コード例 #1
0
        public NeuralNetworkIndividual(
            Random random,
            int inputNodes,
            int outputNodes,
            int maxLayers,
            int maxInnerNodes)
        {
            this.random            = random;
            this.inputNodesEnabled = new int[inputNodes];

            for (int i = 0; i < inputNodes; i++)
            {
                if (random.NextDouble() > 0.5)
                {
                    this.inputNodesEnabled[i] = 1;
                    this.inputNodes++;
                }
            }

            this.maxInnerNodes = maxInnerNodes;

            var innerLayers = random.Next(maxLayers);

            layers = new int[innerLayers + 2];

            layers[0] = this.inputNodes;
            for (int i = 1; i < innerLayers + 1; i++)
            {
                layers[i] = random.Next(maxInnerNodes);
            }
            layers[innerLayers + 1] = outputNodes;

            Network = new MLPNeuralNetwork(layers);
        }
コード例 #2
0
        public NeuralNetworkIndividual(
            Random random,
            int inputNodes,
            int maxInnerNodes,
            int[] layers)
        {
            this.random        = random;
            this.inputNodes    = inputNodes;
            this.maxInnerNodes = maxInnerNodes;
            this.layers        = layers;

            Network = new MLPNeuralNetwork(layers);
        }
コード例 #3
0
        public static MLPNeuralNetwork GetTrainedNetwork(
            List <float[]> inputData,
            List <float[]> expectedOutputData,
            Action <int, int> reportProgress)
        {
            if (!inputData.Any() || !expectedOutputData.Any())
            {
                throw new Exception("Input data and expected output data must contain values.");
            }

            if (inputData.Count() != expectedOutputData.Count())
            {
                throw new Exception("Input data and expected output data are not the same length.");
            }

            var inputNodes  = inputData.First().Length;
            var outputNodes = expectedOutputData.First().Length;


            var innerLayers      = 2;
            var hiddenLayerDepth = 16;

            // Inner Layer + Input + Output = 3
            var layers = new int[innerLayers + 2];

            layers[0] = inputNodes;
            for (int i = 1; i < innerLayers + 1; i++)
            {
                layers[i] = hiddenLayerDepth;
            }
            layers[innerLayers + 1] = outputNodes;

            // Create the NN.
            var network = new MLPNeuralNetwork(layers);

            // Train the network with our known good data.
            network.Train(inputData, expectedOutputData, reportProgress);

            return(network);
        }