[quant][fx] Remove input_output_observed from BinaryOpQuantizeHandler

jerryzh168 · jerryzh168 · commit 80492796d0d1 · 2022-03-25T13:10:25.000-07:00
Summary: when both inputs are scalars, fx tracing will directly calculate the result, instead of generating an op in the fx graph so num_tensor_args will always be greater than 1 for binary ops, so the input_output_observed will always return True for BinaryQuantizeHandler We will remove input_output_observed method after dynamic quantization in qconfig is properly supported Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: 814441c Pull Request resolved: #74776
diff --git a/torch/ao/quantization/fx/quantization_patterns.py b/torch/ao/quantization/fx/quantization_patterns.py
@@ -199,10 +199,6 @@ def __init__(
     def is_general_tensor_value_op(self) -> bool:
         return self.num_tensor_args == 1
 
-    def input_output_observed(self):
-        # for x + y where x and y are scalars, we do not observe anything
-        return self.num_tensor_args > 0
-
     def is_output_quantized(self, qconfig):
         dtypes = get_qconfig_dtypes(qconfig)
         return self.binary_op in binary_op_supported_dtypes and \
