><a href="fann_io-h.html">FANN File Input/<wbr>Output</a></div></div><div class=MEntry><div class=MFile><a href="fann_train-h.html">FANN Training</a></div></div><div class=MEntry><div class=MFile><a href="fann_cpp-h.html">FANN Wrapper for C++</a></div></div><div class=MEntry><div class=MGroup><a href="javascript:ToggleMenu('MGroupContent1')">Index</a><div class=MGroupContent id=MGroupContent1><div class=MEntry><div class=MIndex><a href="../index/General.html">Everything</a></div></div><div class=MEntry><div class=MIndex><a href="../index/Classes.html">Classes</a></div></div><div class=MEntry><div class=MIndex><a href="../index/Constants.html">Constants</a></div></div><div class=MEntry><div class=MIndex><a href="../index/Functions.html">Functions</a></div></div><div class=MEntry><div class=MIndex><a href="../index/Types.html">Types</a></div></div></div></div></div><script type="text/javascript"><!--
var searchPanel = new SearchPanel("searchPanel", "HTML", "../search");
--></script><div id=MSearchPanel class=MSearchPanelInactive><input type=text id=MSearchField value=Search onFocus="searchPanel.OnSearchFieldFocus(true)" onBlur="searchPanel.OnSearchFieldFocus(false)" onKeyUp="searchPanel.OnSearchFieldChange()"><select id=MSearchType onFocus="searchPanel.OnSearchTypeFocus(true)" onBlur="searchPanel.OnSearchTypeFocus(false)" onChange="searchPanel.OnSearchTypeChange()"><option id=MSearchEverything selected value="General">Everything</option><option value="Classes">Classes</option><option value="Constants">Constants</option><option value="Functions">Functions</option><option value="Types">Types</option></select></div></div><!--Menu-->
<!--START_ND_TOOLTIPS-->
<div class=CToolTip id="tt1"><div class=CClass><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td class="prettyprint">struct fann</td></tr></table></blockquote>The fast artificial neural network(fann) structure.</div></div><div class=CToolTip id="tt2"><div class=CClass><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td class="prettyprint">struct fann_train_data</td></tr></table></blockquote>Structure used to store data, for use with training.</div></div><div class=CToolTip id="tt3"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_train_on_data(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann_train_data </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>data,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>unsigned </td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>max_epochs,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>unsigned </td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>epochs_between_reports,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>float </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>desired_error</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Trains on an entire dataset, for a period of time.</div></div><div class=CToolTip id="tt4"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_train_on_file(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>const </td><td class="PType prettyprint " nowrap>char </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>filename,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>unsigned </td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>max_epochs,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>unsigned </td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>epochs_between_reports,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>float </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>desired_error</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Does the same as fann_train_on_data, but reads the training data directly from a file.</div></div><div class=CToolTip id="tt5"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL enum fann_nettype_enum FANN_API fann_get_network_type(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Get the type of neural network it was created as.</div></div><div class=CToolTip id="tt6"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_cascadetrain_on_data(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann_train_data </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>data,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>unsigned </td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>max_neurons,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>unsigned </td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>neurons_between_reports,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>float </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>desired_error</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Trains on an entire dataset, for a period of time using the Cascade2 training algorithm. </div></div><div class=CToolTip id="tt7"><div class=CClass><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td class="prettyprint">struct fann_error</td></tr></table></blockquote>Structure used to store error-related information, both struct fann and struct fann_train_data can be casted to this type.</div></div><div class=CToolTip id="tt8"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_set_training_algorithm(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>enum </td><td class="PType prettyprint " nowrap>fann_train_enum </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>training_algorithm</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Set the training algorithm.</div></div><div class=CToolTip id="tt9"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL enum fann_train_enum FANN_API fann_get_training_algorithm(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Return the training algorithm as described by fann_train_enum. </div></div><div class=CToolTip id="tt10"><div class=CType>The Training algorithms used when training on struct fann_train_data with functions like fann_train_on_data or fann_train_on_file. </div></div><div class=CToolTip id="tt11"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_set_activation_function_hidden(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>enum </td><td class="PType prettyprint " nowrap>fann_activationfunc_enum </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>activation_function</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Set the activation function for all of the hidden layers.</div></div><div class=CToolTip id="tt12"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_set_activation_function_output(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>enum </td><td class="PType prettyprint " nowrap>fann_activationfunc_enum </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>activation_function</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Set the activation function for the output layer.</div></div><div class=CToolTip id="tt13"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_set_activation_function(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>enum </td><td class="PType prettyprint " nowrap>fann_activationfunc_enum </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>activation_function,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>layer,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>neuron</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Set the activation function for neuron number <b>neuron</b> in layer number <b>layer</b>, counting the input layer as layer 0.</div></div><div class=CToolTip id="tt14"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_set_activation_steepness_hidden(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>fann_type </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>steepness</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Set the steepness of the activation steepness in all of the hidden layers.</div></div><div class=CToolTip id="tt15"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_set_activation_steepness_output(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>fann_type </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>steepness</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Set the steepness of the activation steepness in the output layer.