Merge pull request #41 from Sergio0694/dev

Dev

Author: Sergio0694

NeuralNetwork.NET.Cuda/APIS/CuDnnNetworkLayers.cs

@@ -0,0 +1,70 @@
+using JetBrains.Annotations;
+using NeuralNetworkNET.APIs.Enums;
+using NeuralNetworkNET.APIs.Interfaces;
+using NeuralNetworkNET.APIs.Structs;
+using NeuralNetworkNET.Cuda.Layers;
+using NeuralNetworkNET.Networks.Activations;
+
+namespace NeuralNetworkNET.APIs
+{
+    /// <summary>
+    /// A static class that exposes the available cuDNN network layer types
+    /// </summary>
+    public static class CuDnnNetworkLayers
+    {
+        /// <summary>
+        /// Creates a new fully connected layer with the specified number of input and output neurons, and the given activation function
+        /// </summary>
+        /// <param name="input">The input <see cref="TensorInfo"/> descriptor</param>
+        /// <param name="neurons">The number of output neurons</param>
+        /// <param name="activation">The desired activation function to use in the network layer</param>
+        /// <param name="weightsMode">The desired initialization mode for the weights in the network layer</param>
+        /// <param name="biasMode">The desired initialization mode to use for the layer bias values</param>
+        [PublicAPI]
+        [Pure, NotNull]
+        public static INetworkLayer FullyConnected(
+            in TensorInfo input, int neurons, ActivationFunctionType activation,
+            WeightsInitializationMode weightsMode = WeightsInitializationMode.GlorotUniform, BiasInitializationMode biasMode = BiasInitializationMode.Zero) 
+            => new CuDnnFullyConnectedLayer(input, neurons, activation, weightsMode, biasMode);
+
+        /// <summary>
+        /// Creates a fully connected softmax output layer (used for classification problems with mutually-exclusive classes)
+        /// </summary>
+        /// <param name="input">The input <see cref="TensorInfo"/> descriptor</param>
+        /// <param name="outputs">The number of output neurons</param>
+        /// <param name="weightsMode">The desired initialization mode for the weights in the network layer</param>
+        /// <param name="biasMode">The desired initialization mode to use for the layer bias values</param>
+        [PublicAPI]
+        [Pure, NotNull]
+        public static INetworkLayer Softmax(
+            in TensorInfo input, int outputs, 
+            WeightsInitializationMode weightsMode = WeightsInitializationMode.GlorotUniform, BiasInitializationMode biasMode = BiasInitializationMode.Zero) 
+            => new CuDnnSoftmaxLayer(input, outputs, weightsMode, biasMode);
+
+        /// <summary>
+        /// Creates a convolutional layer with the desired number of kernels
+        /// </summary>
+        /// <param name="input">The input volume to process</param>
+        /// <param name="kernel">The volume information of the kernels used in the layer</param>
+        /// <param name="kernels">The number of convolution kernels to apply to the input volume</param>
+        /// <param name="activation">The desired activation function to use in the network layer</param>
+        /// <param name="mode">The desired convolution mode to use</param>
+        /// <param name="biasMode">Indicates the desired initialization mode to use for the layer bias values</param>
+        [PublicAPI]
+        [Pure, NotNull]
+        public static INetworkLayer Convolutional(
+            TensorInfo input, (int X, int Y) kernel, int kernels, ActivationFunctionType activation, 
+            ConvolutionMode mode = ConvolutionMode.Convolution, 
+            BiasInitializationMode biasMode = BiasInitializationMode.Zero) 
+            => new CuDnnConvolutionalLayer(input, kernel, kernels, activation, mode, biasMode);
+
+        /// <summary>
+        /// Creates a pooling layer with a window of size 2 and a stride of 2
+        /// </summary>
+        /// <param name="input">The input volume to pool</param>
+        /// <param name="activation">The desired activation function to use in the network layer</param>
+        [PublicAPI]
+        [Pure, NotNull]
+        public static INetworkLayer Pooling(TensorInfo input, ActivationFunctionType activation) => new CuDnnPoolingLayer(input, activation);
+    }
+}

NeuralNetwork.NET.Cuda/APIs/Enums/ConvolutionMode.cs

@@ -0,0 +1,18 @@
+namespace NeuralNetworkNET.APIs.Enums
+{
+    /// <summary>
+    /// A simple wrapper over the <see cref="Alea.cuDNN.ConvolutionMode"/> <see cref="enum"/>
+    /// </summary>
+    public enum ConvolutionMode
+    {
+        /// <summary>
+        /// The default convolution mode, with the kernel taargeting pixels in the opposite position
+        /// </summary>
+        Convolution = 0,
+
+        /// <summary>
+        /// The cross-correlation mode (equivalent to a convolution with a flipped kernel)
+        /// </summary>
+        CrossCorrelation = 1
+    }
+}