Deep learning con Simulink
R2026aImplemente la funcionalidad de deep learning en modelos de Simulink® usando bloques de las bibliotecas de bloques Deep Neural Networks, Python Neural Networks y Deep Learning Layers (incluidas en Deep Learning Toolbox™) o usando el bloque Deep Learning Object Detector de la biblioteca de bloques Analysis & Enhancement (incluida en Computer Vision Toolbox™).
Para generar un modelo de Simulink que usa la biblioteca de bloques Deep Learning Layers para representar una red, utilice la función exportNetworkToSimulink.
Algunas funcionalidades de deep learning en Simulink usan un bloque MATLAB Function, que requiere un compilador compatible. La mayoría de plataformas incluyen un compilador de C predeterminado con la instalación de MATLAB®. Cuando usa el lenguaje C++, debe instalar un compilador de C++ compatible. Para ver una lista de los compiladores compatibles, abra Supported and Compatible Compilers (Compiladores admitidos y compatibles), haga clic en la pestaña correspondiente a su sistema operativo, busque la tabla Simulink Product Family (Familia de productos Simulink) y vaya a la columna For Model Referencing, Accelerator mode, Rapid Accelerator mode, and MATLAB Function blocks (Para referencias de modelos, modo Acelerador, modo Acelerador rápido y bloques MATLAB Function). Si tiene instalados varios compiladores compatibles con MATLAB en su sistema, puede cambiar el compilador predeterminado usando el comando mex -setup. Consulte Cambiar el compilador predeterminado.
Funciones
exportNetworkToSimulink | Generate Simulink model that contains deep learning layer blocks and subsystems that correspond to deep learning layer objects (Desde R2024b) |
Bloques
 | Classify | Clasificar datos con una red neuronal de deep learning entrenada |
 | Predict | Predecir respuestas usando una red neuronal de deep learning entrenada |
 | Stateful Classify | Classify data using a trained deep learning recurrent neural network |
 | Stateful Predict | Predict responses using a trained recurrent neural network |
 | Deep Learning Object Detector | Detect objects using trained deep learning object detector |
 | TensorFlow Model Predict | Predict responses using pretrained Python TensorFlow model (Desde R2024a) |
 | PyTorch Model Predict | Predict responses using pretrained Python PyTorch model (Desde R2024a) |
 | ONNX Model Predict | Predict responses using pretrained Python ONNX model (Desde R2024a) |
 | Custom Python Model Predict | Predict responses using pretrained custom Python model (Desde R2024a) |
 | Clipped ReLU Layer | Clipped Rectified Linear Unit (ReLU) layer (Desde R2024b) |
 | GELU Layer | Gaussian error linear unit (GELU) layer (Desde R2024b) |
 | Leaky ReLU Layer | Leaky rectified linear unit (ReLU) layer (Desde R2024b) |
 | PReLU Layer | Parameterized rectified linear unit (PReLU) layer (Desde R2026a) |
 | ReLU Layer | Rectified linear unit (ReLU) layer (Desde R2024b) |
 | Sigmoid Layer | Sigmoid layer (Desde R2024b) |
 | Softmax Layer | Softmax layer (Desde R2024b) |
 | Swish Layer | Swish layer (Desde R2026a) |
 | Tanh Layer | Hyperbolic tangent (tanh) layer (Desde R2024a) |
 | Addition Layer | Addition layer (Desde R2024b) |
 | Concatenation Layer | Concatenation layer (Desde R2024b) |
 | Depth Concatenation Layer | Depth concatenation layer (Desde R2024b) |
 | Multiplication Layer | Multiplication layer (Desde R2024b) |
 | Convolution 1D Layer | 1-D convolutional layer (Desde R2024b) |
 | Convolution 2D Layer | 2-D convolutional layer (Desde R2024b) |
 | Convolution 3D Layer | 3-D convolutional layer (Desde R2024b) |
 | Fully Connected Layer | Fully connected layer (Desde R2024b) |
 | Rescale-Symmetric 1D | 1-D input layer with rescale-symmetric normalization (Desde R2024b) |
 | Rescale-Symmetric 2D | 2-D input layer with rescale-symmetric normalization (Desde R2024b) |
 | Rescale-Symmetric 3D | 3-D input layer with rescale-symmetric normalization (Desde R2024b) |
