Using DirectML on WSL2 to accelerate Tensorflow Lite (tflite)

[cminnoy]

Hi,

Tensorflow lite can make use of GPU on Android and iOS.
While Tensorflow can use DirectML do to training, I would like to use Tensorflow Lite for inferencing, also with DirectML GPU acceleration under WSL2 if possible.
Any ideas if this is possible?

Cheers,

Chris