Currently the nn-imx (1.1.9) only support nnapi 1.2 version (Android api 29).
When will nn-imx support nnapi 1.3 (android api 30) ?
Since nnapi1.2 have so many unsupported implementation.
Many tensorflow lite neutral network model are fall back to CPU.
Noticeably Resize(with align_corners enabled), ANEURALNETWORKS_TENSOR_QUANT8_ASYMM_SIGNED etc.
Which makes the NPU running so slow, since many layers are fallback to CPU.