</div></div><div class=CToolTip id="tt16"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "nowrap>FANN_EXTERNAL void FANN_API fann_set_activation_steepness(</td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap>ann,</td></tr><tr><td></td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>fann_type </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap>steepness,</td></tr><tr><td></td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap>layer,</td></tr><tr><td></td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap>neuron</td><td class="PAfterParameters prettyprint "nowrap>)</td></tr></table></td></tr></table></blockquote>Set the activation steepness for neuron number <b>neuron</b> in layer number <b>layer</b>, counting the input layer as layer 0.</div></div><div class=CToolTip id="tt17"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_set_activation_function_layer(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>enum </td><td class="PType prettyprint " nowrap>fann_activationfunc_enum </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>activation_function,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>layer</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Set the activation function for all the neurons in the layer number <b>layer</b>, counting the input layer as layer 0.</div></div><div class=CToolTip id="tt18"><div class=CType>The activation functions used for the neurons during training. </div></div><div class=CToolTip id="tt19"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_set_train_error_function(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>enum </td><td class="PType prettyprint " nowrap>fann_errorfunc_enum </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>train_error_function</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Set the error function used during training.</div></div><div class=CToolTip id="tt20"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL enum fann_errorfunc_enum FANN_API fann_get_train_error_function(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Returns the error function used during training.</div></div><div class=CToolTip id="tt21"><div class=CType>Error function used during training.</div></div><div class=CToolTip id="tt22"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "nowrap>FANN_EXTERNAL fann_type FANN_API fann_get_bit_fail_limit(</td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap>ann</td><td class="PAfterParameters prettyprint "nowrap>)</td></tr></table></td></tr></table></blockquote>Returns the bit fail limit used during training.</div></div><div class=CToolTip id="tt23"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "nowrap>FANN_EXTERNAL void FANN_API fann_set_bit_fail_limit(</td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap>ann,</td></tr><tr><td></td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>fann_type </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap>bit_fail_limit</td><td class="PAfterParameters prettyprint "nowrap>)</td></tr></table></td></tr></table></blockquote>Set the bit fail limit used during training.</div></div><div class=CToolTip id="tt24"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_set_train_stop_function(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>enum </td><td class="PType prettyprint " nowrap>fann_stopfunc_enum </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>train_stop_function</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Set the stop function used during training.</div></div><div class=CToolTip id="tt25"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL enum fann_stopfunc_enum FANN_API fann_get_train_stop_function(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Returns the the stop function used during training.</div></div><div class=CToolTip id="tt26"><div class=CType>Stop criteria used during training.</div></div><div class=CToolTip id="tt27"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "nowrap>FANN_EXTERNAL void FANN_API fann_set_callback(</td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap>ann,</td></tr><tr><td></td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>fann_callback_type </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap>callback</td><td class="PAfterParameters prettyprint "nowrap>)</td></tr></table></td></tr></table></blockquote>Sets the callback function for use during training.</div></div><div class=CToolTip id="tt28"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "nowrap>FANN_EXTERNAL void FANN_API fann_set_error_log(</td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann_error </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap>errdat,</td></tr><tr><td></td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap>FILE </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap>log_file</td><td class="PAfterParameters prettyprint "nowrap>)</td></tr></table></td></tr></table></blockquote>Change where errors are logged to. </div></div><div class=CToolTip id="tt29"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL enum fann_errno_enum FANN_API fann_get_errno(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann_error </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>errdat</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Returns the last error number.</div></div><div class=CToolTip id="tt30"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=4>FANN_EXTERNAL struct fann *FANN_API fann_create_standard(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>unsigned </td><td class="PType prettyprint " nowrap>int </td><td class="PParameter prettyprint " nowrap width=100%>num_layers,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap></td><td class="PType prettyprint " nowrap> </td><td class="PParameter prettyprint " nowrap width=100%>...</td></tr><tr><td class="PAfterParameters prettyprint "colspan=4>)</td></tr></table></td></tr></table></blockquote>Creates a standard fully connected backpropagation neural network.</div></div><div class=CToolTip id="tt31"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "nowrap>FANN_EXTERNAL void FANN_API fann_destroy(</td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap>ann</td><td class="PAfterParameters prettyprint "nowrap>)</td></tr></table></td></tr></table></blockquote>Destroys the entire network and properly freeing all the associated memmory.</div></div><div class=CToolTip id="tt32"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_get_connection_array(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann_connection </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>connections</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Get the connections in the network.</div></div><div class=CToolTip id="tt33"><div class=CFunction><blockquote><table border=0 cellspacing=0 cellpadding=0 class="Prototype"><tr><td><table border=0 cellspacing=0 cellpadding=0><tr><td class="PBeforeParameters prettyprint "colspan=5>FANN_EXTERNAL void FANN_API fann_set_weight_array(</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>ann,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>struct </td><td class="PType prettyprint " nowrap>fann_connection </td><td class="PParameterPrefix prettyprint " nowrap>*</td><td class="PParameter prettyprint " nowrap width=100%>connections,</td></tr><tr><td> </td><td class="PTypePrefix prettyprint " nowrap>unsigned </td><td class="PType prettyprint " nowrap>int </td><td class="PParameterPrefix prettyprint " nowrap></td><td class="PParameter prettyprint " nowrap width=100%>num_connections</td></tr><tr><td class="PAfterParameters prettyprint "colspan=5>)</td></tr></table></td></tr></table></blockquote>Set connections in the network.</div></div><!--END_ND_TOOLTIPS-->
<div id=MSearchResultsWindow><iframe src="" frameborder=0 name=MSearchResults id=MSearchResults></iframe><a href="javascript:searchPanel.CloseResultsWindow()" id=MSearchResultsWindowClose>Close</a></div>
<script language=JavaScript><!--
if (browserType) {if (browserVer) {document.write("</div>"); }document.write("</div>");}// --></script></body></html>
|