diff options
author | Lev Proleev <levp@google.com> | 2020-01-10 16:47:23 +0000 |
---|---|---|
committer | Viet Dang <vddang@google.com> | 2020-01-13 09:56:06 +0000 |
commit | e4c8f5fe23fdb8db4deeae3f921132985e355c36 (patch) | |
tree | 641484eefd0dc6ba0c57c8e0b3c1bc0071a05d12 /neuralnetworks | |
parent | 1fd6b723df993482a589548ccabec88beda6b362 (diff) | |
download | platform_hardware_interfaces-e4c8f5fe23fdb8db4deeae3f921132985e355c36.tar.gz platform_hardware_interfaces-e4c8f5fe23fdb8db4deeae3f921132985e355c36.tar.bz2 platform_hardware_interfaces-e4c8f5fe23fdb8db4deeae3f921132985e355c36.zip |
Fix quantized LSTM doc
The CL changes explicit prefixes for types in types.spec to a macro, so
that the prefixes of tensor types are correct both in NDK and HAL docs.
Bug: 144841609
Test: mma
Change-Id: I6d904bd3a858f555beed9270f141f080f96e429a
Merged-In: I6d904bd3a858f555beed9270f141f080f96e429a
(cherry picked from commit da5079c5cd6093cc6275e1f686b24a6a0c8a925c)
Diffstat (limited to 'neuralnetworks')
-rw-r--r-- | neuralnetworks/1.3/types.hal | 2 | ||||
-rw-r--r-- | neuralnetworks/1.3/types.t | 2 |
2 files changed, 3 insertions, 1 deletions
diff --git a/neuralnetworks/1.3/types.hal b/neuralnetworks/1.3/types.hal index 74c259ca22..62c5833500 100644 --- a/neuralnetworks/1.3/types.hal +++ b/neuralnetworks/1.3/types.hal @@ -4747,7 +4747,7 @@ enum OperationType : int32_t { RESIZE_NEAREST_NEIGHBOR = @1.2::OperationType:RESIZE_NEAREST_NEIGHBOR, /** - * Quantized version of {@link OperationType:LSTM}. + * Quantized version of {@link OperationType::LSTM}. * * The input and the output use asymmetric quantized types, while the rest * use symmetric ones. diff --git a/neuralnetworks/1.3/types.t b/neuralnetworks/1.3/types.t index e06f5d630d..0d20d06f7d 100644 --- a/neuralnetworks/1.3/types.t +++ b/neuralnetworks/1.3/types.t @@ -57,6 +57,8 @@ enum OperationType : int32_t { %insert Operation_1.2 +%insert Operation_1.3 + /** * DEPRECATED. Since NNAPI 1.2, extensions are the preferred alternative to * OEM operation and data types. |