diff options
| author | David Monahan <david.monahan@arm.com> | 2019-05-30 12:07:44 +0100 |
|---|---|---|
| committer | David Monahan <david.monahan@arm.com> | 2019-05-30 12:07:44 +0100 |
| commit | d5bfae1fb06dd2f1227a7518f4951de10874e71a (patch) | |
| tree | 0633ddb86a0c5faa1073fa4a9e428b673ca2684f /NnapiSupport.txt | |
| parent | acf479a1c3841b3036dc63dd84f1d57d29f94f64 (diff) | |
| download | android-nn-driver-d5bfae1fb06dd2f1227a7518f4951de10874e71a.tar.gz | |
IVGCVSW-3146 Update to NNapiSupport.txt to indicate support for
Dequantize, added Dequantize to vector of equivalent 1.0 ops in
1.1/HalPolicy.cpp
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: I4ffe7467ac47dd4b5ebee11d23c960e4346e2f1f
Diffstat (limited to 'NnapiSupport.txt')
| -rw-r--r-- | NnapiSupport.txt | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/NnapiSupport.txt b/NnapiSupport.txt index 7dfb6dd0..314111dd 100644 --- a/NnapiSupport.txt +++ b/NnapiSupport.txt @@ -19,6 +19,7 @@ CONCATENATION (FLOAT32,QUANT8_ASYMM,QUANT16_SYMM) CONV_2D (FLOAT32,QUANT8_ASYMM,QUANT16_SYMM) DEPTHWISE_CONV_2D (FLOAT32,QUANT8_ASYMM,QUANT16_SYMM) DIV (FLOAT32,QUANT8_ASYMM,QUANT16_SYMM) +DEQUANTIZE (FLOAT32,QUANT8_ASYMM,QUANT16_SYMM) FLOOR (FLOAT32) FULLY_CONNECTED (FLOAT32,QUANT8_ASYMM) L2_NORMALIZATION (FLOAT32) @@ -49,7 +50,6 @@ TRANSPOSE (FLOAT32,QUANT8_ASYMM) The following AndroidNN 1.0 operations are currently not supported. DEPTH_TO_SPACE -DEQUANTIZE EMBEDDING_LOOKUP HASHTABLE_LOOKUP LSH_PROJECTION |