[GitHub] [tvm] tmoreau89 commented on a change in pull request #8952: [TVM] Add importer for ONNX QLinearMatMul op

[GitBox <gi...@apache.org>]

tmoreau89 commented on a change in pull request #8952:
URL: https://github.com/apache/tvm/pull/8952#discussion_r703748297



##########
File path: python/tvm/relay/frontend/onnx.py
##########
@@ -3333,6 +3333,44 @@ def get_scalar(x, dtype="float32"):
         return _qnn.op.quantize(out, c_scale, c_zero_point, out_dtype=dtype)
 
 
+class QLinearMatMul(OnnxOpConverter):
+    """Operator converter for QLinearMatMul from Microsoft onnxruntime contrib opset."""
+
+    @classmethod
+    def _impl_v10(cls, inputs, attr, params):
+        def get_scalar(x, dtype="float32"):
+            if isinstance(x, _expr.Var) and x.name_hint in params:
+                return _op.const(params[x.name_hint].numpy(), dtype)
+            rank = len(infer_shape(x))
+            assert rank <= 1, "QLinearMatMul scale and zero_point input must be scalars"
+            if rank == 1:
+                x = _op.squeeze(x, [0])
+            return _op.cast(x, dtype)
+
+        a = inputs[0]
+        a_scale = get_scalar(inputs[1])
+        a_zero_point = get_scalar(inputs[2], "int32")
+
+        b = inputs[3]
+        b_scale = get_scalar(inputs[4])
+        b_zero_point = get_scalar(inputs[5], "int32")
+
+        y_scale = fold_constant(get_scalar(inputs[6]))
+        y_zero_point = get_scalar(inputs[7], "int32")
+
+        dtype = infer_type(a).checked_type.dtype
+
+        a_rank = len(infer_shape(a))
+        b_rank = len(infer_shape(b))
+
+        assert ((a_rank == 2) and (b_rank == 2)), "QLinearMatMul importer currently requires both 'a' and 'b' tensors to be 2D, but rank(a)={}, rank(b)={}".format(a_rank, b_rank)

Review comment:
       @cconvey thanks for this first PR! Could you run the black formatter and linter on these changes, seems like some lines need to be reformatted (e.g. the one above). Thank you!




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org