 | Rescale-Zero-One 1D | 1-D input layer with rescale-zero-one normalization (Desde R2024b) |
 | Rescale-Zero-One 2D | 2-D input layer with rescale-zero-one normalization (Desde R2024b) |
 | Rescale-Zero-One 3D | 3-D input layer with rescale-zero-one normalization (Desde R2024b) |
 | Zerocenter 1D | 1-D input layer with zerocenter normalization (Desde R2024b) |
 | Zerocenter 2D | 2-D input layer with zerocenter normalization (Desde R2024b) |
 | Zerocenter 3D | 3-D input layer with zerocenter normalization (Desde R2024b) |
 | Zscore 1D | 1-D input layer with zscore normalization (Desde R2024b) |
 | Zscore 2D | 2-D input layer with zscore normalization (Desde R2024b) |
 | Zscore 3D | 3-D input layer with zscore normalization (Desde R2024b) |
 | Batch Normalization Layer | Batch normalization layer (Desde R2024b) |
 | Instance Normalization Layer | Instance normalization layer (Desde R2026a) |
 | Layer Normalization Layer | Layer normalization layer (Desde R2024b) |
 | Inverse Zerocenter | Inverse zero-center normalization (Desde R2026a) |
 | Inverse Zscore | Inverse Z-score normalization (Desde R2026a) |
 | Average Pooling 1D Layer | 1-D average pooling layer (Desde R2024b) |
 | Average Pooling 2D Layer | 2-D average pooling layer (Desde R2024b) |
 | Average Pooling 3D Layer | 3-D average pooling layer (Desde R2024b) |
 | Global Average Pooling 1D Layer | 1-D global average pooling layer (Desde R2024b) |
 | Global Average Pooling 2D Layer | 2-D global average pooling layer (Desde R2024b) |
 | Global Average Pooling 3D Layer | 3-D global average pooling layer (Desde R2024b) |
 | Global Max Pooling 1D Layer | 1-D global max pooling layer (Desde R2024b) |
 | Global Max Pooling 2D Layer | 2-D global max pooling layer (Desde R2024b) |
 | Global Max Pooling 3D Layer | 3-D global max pooling layer (Desde R2024b) |
 | Max Pooling 1D Layer | 1-D max pooling layer (Desde R2024b) |
 | Max Pooling 2D Layer | 2-D max pooling layer (Desde R2024b) |
 | Max Pooling 3D Layer | 3-D max pooling layer (Desde R2024b) |
 | Flatten Layer | Flatten layer (Desde R2024b) |
 | GRU Layer | Gated recurrent unit (GRU) layer for recurrent neural network (RNN) (Desde R2025a) |
 | GRU Projected Layer | Gated recurrent unit (GRU) projected layer for recurrent neural network (RNN) (Desde R2025a) |
 | LSTM Layer | Long short-term memory (LSTM) layer for recurrent neural network (RNN) (Desde R2024b) |
 | LSTM Projected Layer | Long short-term memory (LSTM) projected layer for recurrent neural network (RNN) (Desde R2024b) |
 | Dropout Layer | Dropout layer (Desde R2024b) |
 | Identity Layer | Identity layer (Desde R2026a) |
 | Scaling Layer | Scaling layer (Desde R2026a) |
 | Spatial Dropout Layer | Spatial dropout layer (Desde R2026a) |
Temas
Bloques Deep Learning Layer
- List of Deep Learning Layer Blocks and Subsystems
Discover all the deep learning layer blocks and subsystems in Simulink. - Implement Unsupported Deep Learning Layer Blocks
Learn how to implement unsupported deep learning layer blocks.
Imágenes
- Clasificar imágenes en Simulink con GoogLeNet
En este ejemplo se muestra cómo clasificar una imagen en Simulink® usando el bloqueImage Classifier. - Acceleration for Simulink Deep Learning Models
Improve simulation speed with accelerator and rapid accelerator modes. - Lane and Vehicle Detection in Simulink Using Deep Learning
This example shows how to use deep convolutional neural networks inside a Simulink® model to perform lane and vehicle detection. - Classify ECG Signals in Simulink Using Deep Learning
This example shows how to use wavelet transforms and a deep learning network within a Simulink (R) model to classify ECG signals. - Classify Images in Simulink with Imported TensorFlow Network
Import a pretrained TensorFlow™ network usingimportTensorFlowNetwork, and then use the Predict block for image classification in Simulink.
