时序、序列和文本深度学习

创建和训练用于时序分类、回归和预测任务的网络

创建和训练用于时序分类、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

Design, visualize, and train deep learning networks

函数

全部展开

训练网络

`trainingOptions`	Options for training deep learning neural network
`trainNetwork`	Train deep learning neural network
`analyzeNetwork`	Analyze deep learning network architecture

网络层

`sequenceInputLayer`	Sequence input layer
`featureInputLayer`	Feature input layer
`lstmLayer`	Long short-term memory (LSTM) layer
`bilstmLayer`	Bidirectional long short-term memory (BiLSTM) layer
`gruLayer`	Gated recurrent unit (GRU) layer
`sequenceFoldingLayer`	Sequence folding layer
`sequenceUnfoldingLayer`	Sequence unfolding layer
`flattenLayer`	Flatten layer
`fullyConnectedLayer`	Fully connected layer
`reluLayer`	Rectified Linear Unit (ReLU) layer
`leakyReluLayer`	Leaky Rectified Linear Unit (ReLU) layer
`clippedReluLayer`	Clipped Rectified Linear Unit (ReLU) layer
`eluLayer`	Exponential linear unit (ELU) layer
`tanhLayer`	Hyperbolic tangent (tanh) layer
`swishLayer`	Swish layer
`dropoutLayer`	Dropout layer
`softmaxLayer`	Softmax layer
`classificationLayer`	Classification output layer
`regressionLayer`	Create a regression output layer

预测

`classify`	Classify data using a trained deep learning neural network
`predict`	Predict responses using a trained deep learning neural network
`activations`	Compute deep learning network layer activations
`predictAndUpdateState`	Predict responses using a trained recurrent neural network and update the network state
`classifyAndUpdateState`	Classify data using a trained recurrent neural network and update the network state
`resetState`	Reset the state of a recurrent neural network
`confusionchart`	Create confusion matrix chart for classification problem
`sortClasses`	Sort classes of confusion matrix chart

自定义训练循环

padsequences Pad or truncate sequence data to same length

模块

全部展开

深度神经网络

Predict	Predict responses using a trained deep learning neural network
Stateful Predict	Predict responses using a trained recurrent neural network
Stateful Classify	Classify data using a trained deep learning recurrent neural network

属性

ConfusionMatrixChart Properties

Confusion matrix chart appearance and behavior

示例和操作指南

序列和时序

使用深度学习进行序列分类

此示例说明如何使用长短期记忆 (LSTM) 网络对序列数据进行分类。

使用深度学习进行“序列到序列”分类

此示例说明如何使用长短期记忆 (LSTM) 网络对序列数据的每个时间步进行分类。

使用深度学习进行“序列到序列”回归

此示例说明如何使用深度学习预测发动机的剩余使用寿命 (RUL)。

使用深度学习进行时序预测

此示例说明如何使用长期短期记忆 (LSTM) 网络预测时序数据。

使用深度学习对视频进行分类

此示例说明如何通过将预训练图像分类模型和 LSTM 网络相结合来创建视频分类网络。

Classify Videos Using Deep Learning with Custom Training Loop

This example shows how to create a network for video classification by combining a pretrained image classification model and a sequence classification network.

使用深度学习进行语音命令识别

此示例说明如何训练一个深度学习模型来检测音频中是否存在语音命令。此示例使用语音命令数据集 [1] 来训练卷积神经网络，以识别给定的一组命令。

Image Captioning Using Attention

This example shows how to train a deep learning model for image captioning using attention.

使用序列数据的自定义小批量数据存储来训练网络

此示例说明如何使用自定义小批量数据存储基于无法放入内存的序列数据来训练深度学习网络。

Visualize Activations of LSTM Network

This example shows how to investigate and visualize the features learned by LSTM networks by extracting the activations.

Sequence-to-Sequence Classification Using 1-D Convolutions

This example shows how to classify each time step of sequence data using a generic temporal convolutional network (TCN).

Chemical Process Fault Detection Using Deep Learning

This example shows how to use simulation data to train a neural network that can detect faults in a chemical process.