[Mobile]

[taocheng]

Describe the issue

when I use onnxruntime-objc,
env = try ORTEnv(loggingLevel: ORTLoggingLevel.warning)
let options = try ORTSessionOptions()
try options.setLogSeverityLevel(ORTLoggingLevel.warning)
try options.setIntraOpNumThreads(threadCount)
try options.appendExecutionProvider("xnnpack",providerOptions: ["intra_op_num_threads" : "2"])
let session = try ORTSession(env: env, modelPath: modelPath, sessionOptions: options)
the session get error,i have no idea ,why this happen, how can I use func appendExecutionProvider?

To reproduce

when I use onnxruntime-objc,
env = try ORTEnv(loggingLevel: ORTLoggingLevel.warning)
let options = try ORTSessionOptions()
try options.setLogSeverityLevel(ORTLoggingLevel.warning)
try options.setIntraOpNumThreads(threadCount)
try options.appendExecutionProvider("xnnpack",providerOptions: ["intra_op_num_threads" : "2"])
let session = try ORTSession(env: env, modelPath: modelPath, sessionOptions: options)
the session get error,i have no idea ,why this happen, how can I use func appendExecutionProvider?

Urgency

No response

Platform

iOS

OS Version

iOS 14.4

ONNX Runtime Installation

Other / Unknown

Compiler Version (if 'Built from Source')

No response

Package Name (if 'Released Package')

None

ONNX Runtime Version or Commit ID

onnxruntime-objc 1.5.0

ONNX Runtime API

Objective-C/Swift

Architecture

Other / Unknown

Execution Provider

Default CPU

Execution Provider Library Version

onnxruntime-objc 1.5.0

[edgchen1]

What's the error message?

[taocheng]

sorry the parameters "xnnpack" should be"XNNPACK",that is my fault, another question I want switch cpu to gpu, it can be achieve?

[skottmckay]

On iOS you could try the CoreML execution provider to use the Apple Neural Engine. We don't have an execution provider that uses Metal unfortunately. Will depend on the model as to how much of it CoreML can run.