主要内容

readall

R2026b

Read all data in label datastore

Since R2026b

Description

data = readall(ds) returns all the data contained in the label datastore specified by ds.

If all the data in the datastore does not fit in memory, then readall returns an error.

Input Arguments

collapse all

Datastore with labeled data for training a point cloud deep learning network, specified as a semanticPointLabelDatastore or boxLabelDatastore object.

Output Arguments

collapse all

Output data, returned as a cell array of categorical vectors or an N-by-2 cell array, depending on the type of datastore.

DatastoreOutput Description
semanticPointLabelDatastore

Cell array of all P-by-1 categorical vectors in the datastore, where P is the number of points in each point cloud. The number of vectors equals the total number of files in the datastore.

Each element in a vector assigns a semantic class label to the corresponding point in the point cloud.

boxLabelDatastore

N-by-2 cell matrix containing all label data in the datastore, where N is the total number of point clouds in the datastore.

The first column contains bounding boxes. Each element is an M-by-9 matrix of M cuboid bounding boxes in the format [xcenter,ycenter,zcenter,width,height,depth,rx,ry,rz].

The second column contains the label names corresponding to each bounding box. Each element is an M-by-1 categorical vector of label names.

Version History

Introduced in R2026b