Weird behavior of tensorflow-metal on M1 Mac with Object Detection API

Recently days, I installed tensorflow-metal in my M1 MacBook Pro (macOS 12 beta 12.0) following this tutorial.

And then, I run TF2's Object Detection API script from this repository karaage0703/object_detection_tools.

It did great work with apple/tensorflow_macos. But with tensorflow-metal, the accuracies of objects are over 100%.

How can I solve this problem?

Thank you.

Up vote post of PotatoHotate Down vote post of PotatoHotate
747 views
Posted by
PotatoHotate
Reply
Add a Comment

Replies

Hi @PotatoHotate Thanks for reporting the issue. For the object detection tools, did you try the TF 2.x version listed on the website ? We don't support the V1 scripts.

Posted by
Frameworks Engineer
Up vote reply of Frameworks Engineer Down vote reply of Frameworks Engineer
Add a Comment

Thank you for reply.

I used this script and it supports TFv2.

Successful combination:

Failed combination:

  • TensorFlowMetal + EfficientNet for TPU

TensorFlow Metal does not supports TPU Model?

Model download scripts:

Posted by
PotatoHotate
Up vote reply of PotatoHotate Down vote reply of PotatoHotate
Add a Comment