Any possibility of supporting transpose operation?

[ppershing]

According to https://coral.ai/docs/edgetpu/models-intro/#supported-operations coral does not seem to support Transpose op. I would presume that under some limiting circumstances (e.g., not transposing batch dim, etc.) the device should be able to support this. And the operation would be very helpful -- given size limitations, using grouped convolutions might be an easy way to get better accuracy. However, such architectures usually involve a channel "shuffle" so that channels from different groups can get mixed and such shuffle is usually implemented by reshape + transpose + reshape.

[nesherofek]

+1

[Namburger]

Hi guys, I do have a feature request opened for this internally, will keep update if we move forward with it.

[nesherofek]

For any one face this issue
by converting pytorch to keras (using this rep https://github.com/nerox8664/pytorch2keras ) i was able to fully convert my model to NHWC and by that avoiding the TRANSPOE nodes, edge tpu comile was able to map all operations to TPU (:

[MekhailS]

+1, badly need this feature

[manoj7410]

Compiler version 16 supports 'Transpose' operations. Please see the list of supported operations at https://coral.ai/docs/edgetpu/models-intro/#supported-operations.