tflite-micro
tflite-micro copied to clipboard
tensorflow
Adding support for Rfft2d
Hello everyone,
In the issue Can this be done in TFLM, it was mentionned that the kernels Rfft2d and ComplexAbs were not supported by TFLM. (kernels used in tflite for the audio classification model for instance).
I wanted to know if the reason of this is a technical incompatibility or if it's possible to implement by following the step-by-step guide?
Thanks!
TFLM doesn't currently support complex types, which would be needed for both Rfft2d and ComplexAbs. We do have our own signal library (https://github.com/tensorflow/tflite-micro/tree/main/signal), but it does mean models using those operations need to be re-authored using the TFLM signal library.
I too want to know if this op will be supported in future or it can be ported by following the guide.
"This issue is being marked as stale due to inactivity. Remove label or comment to prevent closure in 5 days."
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
"This issue is being closed because it has been marked as stale for 5 days with no further activity."
TFLM doesn't currently support complex types, which would be needed for both
Rfft2dandComplexAbs. We do have our own signal library (https://github.com/tensorflow/tflite-micro/tree/main/signal), but it does mean models using those operations need to be re-authored using the TFLM signal library.
I also want to know if TFLM will support complex types in the future? If not supported, what is the reason?
