nvidia awq only use QuantFormat.QDQ quant format (#22429)

nvidia awq only use QuantFormat.QDQ quant format
microsoft · Oct 15, 2024 · ec7aa63 · ec7aa63
1 parent 6e5e320
commit ec7aa63
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/onnxruntime/python/tools/quantization/matmul_4bits_quantizer.py b/onnxruntime/python/tools/quantization/matmul_4bits_quantizer.py
@@ -280,7 +280,7 @@ def __init__(
 
         super().__init__(
             algorithm="nvidia_awq",
-            quant_format=quant_format,
+            quant_format=QuantFormat.QDQ,
             op_types_to_quantize=None,  # Assuming op_types_to_quantize is handled elsewhere
             quant_axes=None,  # Assuming quant_axes is handled elsewhere
         )