Main Content

Visualize Deep Neural Networks

Plot training progress, assess accuracy, explain predictions, and visualize features learned by an image network

Monitor training progress using built-in plots of network accuracy and loss. Investigate trained networks using visualization techniques such as Grad-CAM, occlusion sensitivity, LIME, and deep dream.


Deep Network DesignerDesign and visualize deep learning networks


alle erweitern

analyzeNetworkAnalyze deep learning network architecture
trainingProgressMonitorMonitor and plot training progress for deep learning custom training loops (Seit R2022b)
updateInfoUpdate information values for custom training loops (Seit R2022b)
recordMetricsRecord metric values for custom training loops (Seit R2022b)
groupSubPlotGroup metrics in training plot (Seit R2022b)
plotPlot neural network architecture
predictCompute deep learning network output for inference (Seit R2019b)
minibatchpredictMini-batched neural network prediction (Seit R2024a)
scores2labelConvert prediction scores to labels (Seit R2024a)
deepDreamImageVisualize network features using deep dream
occlusionSensitivityExplain network predictions by occluding the inputs (Seit R2019b)
imageLIMEExplain network predictions using LIME (Seit R2020b)
gradCAMExplain network predictions using Grad-CAM (Seit R2021a)
confusionchartCreate confusion matrix chart for classification problem
sortClassesSort classes of confusion matrix chart
rocmetricsReceiver operating characteristic (ROC) curve and performance metrics for binary and multiclass classifiers (Seit R2022b)
addMetricsCompute additional classification performance metrics (Seit R2022b)
averageCompute performance metrics for average receiver operating characteristic (ROC) curve in multiclass problem (Seit R2022b)


ConfusionMatrixChart PropertiesConfusion matrix chart appearance and behavior
ROCCurve PropertiesReceiver operating characteristic (ROC) curve appearance and behavior (Seit R2022b)



Training Progress and Performance