commit acb3ec51e51542d3011ed87842f87c2261abaaff
author Teresa Charlin <teresa.charlinreyes@arm.com> Mon Apr 03 19:57:00 2023 +0100
committer Colm Donelan <colm.donelan@arm.com> Tue Apr 18 17:27:41 2023 +0000
tree b1ed73756c1db4a8e71b18a5a8256f42bb49341b
parent 8294e96a2f0f4ad3f5cd261079a6f90eee40142c

GitHub #719 Set quantization parameter scale to 1.0, instead of 0.0.

* Arm NN does not account for int8 or uint8 not quantized types, Tensorflow does.
Not quantized int8 and uint8 is the same as quantized int8 and uint8 with scale = 1.0 and offset= 0
Default offset/zero_point was already 0, this review sets the default scale to 1.0.

Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ibc3eecc281de516c2cc706e17bde01c64ff9556e

src/armnnTfLiteParser/TfLiteParser.cpp

diff --git a/src/armnnTfLiteParser/TfLiteParser.cpp b/src/armnnTfLiteParser/TfLiteParser.cpp
index c787212..ee4cadd 100644
--- a/src/armnnTfLiteParser/TfLiteParser.cpp
+++ b/src/armnnTfLiteParser/TfLiteParser.cpp
@@ -524,7 +524,7 @@
         }
     }
 
-    float quantizationScale = 0.0f;
+    float quantizationScale = 1.0f;
     int32_t quantizationOffset = 0;
 
     if (tensorPtr->quantization.get())