Closed lilh9598 closed 1 year ago
Hi @lilh9598
The following patch solves the problem: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8326
Hope this helps
Hi @lilh9598
The following patch solves the problem: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8326
Hope this helps
Thanks, I get the right result in my test with your code.
Output of 'strings libarm_compute.so | grep arm_compute_version': arm_compute_version=v22.05 Build options: {'Werror': '0', 'debug': '0', 'neon': '1', 'opencl': '0', 'embed_kernels': '0', 'os': 'linux', 'arch': 'armv8a', 'build': 'native', 'multi_isa': '1'} Git hash=b'a175e887d64450decf80ea47d4049832c5805565'
Platform: Neoverse-N1
Operating System: debian
Problem description: The output overflows when using the NEActivationLayer with TANH and F16. The following program can reproduce this bug.
The logs of this program on my platform is as follows:
I did some checks and there is a bug with tanhq_f16 I think.
https://github.com/ARM-software/ComputeLibrary/blob/aabef6c0584f06f4c0f4b61fb787d80374240619/src/core/NEON/NEMath.inl#L484-L497