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 trainnet or define your own custom training function. Monitor your progress by using training plots. Use confusion matrices and custom metric functions to evaluate your trained network.

This page contains information about experiments for your AI workflows. For general information about using the app, see Experiment Manager.

Apps

Experiment Manager

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

Objects

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

Functions

`groupSubPlot`	Group metrics in experiment training plot (Since R2021a)
`recordMetrics`	Record metric values in experiment results table and training plot (Since R2021a)
`updateInfo`	Update information columns in experiment results table (Since R2021a)
`yscale`	Set training plot y-axis scale (linear or logarithmic) (Since R2024a)

Topics

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

Troubleshooting

Debug Deep Learning Experiments

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

Related Information

Featured Examples

Try Multiple Pretrained Networks for Transfer Learning

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

Since R2020a
Open 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.

Since R2020a
Open 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.

Open Script

Choose Training Configurations for LSTM Using Bayesian Optimization

Find optimal data architecture and network configurations for sequence-to-sequence regression using Bayesian optimization.

Since R2020b
Open Script

Run a Custom Training Experiment for Image Comparison

Train a twin neural network to identify similar images of handwritten characters.

Since R2021a
Open Script

Use Experiment Manager to Train Generative Adversarial Networks (GANs)

Create a custom training experiment to generate images of flowers.

Since R2021a
Open Script

Custom Training with Multiple GPUs in Experiment Manager

Configure multiple parallel workers to collaborate on each trial of a custom training experiment.

Since R2021a
Open Script