Exemplo n.º 1
0
        private void numNeurons_ValueChanged(object sender, EventArgs e)
        {
            int n = (int)numNeurons.Value;

            if (_Layer.NumberOfNeurons != n)
            {
                _Layer.NumberOfNeurons = n;
                panelNeurons.Controls.Clear();
                Point p = new Point(50, 0);
                for (int i = 0; i < _Layer.Neurons.Count; i++)
                {
                    NeuronControl nc = new NeuronControl();
                    nc.Neuron = _Layer.Neurons[i];
                    nc.NeuronControlClicked = new NeuronControl.NeuronControlClickedDelegate(SetCurrentNeuron);
                    panelNeurons.Controls.Add(nc);
                    nc.Location = p;
                    p.Y        += 55;
                }
            }
        }
Exemplo n.º 2
0
 void SetCurrentNeuron(NeuronControl neuron_control)
 {
     if (!txtNeuronThreshold.Enabled)
     {
         txtNeuronThreshold.Enabled = true;
     }
     if (!btnRandThreshold.Enabled)
     {
         btnRandThreshold.Enabled = true;
     }
     foreach (NeuronControl nc in panelNeurons.Controls)
     {
         if (!nc.Equals(neuron_control))
         {
             nc.Selected = false;
         }
     }
     neuron_control.Selected  = true;
     SelectedNeuron           = neuron_control.Neuron;
     txtNeuronThreshold.Value = (decimal)SelectedNeuron.Threshold;
 }
Exemplo n.º 3
0
        void SetUI()
        {
            if (_Layer != null)
            {
                prenn = _Layer.NumberOfNeurons;
                this.labLayerType.Text = "Layer type: " + _Layer.LayerType.ToString() + " layer";
                numNeurons.Value       = _Layer.NumberOfNeurons;
                Point p = new Point(50, 0);
                for (int i = 0; i < _Layer.Neurons.Count; i++)
                {
                    NeuronControl nc = new NeuronControl();
                    nc.Neuron = _Layer.Neurons[i];
                    nc.NeuronControlClicked = new NeuronControl.NeuronControlClickedDelegate(SetCurrentNeuron);
                    panelNeurons.Controls.Add(nc);
                    nc.Location = p;
                    p.Y        += 55;
                }
                switch (_Layer.InputCombinationFunction)
                {
                case InputCombinationFunctionType.LinearProduct:
                    radioCFLP.Checked = true;
                    labCFHelp.Text    = "The net input is the sum of the linear products of outputs of the neurons in the previous layer and weights of the pre synapses.";
                    break;

                case InputCombinationFunctionType.EucledianDistance:
                    radCFED.Checked = true;
                    labCFHelp.Text  = "The net input is the sum of eucledian distance between the synapses weights and outputs of the previous layers neuron.";
                    break;

                case InputCombinationFunctionType.ManhattanDistance:
                    radCFMD.Checked = true;
                    labCFHelp.Text  = "The net input is the sum of manhattan distance between the synapses weights and outputs of the previous layers neuron.";
                    break;
                }
                if (_Layer.ActivationFunction != null)
                {
                    switch (_Layer.ActivationFunction.FunctionName)
                    {
                    case "Unipolar Sign Function":
                        USF = (UnipolarSignFunction)_Layer.ActivationFunction;
                        radAFUSF.Checked = true;
                        break;

                    case "Bipolar Sign Function":
                        BSF = (BipolarSignFunction)_Layer.ActivationFunction;
                        radAFBSF.Checked = true;
                        break;

                    case "Unipolar Linear Function":
                        ULF = (UnipolarLinearFunction)_Layer.ActivationFunction;
                        radAFULF.Checked = true;
                        break;

                    case "Bipolar Linear Function":
                        BLF = (BipolarLinearFunction)_Layer.ActivationFunction;
                        radAFBLF.Checked = true;
                        break;

                    case "Sigmoidal Unipolar Function":
                        SUF = (SigmoidalUnipolarFunction)_Layer.ActivationFunction;
                        radAFSUF.Checked = true;
                        break;

                    case "Sigmoidal Bipolar Function":
                        SBF = (SigmoidalBipolarFunction)_Layer.ActivationFunction;
                        radAFSBF.Checked = true;
                        break;

                    case "Tangent Hyperbolic Function":
                        THF = (TangentHyperbolicFunction)_Layer.ActivationFunction;
                        radAFTHF.Checked = true;
                        break;

