Add an "is_signed" attribute to the quant_ConstFakeQuant op

Some TensorFlow simulated quantize ops such as QuantizeAndDequantizeV2Op have
attribute for the sign of the quantization, so quant_ConstFakeQuant should be
able to represent it with the new attribute is added.

The method for converting these attributes to an QuantizedType is updated to
handle this new argument.

PiperOrigin-RevId: 258810290

1 files changed, 1 insertions, 1 deletions

diff --git a/mlir/lib/Dialect/QuantOps/Transforms/ConvertSimQuant.cpp b/mlir/lib/Dialect/QuantOps/Transforms/ConvertSimQuant.cpp
index 0c93146a232..32d8c8a81c1 100644
--- a/mlir/lib/Dialect/QuantOps/Transforms/ConvertSimQuant.cpp
+++ b/mlir/lib/Dialect/QuantOps/Transforms/ConvertSimQuant.cpp
@@ -70,7 +70,7 @@ public:
     UniformQuantizedType uniformElementType = fakeQuantAttrsToType(
         fqOp.getLoc(), fqOp.num_bits().getSExtValue(),
         fqOp.min().convertToFloat(), fqOp.max().convertToFloat(),
-        fqOp.narrow_range(), converter.expressedType);
+        fqOp.narrow_range(), converter.expressedType, fqOp.is_signed());
 
     if (!uniformElementType) {
       // Note that the fakeQuantAttrsToType will have emitted the error.