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)

Minimal steps to reproduce

Any log messages given by the failure

Expected/desired behavior

OS and Version?

Windows 7, 8 or 10. Linux (which distribution). macOS (Yosemite? El Capitan? Sierra?)
iOS 12

Versions

Mention any other details that might be useful

I trained an object detection model from Custom Vision export to CoreML format. And the model output multidimensional array of doubles. And I don't know how to get the tagId, tagName, probability & bounding box? Are there any document or example to show how to do it? Thanks.

Is there a reason the CVSInference source code is not available? Normally examples and samples like this have all the code necessary to complete the objective. This allows people to learn and understand what is going on behind the scenes, so they can replicate it in their own code.

However, this is not the case here.

Is there a reason for this?

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)

Minimal steps to reproduce

Any log messages given by the failure

Expected/desired behavior

OS and Version?

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

Versions

Mention any other details that might be useful

When I export the object detection model the output that I get from xCode for the neural network is a two-dimensional multiArray. Can someone please provide documentation as to how we can extract the prediction values from this variable? Or can the model export be modified to clearly indicate outputs that are most valuable for the prediction output?

I threw my model in the sample code and it works great recognizing the three object in the frame. When I try to use the pod and send the model a UIImage taken from a AVCapturePhotoOutput I have very poor performance. Is there a trick to creating the UIImage from a camera that will make it work with your pod?

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)

Minimal steps to reproduce

Any log messages given by the failure

Expected/desired behavior

OS and Version?

macOS (Mojave)

Versions

10.14.6

Mention any other details that might be useful

I tried to replace the model with the object detection model, but seem it failed to do so. I saw there is a object detection feature in the pull request. However, it requires a Microsoft developer account to do the pod install. Therefore is there any other way to import a object detection model rather than just classification model?

Feature request
Hello ,
I checked this demo, and it's working fine. But I am working on an app where I have to recognize multiple person's faces at once on iOS 11. So here can we recognize multiple fruits at once?
Thanks in advance.