                    case "Gaussian Activation Function":
                        GAF = (GaussianActivationFunction)_Layer.ActivationFunction;
                        radAFGAF.Checked = true;
                        break;
                    }
                    SetAFParams();
                }
                else
                {
                    radAFNone.Checked = true;
                }
            }
        }
Exemplo n.º 4
0
 void SetCurrentNeuron(NeuronControl neuron_control)
 {
     if (!txtNeuronThreshold.Enabled) txtNeuronThreshold.Enabled = true;
     if (!btnRandThreshold.Enabled) btnRandThreshold.Enabled = true;
     foreach (NeuronControl nc in panelNeurons.Controls)
     {
         if (!nc.Equals(neuron_control))
         {
             nc.Selected = false;
         }
     }
     neuron_control.Selected = true;
     SelectedNeuron = neuron_control.Neuron;
     txtNeuronThreshold.Value = (decimal)SelectedNeuron.Threshold;
 }
Exemplo n.º 5
0
 private void numNeurons_ValueChanged(object sender, EventArgs e)
 {
     int n = (int)numNeurons.Value;
     if (_Layer.NumberOfNeurons != n)
     {
         _Layer.NumberOfNeurons = n;
         panelNeurons.Controls.Clear();
         Point p = new Point(50, 0);
         for (int i = 0; i < _Layer.Neurons.Count; i++)
         {
             NeuronControl nc = new NeuronControl();
             nc.Neuron = _Layer.Neurons[i];
             nc.NeuronControlClicked = new NeuronControl.NeuronControlClickedDelegate(SetCurrentNeuron);
             panelNeurons.Controls.Add(nc);
             nc.Location = p;
             p.Y += 55;
         }
     }
 }
Exemplo n.º 6
0
 void SetUI()
 {
     if (_Layer != null)
     {
         prenn = _Layer.NumberOfNeurons;
         this.labLayerType.Text = "Layer type: " + _Layer.LayerType.ToString() + " layer";
         numNeurons.Value = _Layer.NumberOfNeurons;
         Point p = new Point(50, 0);
         for (int i = 0; i < _Layer.Neurons.Count; i++)
         {
             NeuronControl nc = new NeuronControl();
             nc.Neuron = _Layer.Neurons[i];
             nc.NeuronControlClicked = new NeuronControl.NeuronControlClickedDelegate(SetCurrentNeuron);
             panelNeurons.Controls.Add(nc);
             nc.Location = p;
             p.Y += 55;
         }               
         switch (_Layer.InputCombinationFunction)
         {
             case InputCombinationFunctionType.LinearProduct:
                 radioCFLP.Checked = true;
                 labCFHelp.Text = "The net input is the sum of the linear products of outputs of the neurons in the previous layer and weights of the pre synapses.";
                 break;
             case InputCombinationFunctionType.EucledianDistance:
                 radCFED.Checked = true;
                 labCFHelp.Text = "The net input is the sum of eucledian distance between the synapses weights and outputs of the previous layers neuron.";
                 break;
             case InputCombinationFunctionType.ManhattanDistance:
                 radCFMD.Checked = true;
                 labCFHelp.Text = "The net input is the sum of manhattan distance between the synapses weights and outputs of the previous layers neuron.";
                 break;
         }
         if (_Layer.ActivationFunction != null)
         {
             switch (_Layer.ActivationFunction.FunctionName)
             {
                 case "Unipolar Sign Function":
                     USF = (UnipolarSignFunction)_Layer.ActivationFunction;
                     radAFUSF.Checked = true;
                     break;
                 case "Bipolar Sign Function":
                     BSF = (BipolarSignFunction)_Layer.ActivationFunction;
                     radAFBSF.Checked = true;
                     break;
                 case "Unipolar Linear Function":
                     ULF = (UnipolarLinearFunction)_Layer.ActivationFunction;
                     radAFULF.Checked = true;
                     break;
                 case "Bipolar Linear Function":
                     BLF = (BipolarLinearFunction)_Layer.ActivationFunction;
                     radAFBLF.Checked = true;
                     break;
                 case "Sigmoidal Unipolar Function":
                     SUF = (SigmoidalUnipolarFunction)_Layer.ActivationFunction;
                     radAFSUF.Checked = true;
                     break;
                 case "Sigmoidal Bipolar Function":
                     SBF = (SigmoidalBipolarFunction)_Layer.ActivationFunction;
                     radAFSBF.Checked = true;
                     break;
                 case "Tangent Hyperbolic Function":
                     THF = (TangentHyperbolicFunction)_Layer.ActivationFunction;
                     radAFTHF.Checked = true;
                     break;
                 case "Gaussian Activation Function":
                     GAF = (GaussianActivationFunction)_Layer.ActivationFunction;
                     radAFGAF.Checked = true;
                     break;
             }
             SetAFParams();
         }
         else
         {
             radAFNone.Checked = true;
         }
        
     }
 }