La traduction de cette page n'est pas à jour. Cliquez ici pour voir la dernière version en anglais.

Ajustement du Deep Learning

Ajustez de manière programmatique des options d’apprentissage, reprendre l’apprentissage depuis un point de contrôle et analyser des exemples contradictoires

Pour découvrir comment définir des options avec la fonction trainingOptions, veuillez consulter Set Up Parameters and Train Convolutional Neural Network. Après avoir identifié de bonnes options de départ, vous pouvez automatiser le balayage d’hyperparamètres ou essayer l’optimisation bayésienne avec Experiment Manager.

Analysez la robustesse du réseau en générant des exemples contradictoires. Vous pouvez ensuite utiliser l’apprentissage antagoniste avec la méthode FGSM (Fast Gradient Sign Method) pour entraîner un réseau robuste aux perturbations antagonistes.

Applications

Deep Network Designer

Concevoir, visualiser et entraîner des réseaux de Deep Learning

Fonctions

`trainingOptions`	Options d’un réseau de neurones d’apprentissage pour le Deep Learning
`trainNetwork`	(Not recommended) Train neural network

Rubriques

Set Up Parameters and Train Convolutional Neural Network
Learn how to set up training parameters for a convolutional neural network.
Deep Learning Using Bayesian Optimization
This example shows how to apply Bayesian optimization to deep learning and find optimal network hyperparameters and training options for convolutional neural networks.
Train Deep Learning Networks in Parallel
This example shows how to run multiple deep learning experiments on your local machine.
Train Network Using Custom Training Loop
This example shows how to train a network that classifies handwritten digits with a custom learning rate schedule.
Compare Activation Layers
This example shows how to compare the accuracy of training networks with ReLU, leaky ReLU, ELU, and swish activation layers.
Deep Learning Tips and Tricks
Learn how to improve the accuracy of deep learning networks.
Train Robust Deep Learning Network with Jacobian Regularization
Train a neural network that is robust to adversarial examples using a Jacobian regularization scheme.
Specify Custom Weight Initialization Function
This example shows how to create a custom He weight initialization function for convolution layers followed by leaky ReLU layers.
Compare Layer Weight Initializers
This example shows how to train deep learning networks with different weight initializers.

Informations connexes

Application Experiment Manager

Exemples présentés

Resume Training from Checkpoint Network

Save checkpoint networks while training a deep learning network and resume training from a previously saved network.

Ouvrir le live script

Generate Untargeted and Targeted Adversarial Examples for Image Classification

Use the fast gradient sign method (FGSM) and the basic iterative method (BIM) to generate adversarial examples for a pretrained neural network.

Ouvrir le live script

Generate Adversarial Examples for Semantic Segmentation

Generate adversarial examples for a semantic segmentation network using the basic iterative method (BIM).

Ouvrir le live script

Train Image Classification Network Robust to Adversarial Examples

Train a neural network that is robust to adversarial examples using fast gradient sign method (FGSM) adversarial training.

Ouvrir le live script

Train Robust Deep Learning Network with Jacobian Regularization

Train a neural network that is robust to adversarial examples using a Jacobian regularization scheme.

Ouvrir le live script