Main Content

本页的翻译已过时。点击此处可查看最新英文版本。

使用时间序列和序列数据进行深度学习

创建和训练用于时间序列分类、回归和预测任务的网络

创建和训练用于时间序列分类、回归和预测任务的网络。训练用于“序列到单个”或“序列到标签”的分类和回归问题的长短期记忆 (LSTM) 网络。您可以使用单词嵌入层对文本数据训练 LSTM 网络(需要 Text Analytics Toolbox™),或使用频谱图对音频数据训练卷积神经网络(需要 Audio Toolbox™)。

App

深度网络设计器设计、可视化和训练深度学习网络

函数

全部展开

trainingOptionsOptions for training deep learning neural network
trainNetworkTrain deep learning neural network
analyzeNetworkAnalyze deep learning network architecture

输入层

sequenceInputLayerSequence input layer
featureInputLayerFeature input layer

循环层

lstmLayerLong short-term memory (LSTM) layer
bilstmLayerBidirectional long short-term memory (BiLSTM) layer
gruLayerGated recurrent unit (GRU) layer

卷积和全连接层

convolution1dLayer1-D convolutional layer
fullyConnectedLayerFully connected layer

池化层

maxPooling1dLayer1-D max pooling layer
averagePooling1dLayer1-D average pooling layer
globalMaxPooling1dLayer1-D global max pooling layer
globalAveragePooling1dLayer1-D global average pooling layer

激活层和丢弃层

reluLayerRectified Linear Unit (ReLU) layer
leakyReluLayerLeaky Rectified Linear Unit (ReLU) layer
clippedReluLayerClipped Rectified Linear Unit (ReLU) layer
eluLayerExponential linear unit (ELU) layer
tanhLayerHyperbolic tangent (tanh) layer
swishLayerSwish layer
softmaxLayerSoftmax 层
dropoutLayerDropout layer
functionLayerFunction layer

数据操作

sequenceFoldingLayerSequence folding layer
sequenceUnfoldingLayerSequence unfolding layer
flattenLayerFlatten layer

输出层

classificationLayerClassification output layer
regressionLayer创建回归输出层
classifyClassify data using trained deep learning neural network
predictPredict responses using trained deep learning neural network
activationsCompute deep learning network layer activations
predictAndUpdateStatePredict responses using a trained recurrent neural network and update the network state
classifyAndUpdateStateClassify data using a trained recurrent neural network and update the network state
resetStateReset state parameters of neural network
confusionchartCreate confusion matrix chart for classification problem
sortClassesSort classes of confusion matrix chart
padsequencesPad or truncate sequence data to same length

模块

全部展开

PredictPredict responses using a trained deep learning neural network
Stateful PredictPredict responses using a trained recurrent neural network
Stateful ClassifyClassify data using a trained deep learning recurrent neural network

属性

ConfusionMatrixChart PropertiesConfusion matrix chart appearance and behavior

示例和操作指南

序列和时间序列

概念