diff options
author | Georgios Pinitas <georgios.pinitas@arm.com> | 2017-07-12 12:30:40 +0100 |
---|---|---|
committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-09-17 14:16:42 +0100 |
commit | e222941f931b1d44bb29e5827b6df748e60cefc4 (patch) | |
tree | 1a54a60750caa998527e5c6b2a494047cb3ccc23 /tests/validation/Reference.h | |
parent | bbd3d6045ba6480be350c0b1610048ee6c9c050f (diff) | |
download | ComputeLibrary-e222941f931b1d44bb29e5827b6df748e60cefc4.tar.gz |
COMPMID-401: Implement FixedPointPosition conversion for NEON.
Adds support of changing the fixed point position of a tensor in
DepthConvert.
Change-Id: Ic3b50a4628fac7497a0217d92941c9d6f64d21cb
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/80438
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Diffstat (limited to 'tests/validation/Reference.h')
-rw-r--r-- | tests/validation/Reference.h | 16 |
1 files changed, 9 insertions, 7 deletions
diff --git a/tests/validation/Reference.h b/tests/validation/Reference.h index 3ad9814439..a3ae3b6d02 100644 --- a/tests/validation/Reference.h +++ b/tests/validation/Reference.h @@ -168,16 +168,18 @@ public: static RawTensor compute_reference_box3x3(const TensorShape &shape, BorderMode border_mode, uint8_t constant_border_value); /** Compute reference depth convert. * - * @param[in] shape Shape of the input and output tensors. - * @param[in] dt_in Data type of input tensor. - * @param[in] dt_out Data type of the output tensor. - * @param[in] policy Overflow policy of the operation. - * @param[in] shift Value for down/up conversions. Must be 0 <= shift < 8. - * @param[in] fixed_point_position Fixed point position. + * @param[in] shape Shape of the input and output tensors. + * @param[in] dt_in Data type of input tensor. + * @param[in] dt_out Data type of the output tensor. + * @param[in] policy Overflow policy of the operation. + * @param[in] shift Value for down/up conversions. Must be 0 <= shift < 8. + * @param[in] fixed_point_position_in (Optional) Fixed point position for the input tensor. + * @param[in] fixed_point_position_out (Optional) Fixed point position for the output tensor. * * @return Computed raw tensor. */ - static RawTensor compute_reference_depth_convert(const TensorShape &shape, DataType dt_in, DataType dt_out, ConvertPolicy policy, uint32_t shift, uint32_t fixed_point_position); + static RawTensor compute_reference_depth_convert(const TensorShape &shape, DataType dt_in, DataType dt_out, ConvertPolicy policy, + uint32_t shift, uint32_t fixed_point_position_in = 0, uint32_t fixed_point_position_out = 0); /** Compute reference gaussian3x3 filter. * * @param[in] shape Shape of the input and output tensors. |