diff options
| author | giuros01 <giuseppe.rossini@arm.com> | 2019-07-24 16:29:53 +0100 |
|---|---|---|
| committer | Giuseppe Rossini <giuseppe.rossini@arm.com> | 2019-07-24 21:21:50 +0000 |
| commit | b5e75dbc1543ad7f549bf3351ac753614fca25e4 (patch) | |
| tree | c8cc96219f427571ade5dfc4d08c0c583331d697 /arm_compute/runtime | |
| parent | 7faa0ec6d74883f29924efaf86c15a9634fdec64 (diff) | |
| download | ComputeLibrary-b5e75dbc1543ad7f549bf3351ac753614fca25e4.tar.gz | |
COMPMID-2491: (Nightly) Bug in NEON LSTMLayerQuantized
Bug fix: We had to introduce a +1/-1 tolerance for 32-bit.
This is already done for other layers on NEON (e.g., Activation)
Change-Id: I449cd3c394ccdb46517f26ceff5d1254734a476a
Signed-off-by: giuros01 <giuseppe.rossini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/1607
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: VidhyaSudhan Loganathan <vidhyasudhan.loganathan@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/runtime')
0 files changed, 0 insertions, 0 deletions