scores2label

Convert prediction scores to labels

Since R2024a

Syntax

labels = scores2label(scores,classNames)

labels = scores2label(scores,classNames,dim)

[labels,topScores] = scores2label(___)

Description

labels = scores2label(scores,classNames) converts the specified prediction scores to labels by returning the label with the highest score in the channel dimension.

labels = scores2label(scores,classNames,dim) additionally specifies the channel dimension of the prediction scores.

[labels,topScores] = scores2label(___) also returns the scores that correspond to the output labels using any of the previous syntaxes.

Examples

Classify Image Using Pretrained Neural Network

Load a pretrained SqueezeNet neural network into the workspace.

[net,classNames] = imagePretrainedNetwork;

Read an image from a PNG file and classify it. To classify the image, first convert it to the data type single.

im = imread("peppers.png");
figure
imshow(im)

X = single(im);
scores = predict(net,X);
[label,score] = scores2label(scores,classNames);

Display the image with the predicted label and corresponding score.

figure
imshow(im)
title(string(label) + " (Score: " + score + ")")

Input Arguments

scores — Prediction scores

numeric array | dlarray object

Prediction scores, specified as a numeric array or dlarray object.

For a classification neural network, the elements of the output correspond to the scores for each class. The order of the scores matches the order of the categories in the training data. For example, if you train the neural network using the categorical labels TTrain, then the order of the scores matches the order of the categories given by categories(TTrain).

classNames — Class names

string array | categorical array | cell array of character vectors

Class names, specified as a string array, categorical array, or cell array of character vectors.

Data Types: string | cell | categorical

dim — Channel dimension

"auto" (default) | positive integer

Channel dimension, specified as one of these values:

• "auto" — Automatically determine the channel dimension of scores based on the number of elements in classNames. If scores contains exactly one dimension of size numel(classNames) that has elements that sum to one, then the function uses that dimension. Otherwise, the function throws an error.

• Positive integer — Use the specified dimension as the channel dimension. Use this option when scores contains multiple dimensions of size numel(classNames).

The scores2label function finds the top scores in the specified dimension.

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | char | string

Output Arguments

labels — Labels with highest score

categorical array

Labels with the highest score in the channel dimension of scores, returned as a categorical array.

topScores — Scores that correspond to the output labels

numeric array | dlarray object

Scores that correspond to the output labels, returned as a numeric array or dlarray object.

This argument has the same data type and format as scores.

Extended Capabilities

C/C++ Code Generation

Generate C and C++ code using MATLAB® Coder™.

Usage notes and limitations:

• Code generation supports these syntaxes only:

  • labels = scores2label(scores,classNames,dim)

  • [labels,topScores] = scores2label(scores,classNames,dim)

GPU Code Generation

Generate CUDA® code for NVIDIA® GPUs using GPU Coder™.

Usage notes and limitations:

• GPU code generation supports these syntaxes only:

  • labels = scores2label(scores,classNames,dim)

  • [labels,topScores] = scores2label(scores,classNames,dim)

Version History

Introduced in R2024a