HDK
Main Page
Related Pages
Modules
Namespaces
Classes
Files
Examples
File List
File Members
All
Classes
Namespaces
Files
Functions
Variables
Typedefs
Enumerations
Enumerator
Friends
Macros
Groups
Pages
ML_Types.h
Go to the documentation of this file.
1
/*
2
* PROPRIETARY INFORMATION. This software is proprietary to
3
* Side Effects Software Inc., and is not to be reproduced,
4
* transmitted, or disclosed in any way without written permission.
5
*
6
* COMMENTS:
7
*
8
*/
9
10
#pragma once
11
12
#include "
ML_API.h
"
13
14
/*
15
* Enumeration of supported ONNX execution providers
16
*/
17
enum class
ML_ExecutionProvider
18
{
19
/// Automatically determine the best provider based on the platform
20
/// and hardware available
21
AUTOMATIC
,
22
23
/// CPU inference provider, works on all platforms
24
CPU
,
25
26
/// CUDA/cudNN inference provider, works on Windows and Linux assuming
27
/// the GPU driver is new enough and cudNN is installed on the system
28
CUDA
,
29
30
/// Uses DirectML, only supported on Windows
31
DIRECTML
,
32
33
/// Uses CoreML, only supported on macOS
34
COREML
,
35
36
/// The total number of providers
37
COUNT
38
};
ML_ExecutionProvider
ML_ExecutionProvider
Definition:
ML_Types.h:17
ML_ExecutionProvider::COREML
Uses CoreML, only supported on macOS.
ML_ExecutionProvider::CUDA
ML_ExecutionProvider::AUTOMATIC
ML_ExecutionProvider::CPU
CPU inference provider, works on all platforms.
ML_API.h
ML_ExecutionProvider::DIRECTML
Uses DirectML, only supported on Windows.
ML_ExecutionProvider::COUNT
The total number of providers.
ML
ML_Types.h
Generated on Thu Sep 4 2025 02:37:55 for HDK by
1.8.6