crashed at TfLiteInterpreterCreate

[jucysoft]

• iPhone11 pro

• Podfile
  pod 'TensorFlowLiteSelectTfOps', '> 0.0.1-nightly'
  pod 'TensorFlowLiteSwift', '> 0.0.1-nightly', :subspecs => ['Metal']

• crashed at TfLiteInterpreterCreate.
  My custom model is used at StyleTranfer demo project.

guard let cInterpreter = TfLiteInterpreterCreate(model.cModel, cInterpreterOptions) else {
--> Thread 8: EXC_BAD_ACCESS (code=1, address=0x0)

Thread 8 Queue : org.tensorflow.examples.lite.style_transfer (serial)
#0 0x00000001f0bce618 in _platform_memmove ()
#1 0x00000001007d406c in ___lldb_unnamed_symbol7040$$TFL Style Transfer ()
#2 0x00000001007c82e4 in ___lldb_unnamed_symbol6806$$TFL Style Transfer ()
#3 0x00000001007d0f90 in ___lldb_unnamed_symbol6968$$TFL Style Transfer ()
#4 0x00000001007c3744 in ___lldb_unnamed_symbol6746$$TFL Style Transfer ()
#5 0x00000001007c4304 in ___lldb_unnamed_symbol6748$$TFL Style Transfer ()
#6 0x00000001007b94bc in ___lldb_unnamed_symbol6574$$TFL Style Transfer ()
#7 0x0000000100429cfc in ___lldb_unnamed_symbol129$$TFL Style Transfer ()
#8 0x0000000100429924 in ___lldb_unnamed_symbol128$$TFL Style Transfer ()
#9 0x0000000100429678 in ___lldb_unnamed_symbol127$$TFL Style Transfer ()
#10 0x00000001007b9420 in ___lldb_unnamed_symbol6573$$TFL Style Transfer ()
#11 0x000000010042cc80 in ___lldb_unnamed_symbol169$$TFL Style Transfer ()
#12 0x00000001005638fc in ___lldb_unnamed_symbol2108$$TFL Style Transfer ()
#13 0x0000000100524554 in ___lldb_unnamed_symbol1773$$TFL Style Transfer ()
#14 0x000000010052431c in TfLiteInterpreterCreate ()
#15 0x000000010040e854 in Interpreter.init(modelPath:options:delegates:) at /Users/user/git/examples-master/lite/examples/style_transfer/ios/Pods/TensorFlowLiteSwift/tensorflow/lite/experimental/swift/Sources/Interpreter.swift:98
#16 0x000000010040e144 in Interpreter.__allocating_init(modelPath:options:delegates:) ()
#17 0x00000001003fe3a0 in closure #1 in static Inpainting.newInstance(inpaintingModel:useMetalDelegate:completion:) at /Users/user/git/examples-master/lite/examples/style_transfer/ios/StyleTransfer/Inpainting.swift:97
#18 0x00000001003fee34 in thunk for @escaping @callee_guaranteed () -> () ()
#19 0x000000010096bb68 in _dispatch_call_block_and_release ()
#20 0x000000010096d5f0 in _dispatch_client_callout ()
#21 0x0000000100974fa8 in _dispatch_lane_serial_drain ()
#22 0x0000000100975cb4 in _dispatch_lane_invoke ()
#23 0x0000000100981e38 in _dispatch_workloop_worker_thread ()
#24 0x00000001f0bd4908 in _pthread_wqthread ()

[jvishnuvardhan]

@jucysoft Can you please create a simple standalone code to reproduce the issue? Thanks!

[google-ml-butler]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you.

[sdgbraincraft]

this issue still persists. please see this project Link. I cloned the project from examples.

[jucysoft]

When a tflite model includes a TransposeConv node, this issue happens. Could you test it?

[jucysoft]

This issue was resolved at nightly build version by khanh.

[nimishpatel19]

Still it is crashing at my end. @jucysoft can you guide me how to resolve this?

[jucysoft]

I just used the nightly build version of at that time. no other resolution. Thank you,

…

2021. 5. 22. 오후 2:17, Nimish Patel ***@***.***> 작성:  Still it is crashing at my end. @jucysoft can you guide me how to resolve this? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[nimishpatel19]

It is crashing here.

pod 'TensorFlowLiteSwift', '~> 0.0.1-nightly'

I have used this. Is it fine @jucysoft?

[nimishpatel19]

this issue still persists. please see this project Link. I cloned the project from examples.

Did you solve this issue?

[jucysoft]

I ran the my old project with nightly version, but current nightly version reproduced this error. so, I changed pod file like following. It resolved this error. pod 'TensorFlowLiteSwift', '~> 2.3.0', :subspecs => ['Metal']

…

2021. 5. 22. 오후 3:20, Nimish Patel ***@***.***> 작성:  this issue still persists. please see this project Link. I cloned the project from examples. Did you solve this issue? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[nimishpatel19]

I ran the my old project with nightly version, but current nightly version reproduced this error. so, I changed pod file like following. It resolved this error. pod 'TensorFlowLiteSwift', '~> 2.3.0', :subspecs => ['Metal']
…
2021. 5. 22. 오후 3:20, Nimish Patel @.***> 작성:  this issue still persists. please see this project Link. I cloned the project from examples. Did you solve this issue? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

Thanks. Let me try it right now.

[araii]

I ran the my old project with nightly version, but current nightly version reproduced this error. so, I changed pod file like following. It resolved this error. pod 'TensorFlowLiteSwift', '~> 2.3.0', :subspecs => ['Metal']
…
2021. 5. 22. 오후 3:20, Nimish Patel @.***> 작성:  this issue still persists. please see this project Link. I cloned the project from examples. Did you solve this issue? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

....
hi, I'm facing the same EXC_BAD_ACCESS error

and I tried changing the pod file, but there is new error. Lines of
" TensorFlow Lite Error: Unsupported data type 13 in tensor" followed by
"Failed to create the interpreter with error: Failed to create the interpreter."

any help? did changing the pod file work you?

[nimishpatel19]

I ran the my old project with nightly version, but current nightly version reproduced this error. so, I changed pod file like following. It resolved this error. pod 'TensorFlowLiteSwift', '~> 2.3.0', :subspecs => ['Metal']
…
2021. 5. 22. 오후 3:20, Nimish Patel @.***> 작성:  this issue still persists. please see this project Link. I cloned the project from examples. Did you solve this issue? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

....
hi, I'm facing the same EXC_BAD_ACCESS error

and I tried changing the pod file, but there is new error. Lines of
" TensorFlow Lite Error: Unsupported data type 13 in tensor" followed by
"Failed to create the interpreter with error: Failed to create the interpreter."

any help? did changing the pod file work you?

Nope. But I changed the approch of implementation. I was using custom code to classify. But now I am using TensorFlowLiteTaskText for Text classification in project like this,

pod 'TensorFlowLiteSwift'
pod 'TensorFlowLiteTaskText', '~> 0.0.1-nightly'

Tensorflow released TensorFlowLiteTaskText prereleased version.
Check this examples here
https://github.com/tensorflow/examples/tree/master/lite/examples
I hope this might be useful for you.
Thanks

[mralihaidar]

how to resolve this issue