Secuencias
- Predecir y actualizar el estado de una red en Simulink
En este ejemplo se muestra cómo predecir respuestas de una red neuronal recurrente entrenada en Simulink® mediante el bloqueStateful Predict. - Clasificar y actualizar el estado de una red en Simulink
En este ejemplo se muestra cómo clasificar datos de una red neuronal recurrente entrenada en Simulink® mediante el bloqueStateful Classify. - Speech Command Recognition in Simulink
Detect the presence of speech commands in audio using a Simulink model. - Time Series Prediction in Simulink Using Deep Learning Network
This example shows how to use an LSTM deep learning network inside a Simulink® model to predict the remaining useful life (RUL) of an engine. - Simulate Calorie Burn Using Neural Network in Simulink
This example shows how to include a simple fully connected neural network in a Simulink® model that predicts calorie burn when given five time steps of sensor readings from a smart watch. (Desde R2024b) - Estimación del estado de carga de una batería utilizando deep learning
Defina los requisitos, prepare los datos, entrene las redes de deep learning, verifique su solidez, integre redes en Simulink y despliegue modelos. (Desde R2024b) - Physical System Modeling Using LSTM Network in Simulink
This example shows how to create a reduced order model (ROM) that acts as a virtual sensor in a Simulink® model using a long short-term memory (LSTM) neural network. - Classify Motor Faults Using Deep Learning
This example shows how to train a deep learning model to classify faults in a permanent magnet synchronous motor (PMSM) using simulated data across various revolutions per minute (RPM). (Desde R2025a) - Improve Performance of Deep Learning Simulations in Simulink
This example shows how to use code generation to improve the performance of deep learning simulations in Simulink®.
Reinforcement learning
- Control Water Level in a Tank Using a DDPG Agent (Reinforcement Learning Toolbox)
Use DDPG agent to control water level in a tank modeled in Simulink. - Train DDPG Agent for Adaptive Cruise Control (Reinforcement Learning Toolbox)
Train a DDPG agent for an adaptive cruise control application. - Train DQN Agent for Lane Keeping Assist Using Parallel Computing (Reinforcement Learning Toolbox)
Train a DQN agent for an automated driving application using parallel computing. - Train DDPG Agent for Path-Following Control (Reinforcement Learning Toolbox)
Train a DDPG agent for lane following control.
Coejecución de Python
- Classify Images Using TensorFlow Model Predict Block
Classify images using TensorFlow Model Predict block. - Classify Images Using ONNX Model Predict Block
Classify images using ONNX Model Predict block. - Classify Images Using PyTorch Model Predict Block
Classify images using PyTorch Model Predict block. - Predict Responses Using TensorFlow Model Predict Block
Predict Responses Using TensorFlow Model Predict block. - Predict Responses Using ONNX Model Predict Block
Predict Responses Using ONNX Model Predict block. - Predict Responses Using PyTorch Model Predict Block
Predict Responses Using PyTorch Model Predict block. - Predict Responses Using Custom Python Model in Simulink (Statistics and Machine Learning Toolbox)
This example shows how to use the Custom Python Model Predict (Statistics and Machine Learning Toolbox) block for prediction in Simulink®.
Generación de código
- Generación de código de deep learning a partir de aplicaciones de Simulink
Genere código C/C++ y GPU para despliegues en objetivos de escritorio o embebidos - Export Network to FMU
This example shows how to export a trained network as a Functional Mock-up Unit (FMU). (Desde R2023b)
Ejemplos destacados
Lane and Vehicle Detection in Simulink Using Deep Learning
Use deep convolutional neural networks inside a Simulink® model to perform lane and vehicle detection. This example takes the frames from a traffic video as an input, outputs two lane boundaries that correspond to the left and right lanes of the ego vehicle, and detects vehicles in the frame.
Classify ECG Signals in Simulink Using Deep Learning
Use wavelet transforms and a deep learning network within a Simulink (R) model to classify ECG signals. This example uses the pretrained convolutional neural network from the Classify Time Series Using Wavelet Analysis and Deep Learning example of the Wavelet Toolbox™ to classify ECG signals based on images from the CWT of the time series data. For information on training, see Classify Time Series Using Wavelet Analysis and Deep Learning (Wavelet Toolbox).
Time Series Prediction in Simulink Using Deep Learning Network
Use an LSTM deep learning network inside a Simulink® model to predict the remaining useful life (RUL) of an engine. You include the network inside the Simulink model by using a Stateful Predict block, which predicts the RUL at every simulation step.
Estimación del estado de carga de una batería en Simulink utilizando una red neuronal prealimentada
En este ejemplo se muestra cómo utilizar una red de deep learning prealimentada dentro de un modelo de Simulink® para predecir el estado de carga (SOC) de una batería. Puede incluir la red dentro del modelo de Simulink utilizando un bloque Predict, que predice el SOC en cada paso de la simulación.
Profile Deep Learning Network for Deployment
Profile a deep learning network in Simulink. Profiling your network before training is a good way of checking that the network meets your requirements, before spending time on training. This example shows how to check:
