API Docs / Microsoft.VisualBasic.DeepLearning 6.0.9753.4776 6.0.9762.11310
Microsoft.VisualBasic.DeepLearning 6.0.9753.4776 .
Namespaces 10
Types 67
Members 508
01 Namespaces
Microsoft.VisualBasic.MachineLearning
Microsoft.VisualBasic.MachineLearning.CNN
Microsoft.VisualBasic.MachineLearning.CNN.data
Type Summary
Members
BackPropResult When we have done a back propagation of the network we will receive a result of weight adjustments required to learn.
7
DataBlock Holding all the data handled by the network.
44
OutputDefinition This class will hold the definitions that bridge two layers.
5
TrainResult Created by danielp on 1/27/17.
2
Microsoft.VisualBasic.MachineLearning.CNN.layers
Type Summary
Members
Conv2DTransposeLayer 5
ConvolutionLayer This layer uses different filters to find attributes of the data that affects the result.
8
DropoutLayer This layer will remove some random activations in order to defeat over-fitting.
5
FourierFeatureLayer Based on the paper "Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains" (2020) presented at NeurIPS (https://bmild.github.io/fourfeat/index.…
5
FullyConnectedLayer Neurons in a fully connected layer have full connections to all activations in the previous layer, as seen in regular Neural Networks.
7
InputLayer The input layer is a simple layer that will pass the data though and create a window into the full training data set.
8
Layer A convolution neural network is built of layers that the data traverses back and forth in order to predict what the network sees in the data.
4
LocalResponseNormalizationLayer This layer is useful when we are dealing with ReLU neurons.
5
MaxoutLayer Implements Maxout nonlinearity that computes x to max(x) where x is a vector of size group_size.
5
PoolingLayer This layer will reduce the dataset by creating a smaller zoomed out version.
6
RectifiedLinearUnitsLayer This is a layer of neurons that applies the non-saturating activation function f(x)=max(0,x).
5
SigmoidLayer Implements Sigmoid nonlinearity elementwise x to 1/(1+e^(-x)) so the output is between 0 and 1.
5
TanhLayer Implements Tanh nonlinearity elementwise x to tanh(x) so the output is between -1 and 1.
5
Microsoft.VisualBasic.MachineLearning.CNN.losslayers
Type Summary
Members
LossLayer Created by danielp on 1/25/17.
6
RegressionLayer Regression layer is used when your output is an area of data.
4
SoftMaxLayer This layer will squash the result of the activations in the fully connected layer and give you a value of 0 to 1 for all output activations.
4
SVMLayer This layer uses the input area trying to find a line to separate the correct activation from the incorrect ones.
4
Microsoft.VisualBasic.MachineLearning.CNN.trainers
Type Summary
Members
AdaDeltaTrainer Adaptive delta will look at the differences between the expected result and the current result to train the network.
3
AdaGradTrainer The adaptive gradient trainer will over time sum up the square of the gradient and use it to change the weights.
2
AdamTrainer Adaptive Moment Estimation is an update to RMSProp optimizer.
3
NesterovTrainer Another extension of gradient descent is due to Yurii Nesterov from 1983,[7] and has been subsequently generalized @author Daniel Persson (mailto.woden@gmail.com)
1
SGDTrainer Stochastic gradient descent (often shortened in SGD), also known as incremental gradient descent, is a stochastic approximation of the gradient descent optimization method for min…
1
TrainerAlgorithm Trainers take the generated output of activations and gradients in order to modify the weights in the network to make a better prediction the next time the network runs with a da…
11
WindowGradTrainer This is AdaGrad but with a moving window weighted average so the gradient is not accumulated over the entire history of the run.
1
Microsoft.VisualBasic.MachineLearning.Convolutional
feed-forward phase of deep Convolutional Neural Networks > https://github.com/atasoyhus/CeNiN
Type Summary
Members
CeNiN CeNiN (means "fetus" in Turkish) is a minimal implementation of feed-forward phase of deep Convolutional Neural Networks
11
Input the layer for image inputs
8
Layer 16
Output 6
Microsoft.VisualBasic.MachineLearning.NeuralNetwork
Type Summary
Members
HiddenLayersView 只读视图：表示网络之中的全部隐藏层集合。 保留旧 "HiddenLayers" 公开接口之中的 HiddenLayersView.Count 与索引器， 但其内部不再维护遗留的 Layer/Neuron/Synapse 数据图。
4
Network 人工神经网络计算用的对象模型。 本类以 CNN 的 ConvolutionalNN 全连接网络为统一的内部计算内核： 前向传播与反向传播（含参数更新）均由 CNN 内核完成； 对外公开的 Network.InputLayer / Network.HiddenLayer / Network.OutputLayer / Network.Activat…
24
NetworkKernel 以 CNN 的 ConvolutionalNN 全连接网络作为 NeuralNetwork 的统一计算内核。 本模块通过 LayerBuilder 构建等价的 CNN 全连接网络， 供 Netz 与 Network 两个公开类作为内部计算内核使用。
2
NetworkLayerView 只读视图：表示一个神经网络层（输入层 / 隐藏层 / 输出层）。 本视图不依赖任何遗留计算类（Layer/Neuron/Synapse）， 其规模（NetworkLayerView.Count）与输出（NetworkLayerView.Output）均从 CNN 内核派生， 用于支撑 Network 的 Network.InputLayer /…
3
Netz Neural Network for regression analysis.
10
Microsoft.VisualBasic.MachineLearning.RNN
nuget