Closed bknutson77 closed 3 years ago
Thanks for the feedback, and sorry for late response. I will take a look at it today.
Thanks for the feedback, and sorry for late response. I will take a look at it today.
No worries! Just thought I'd let you know. This stuff is constantly changing.. I tried creating a model with Google AutoML and exported that to TFJS, used that in combination with the library of TFJS-AutoML and it worked. Custom Vision might be exporting a totally different, new idea of the model and it may have changed on ya.
Hi,
I've tried 3 models (Classification, non-S1 Object Detection, S1 Object Detection) with the sample scripts but I couldn't reproduce the issue. https://github.com/Azure-Samples/customvision-export-samples/tree/main/samples/javascript/tensorflowjs
Could you provide more details about the issue?
This is strange.. when I used the code from here: https://github.com/Azure-Samples/customvision-export-samples/tree/main/samples/javascript/tensorflowjs it works, but if I used your repo directly it was giving me the error.. I'm not sure what's going on but I got it to work with the model, cool!
Literally everything is broken.. try it out, the tfjs models being exported by custom vision are completely different than what your code is expecting. Specifically the postprocessing method is trying to index into the model, but the model's structure is different and so it tries to get index 0 of null at some point.