Manage Experiments

Train networks under multiple initial conditions, interactively tune training options, and evaluate your results

Use the Experiment Manager app to find optimal training options for neural networks by sweeping through a range of hyperparameter values or by using Bayesian optimization. Use the built-in function trainNetwork or define your own custom training function. Test different training configurations at the same time by running your experiment in parallel. Offload experiments as batch jobs in a remote cluster so that you can continue working or close your MATLAB^® session while your experiment is running. Monitor your progress by using training plots. Use confusion matrices and custom metric functions to evaluate your trained network. Use visualizations, filters, and annotations to manage your experiment results and record your observations. Access past experiment definitions to keep track of the combinations of hyperparameters that produce each of your results.

App

Experiment Manager

Design and run experiments to train and compare deep learning networks (Da R2020a)

Oggetti

experiments.Monitor Update results table and training plots for custom training experiments (Da R2021a)

Funzioni

`groupSubPlot`	Group metrics in experiment training plot (Da R2021a)
`recordMetrics`	Record metric values in experiment results table and training plot (Da R2021a)
`updateInfo`	Update information columns in experiment results table (Da R2021a)

Argomenti

Create a Deep Learning Experiment for Classification
Train a deep learning network for classification using Experiment Manager. (Da R2020a)
Create a Deep Learning Experiment for Regression
Train a deep learning network for regression using Experiment Manager. (Da R2020a)
Use Experiment Manager to Train Networks in Parallel
Run multiple simultaneous trials or one trial at a time on multiple workers. (Da R2020b)
Offload Deep Learning Experiments as Batch Jobs to a Cluster
Run experiments on a cluster so you can continue working or close MATLAB. (Da R2022a)
Evaluate Deep Learning Experiments by Using Metric Functions
Use metric functions to evaluate the results of an experiment. (Da R2020a)
Tune Experiment Hyperparameters by Using Bayesian Optimization
Find optimal network hyperparameters and training options for convolutional neural networks. (Da R2020b)
Use Bayesian Optimization in Custom Training Experiments
Create custom training experiments that use Bayesian optimization. (Da R2021b)
Generate Experiment Using Deep Network Designer
Use Experiment Manager to tune the hyperparameters of a network trained in Deep Network Designer.
Keyboard Shortcuts for Experiment Manager
Navigate Experiment Manager using only your keyboard.

Risoluzione dei problemi

Debug Deep Learning Experiments

Diagnose problems in your setup, training, and metric functions. (Da R2023a)

Informazioni complementari

Esempi in primo piano

Try Multiple Pretrained Networks for Transfer Learning

Configure an experiment that replaces layers of different pretrained networks for transfer learning.

Da R2020a

Apri script

Experiment with Weight Initializers for Transfer Learning

Configure an experiment that initializes the weights of convolution and fully connected layers using different weight initializers.

Da R2020a

Apri script

Audio Transfer Learning Using Experiment Manager

Configure an experiment that compares the performance of multiple pretrained networks applied to a speech command recognition task using transfer learning.

Apri script