内置训练
使用内置训练函数训练深度学习网络
定义网络架构后，您可以使用 trainingOptions 函数定义训练参数。然后，您可以使用 trainnet 函数训练网络。使用经过训练的网络预测类标签或数值响应。
App
| 时间序列建模器 | Train models for time series prediction (自 R2026a 起) |
函数
主题
训练基础
- 创建简单的深度学习神经网络以用于分类
此示例说明如何创建和训练简单的卷积神经网络来进行深度学习分类。 - 针对回归训练卷积神经网络
此示例说明如何训练卷积神经网络来预测手写数字的旋转角度。 - Create Custom Deep Learning Training Plot
This example shows how to create a custom training plot that updates at each iteration during training of deep learning neural networks usingtrainnet. (自 R2023b 起) - Custom Stopping Criteria for Deep Learning Training
This example shows how to stop training of deep learning neural networks based on custom stopping criteria usingtrainnet. (自 R2023b 起) - Speed Up Deep Neural Network Training
Learn how to accelerate deep neural network training. - Define Custom Learning Rate Schedule
This example shows how to define a time-based decay learning rate schedule and use it to train a neural network. - 训练具有多个输出的网络
此示例说明如何训练具有多个输出的深度学习网络，来预测手写数字的标签和旋转角度。 - Train Network with Complex-Valued Data
This example shows how to predict the frequency of a complex-valued waveform using a 1-D convolutional neural network. - Denoise Complex-Valued Signal Using Multilayer Perceptron
Create and train a complex-valued neural network for denoising complex-valued signals. (自 R2026a 起) - Data Sets for Deep Learning
Discover data sets for various deep learning tasks. - Deep Learning Metrics
Comparison of metrics for deep learning tasks. - Transition Legacy Neural Network Code to dlnetwork Workflows
Learn how to transition legacy neural network code todlnetworkworkflows.
表格数据工作流
- Training Neural Networks with Tabular Data
Learn about training neural networks with tabular data. - Train Neural Network with Tabular Data
This example shows how to train a neural network with tabular data. (自 R2023b 起) - In-Context Learning for Tabular Classification Using a Prior-Data Fitted Network
This example shows how to train a prior-data fitted network (PFN) to classify tabular data. (自 R2026a 起)
序列数据工作流
- Training Neural Networks with Time Series Data
Learn about training neural networks with time series data. (自 R2026a 起) - 使用深度学习进行时间序列预测
此示例说明如何使用长期短期记忆 (LSTM) 网络预测时间序列数据。 - Compare Deep Learning and ARMA Models for Time Series Modeling
This example shows how to train and compare different models for time series modeling using the Time Series Modeler app. (自 R2026a 起) - 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“序列到单个”回归
此示例说明如何使用长短期记忆 (LSTM) 神经网络来预测波形的频率。 - 使用一维卷积进行序列分类
此示例说明如何使用一维卷积神经网络对序列数据进行分类。 - Train Network with LSTM Projected Layer
Train a deep learning network with an LSTM projected layer for sequence-to-label classification. - Train Sequence Classification Network Using Data with Imbalanced Classes
This example shows how to classify sequences with a 1-D convolutional neural network using class weights to modify the training to account for imbalanced classes. - 使用一维卷积进行“序列到序列”分类
此示例说明如何使用通用时序卷积网络 (TCN) 对序列数据的每个时间步进行分类。 - Sequence Classification Using CNN-LSTM Network
This example shows how to create a 2-D CNN-LSTM network for speech classification tasks by combining a 2-D convolutional neural network (CNN) with a long short-term memory (LSTM) layer. - 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。 - 使用深度学习训练语音命令识别模型
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，以识别一组命令。 - Chemical Process Fault Detection Using Deep Learning
Use simulation data to train a neural network than can detect faults in a chemical process. - 使用深度学习进行电池荷电状态估计
定义需求、准备数据、训练深度学习网络、验证稳健性、将网络集成到 Simulink 中以及部署模型。 (自 R2024b 起) - Autoregressive Time Series Prediction Using Deep Learning
This example shows how to interactively train an autoregressive deep neural network using the Time Series Modeler app to predict electricity consumption. (自 R2026a 起) - Denoise ECG Signals Using Deep Learning
This example shows how to interactively train deep neural networks to remove noise from heartbeat electrocardiogram (ECG) signals using the Time Series Modeler app. (自 R2026a 起) - Build Transformer Network for Time Series Regression
This example shows how to interactively build and train a transformer network to predict engine torque using the Time Series Modeler app. (自 R2026a 起) - Create Virtual Sensors Interactively Using Deep Learning and Generate C Code for Deployment
This example shows how to interactively train a deep neural network as a virtual sensor to predict battery state of charge using the Time Series Modeler app. (自 R2026a 起)
图像数据工作流
- 基于图像和特征数据训练网络
此示例说明如何使用图像和特征输入数据来训练对手写数字进行分类的网络。 - Multilabel Image Classification Using Deep Learning
This example shows how to use transfer learning to train a deep learning model for multilabel image classification. - Build Image-to-Image Regression Network Using Deep Network Designer
This example shows how to use Deep Network Designer to construct an image-to-image regression network for super resolution.
精选示例
使用深度网络设计器为迁移学习准备网络
以交互方式微调预训练的深度学习网络以学习新的图像分类任务。
Time Series Anomaly Detection Using Deep Learning
Detect anomalies in sequence or time series data.
训练残差网络进行图像分类
此示例说明如何创建包含残差连接的深度学习神经网络，并针对 CIFAR-10 数据对其进行训练。残差连接是卷积神经网络架构中的常见元素。使用残差连接可以改善网络中的梯度流，从而可以训练更深的网络。
Adapt Pretrained Audio Network for New Data Using Deep Network Designer
Interactively adapt a pretrained network to classify new audio signals using Deep Network Designer.
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。
Train Neural ODE Network
Train an augmented neural ordinary differential equation (ODE) network.
Create and Train NARX Network for Time Series Forecasting
Create and train a NARX network for time series forecasting and test its predictions in MATLAB® and Simulink®.
Compare Deep Learning Networks for Credit Default Prediction
Create, train, and compare three deep learning networks for predicting credit default probability.
使用深度学习训练语音命令识别模型
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，以识别一组命令。
Chemical Process Fault Detection Using Deep Learning
Use simulation data to train a neural network than can detect faults in a chemical process.
使用傅里叶神经算子求解 PDE
此示例说明如何训练傅里叶神经算子 (FNO) 神经网络来输出偏微分方程 (PDE) 的解。
