Please provide us with the following information:

This issue is for a: (mark with an x)

- [ ] bug report -> please search issues before submitting
- [x] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)

Minimal steps to reproduce

N/A

Any log messages given by the failure

N/A

Expected/desired behavior

N/A

OS and Version?

Windows 7, 8 or 10. Linux (which distribution). macOS (Yosemite? El Capitan? Sierra?)

Mention any other details that might be useful

---

Thanks! We'll be in touch soon.

This issue is for a: (mark with an x)

- [x] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)

Minimal steps to reproduce

1. Create new project with domain General (compact) or General (compact) S1
2. Train model using Quick Training
3. Export TensorFlow Lite model
4. Test model using the sample code in this repo https://github.com/Azure-Samples/customvision-export-samples/blob/main/samples/python/tensorflow_lite/classification/predict.py

Any log messages given by the failure

This is the error provided:

Traceback (most recent call last):
  File "predict.py", line 47, in <module>
    main()
  File "predict.py", line 41, in main
    model = Model(args.model_filepath)
  File "predict.py", line 10, in __init__
    self.interpreter = tensorflow.lite.Interpreter(model_path=str(model_filepath))
  File "/Users/{username}/anaconda3/lib/python3.8/site-packages/tensorflow/lite/python/interpreter.py", line 463, in __init__
    _interpreter_wrapper.CreateWrapperFromFile(
ValueError: Model provided has model identifier '

I also tried adjusting the code to use the tflite runtime instead and it produced the same error:

Traceback (most recent call last):
  File "predict.py", line 47, in <module>
    main()
  File "predict.py", line 41, in main
    model = Model(args.model_filepath)
  File "predict.py", line 10, in __init__
    self.interpreter = tflite.Interpreter(model_path=str(model_filepath))
  File "/Users/{username}/anaconda3/lib/python3.8/site-packages/tflite_runtime/interpreter.py", line 348, in __init__
    _interpreter_wrapper.CreateWrapperFromFile(
ValueError: Model provided has model identifier '

Expected/desired behavior

Output of predicted tag

OS and Version?

macOS 10.15.7
python 3.8.12
tensorflow 2.7.0
tflite-runtime 2.5.0.post1

Please provide us with the following information:

This issue is for a: (mark with an x)

- [ ] bug report -> please search issues before submitting
- [ X] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)

Do you have any example for Android and custom S1 model TFLite?

Please provide us with the following information:

This issue is for a: (mark with an x)

- [ ] bug report -> please search issues before submitting
- [ ] feature request
- [x ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)

OS and Version?

Windows10

Hi everyone,

my question/remark is two-folded and concerns the provided predict.py functions for different model types, tasks and domains for python(find here: https://github.com/Azure-Samples/customvision-export-samples/tree/main/samples/python):

1. Which is the correct color space of images used for inference with exported models from custom vision?

When i download models, the metadata_properties.json always says that the pixelformat is RGB8. The predict functions on the other hand (almost - see 2nd point) all contain a convertion to BGR like: input_array = input_array[:, :, :, (2, 1, 0)] # => BGR.
Why is that and which color space/pixelformat is used for training?

2. The mentiond color space conversion is not part of the predict function for the following combination: model type ->tensorflow, task->classification, domain -> General compact [S1] (find here: https://github.com/Azure-Samples/customvision-export-samples/blob/main/samples/python/tensorflow/classification_s1/predict.py).
   Why is that? Or is it just missing?

Thanks alot in advance

Uncaught (in promise) TypeError: webglBackend.incRef is not a function
at Object.reshape [as kernelFunc] (Reshape.ts:50)
at kernelFunc (engine.ts:576)
at engine.ts:634
at Engine.scopedRun (engine.ts:439)
at Engine.runKernelFunc (engine.ts:631)
at reshape_ (reshape.ts:71)
at reshape__op (operation.ts:50)
at expandDims_ (expand_dims.ts:56)
at expandDims__op (operation.ts:50)
at Tensor.push../node_modules/@tensorflow/tfjs/node_modules/@tensorflow/tfjs-core/dist/public/chained_ops/expand_dims.js.tensor__WEBPACK_IMPORTED_MODULE_1_.Tensor.expandDims (expand_dims.ts:29)

Attempt to load model in react.js. Failed with message shown

Actually i just need the algorithm so calculate the output

This issue is for a: (mark with an x)

- [ ] bug report -> please search issues before submitting
- [ ] feature request
- [x] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)

Minimal steps to reproduce

Examples of Classification (C#) is no longer available

Any log messages given by the failure

Github error 404 not found

Expected/desired behavior

Code for Classification (C#)

Mention any other details that might be useful

It seems that the code has been removed or unavailable