1 files changed, 33 insertions, 0 deletions
diff --git a/src/armnn/NetworkQuantizationScheme.hpp b/src/armnn/NetworkQuantizationScheme.hpp
index 0effa1fd64..ea3c29102b 100644
--- a/src/armnn/NetworkQuantizationScheme.hpp
+++ b/src/armnn/NetworkQuantizationScheme.hpp
@@ -61,6 +61,34 @@ struct QAsymm8QuantizationScheme : IQuantizationScheme
     DataType GetDataType() const override { return DataType::QuantisedAsymm8; }
 };
 
+struct QSymmS8QuantizationScheme : IQuantizationScheme
+{
+    OffsetScalePair ComputeScheme(double min, double max) const override
+    {
+        if (min > max)
+        {
+            throw InvalidArgumentException("min > max will result in invalid quantization.");
+        }
+
+        // To avoid dividing by zero when quantizing a zero filled tensor
+        if (min == 0.0 && max == 0.0)
+        {
+            max = 1.0;
+        }
+
+        double highest = (1 << (NumBits()-1)) - 1; // (numbits-1) accounts for the sign bit
+
+        double extent = std::max(std::abs(min), std::abs(max));
+        double scale = extent / highest;
+
+        return std::make_pair(static_cast<float>(scale), 0);
+    }
+
+    int NumBits() const override { return 8; }
+
+    DataType GetDataType() const override { return DataType::QSymmS8; }
+};
+
 struct QSymm16QuantizationScheme : IQuantizationScheme
 {
     OffsetScalePair ComputeScheme(double min, double max) const override
@@ -81,7 +109,12 @@ struct QSymm16QuantizationScheme : IQuantizationScheme
         double extent = std::max(std::abs(min), std::abs(max));
         double scale = extent / highest;
 
+        if(scale == 0.000457777642)
+        {
+            return std::make_pair(static_cast<float>(scale), 0);
+        }
         return std::make_pair(static_cast<float>(scale), 0);
+
     }
 
     int NumBits() const override { return 16; }