diff --git a/python/tvm/relay/frontend/pytorch.py b/python/tvm/relay/frontend/pytorch.py
index 683b94dd9290..9ddd04b5b4ee 100644
--- a/python/tvm/relay/frontend/pytorch.py
+++ b/python/tvm/relay/frontend/pytorch.py
@@ -4424,7 +4424,7 @@ def _create_typed_const(data, dtype):
     dtype should be a TVM dtype"""
 
     if dtype == "float64":
-        typed_data = _expr.const(np.float64(data), dtype=dtype)
+        typed_data = _expr.const(np.asarray(data, dtype="float64"), dtype=dtype)
     elif dtype == "float32":
         typed_data = _expr.const(np.float32(data), dtype=dtype)
     elif dtype == "float16":
diff --git a/tests/python/frontend/pytorch/test_forward.py b/tests/python/frontend/pytorch/test_forward.py
index 8c1cdbb0cf0b..9ee03512e7ae 100644
--- a/tests/python/frontend/pytorch/test_forward.py
+++ b/tests/python/frontend/pytorch/test_forward.py
@@ -3449,6 +3449,20 @@ def test_forward_adaptive_max_pool1d():
     verify_model(m.float().eval(), input_data=input_data)
 
 
+@tvm.testing.uses_gpu
+def test_forward_instance_norm():
+    """test_forward_instance_norm"""
+
+    class instance_norm(Module):
+        def forward(self, *args):
+            return torch.nn.functional.instance_norm(args[0], use_input_stats=True)
+
+    m = instance_norm().float().eval()
+    input_data = torch.randn([1, 1, 1, 2], dtype=torch.float64)
+
+    verify_model(m.float().eval(), input_data=input_data)
+
+
 @tvm.testing.uses_gpu
 def test_forward_full_like():
     """test_forward_full_like"""