Main Content

HDF5 Files

Hierarchical Data Format, Version 5

High-level access functions make it easy to read and view data in an HDF5 file or write a variable from the MATLAB® workspace into an HDF5 file.

Low-level functions in the HDF5 library package provide direct access to the more than 300 functions in the HDF5 C library of the HDF Group. To create or write to non-numeric datasets or attributes, you must use the low-level functions.

To use the MATLAB low-level HDF5 functions, you must be familiar with HDF5 C API programming concepts, described at The HDF Group website.


expand all

h5createCreate HDF5 dataset
h5dispDisplay contents of HDF5 file
h5infoInformation about HDF5 file
h5readRead data from HDF5 dataset
h5readattRead attribute from HDF5 file
h5writeWrite to HDF5 dataset
h5writeattWrite HDF5 attribute
Library (H5)General-purpose functions for use with entire HDF5 library
Attribute (H5A)Metadata associated with datasets or groups
Dataset (H5D)Multidimensional arrays of data elements and supporting metadata
Dimension Scale (H5DS)Dimension scale associated with dataset dimensions
Error (H5E)Error handling
File (H5F)HDF5 file access
Group (H5G)Organization of objects in file
Identifier (H5I)HDF5 object identifiers
Link (H5L)Links in HDF5 file
MATLAB (H5ML)MATLAB Utility functions not part of HDF5 C library
Object (H5O)Objects in file
Property (H5P)Object property lists
Reference (H5R)HDF5 references
Dataspace (H5S)Dimensionality of dataset
Datatype (H5T)Datatype of elements in a dataset
Filters and Compression (H5Z)Inline data filters, data compression


Importing HDF5 Files

Reading and writing data and metadata using the Hierarchical Data Format (HDF5) file format.

Exporting to HDF5 Files

Hierarchical Data Format, Version 5, (HDF5) is a general-purpose, machine-independent standard for storing scientific data in files, developed by the National Center for Supercomputing Applications (NCSA).

Working with Non-ASCII Characters in HDF5 Files

MATLAB support for non-ASCII data and metadata in HDF5 files.

Read and Write Data Concurrently Using Single-Writer/Multiple-Reader (SWMR)

Write data to an HDF5 file in one process while you concurrently read from the file in one or more reader processes.

Work with HDF5 Virtual Datasets (VDS)

Access data stored across multiple HDF5 files as a single, unified HDF5 dataset.