half - Construct half-precision numeric object - MATLAB (original) (raw)

• Documentation
• Examples
• Functions
• Blocks
• Apps
• Videos
• Answers

Main Content

Construct half-precision numeric object

Description

Use the half constructor to assign a half-precision data type to a number or variable. A half-precision data type occupies 16 bits of memory, but its floating-point representation enables it to handle wider dynamic ranges than integer or fixed-point data types of the same size. For more information, see Floating-Point Numbers (Fixed-Point Designer) andWhat Is Half Precision? (Fixed-Point Designer).

For a list of functions that support code generation with half-precision inputs, see Half Precision Code Generation Support.

Creation

Syntax

Description

`a` = half([v](#mw%5F8c5fdae8-31a5-4dd4-a731-85e7d90fc5e8)) converts the values in v to half-precision.

Input Arguments

expand all

v — Input array

scalar | vector | matrix | multidimensional array

Input array, specified as a scalar, vector, matrix, or multidimensional array.

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | logical
Complex Number Support: Yes

Object Functions

These functions are supported for simulation with half-precision inputs in MATLAB®. MATLAB System object™ supports half-precision data type and MATLAB System (Simulink) block supports half-precision data type with real values. For a list of functions that support code generation with half-precision inputs, see Half Precision Code Generation Support (Fixed-Point Designer).

expand all

Math and Arithmetic

Data Types

Relational and Logical Operators

Array and Matrix Operations

Graphics

Deep Learning

To display a list of supported functions, at the MATLAB Command Window, enter:

Examples

collapse all

Convert Value to Half Precision

To cast a double-precision number to half precision, use thehalf function.

You can also use the half function to cast an existing variable to half-precision.

v = 3x3 single matrix

 8     1     6
 3     5     7
 4     9     2

a =

3x3 half matrix

 8     1     6
 3     5     7
 4     9     2

Limitations

• Arithmetic operations which combine half-precision and logical types are not supported.
• For additional usage notes and limitations, see Half Precision Code Generation Support.

Extended Capabilities

C/C++ Code Generation

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

• For a list of functions that support code generation with half-precision inputs and associated limitations, see Half Precision Code Generation Support.
• If your target hardware does not have native support for half-precision, thenhalf is used as a storage type, with arithmetic operations performed in single-precision.
• Some functions use half only as a storage type and the arithmetic is performed in single-precision, regardless of the target hardware.
• For deep learning code generation, half inputs are cast to single precision and computations are performed in single precision.
• In MATLAB, the isobject function returns true with a half-precision input. In generated code, this function returns false.

GPU Code Generation

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

• For a list of functions that support code generation with half-precision inputs and associated limitations, see Half Precision Code Generation Support.
• CUDA® compute capability of 5.3 or higher is required for generating and executing code with half-precision data types.
• CUDA toolkit version of 10.0 or later is required for generating and executing code with half-precision data types.
• You must set the memory allocation (malloc) mode to'Discrete' for generating CUDA code.
• Half-precision complex data types are not supported for GPU code generation.
• If your target hardware does not have native support for half-precision, thenhalf is used as a storage type, with arithmetic operations performed in single-precision.
• Some functions use half only as a storage type and the arithmetic is performed in single-precision, regardless of the target hardware.
• For deep learning code generation, half inputs are cast to single precision and computations are performed in single precision. To perform computations in half, set the library target to 'tensorrt' and set the data type to'FP16' in coder.DeepLearningConfig.
• In MATLAB, the isobject function returns true with a half-precision input. In generated code, this function returns false.

Version History

Introduced in R2018b