[ONNX] Wrap 'If' if it has multiple outputs (apache#8385)

* [ONNX] Wrap 'If' if it has multiple outputs

Without this wrapper, an assertion in from_onnx() will fail with the
error message showing ""Number of output mismatch"

* [ONNX] Test If nodes with multiple output tensors

* Fix formatting issues

python/tvm/relay/frontend/onnx.py

@@ -2879,7 +2879,10 @@ def _impl_v1(cls, inputs, attr, params):
             graph_scope._nodes.update({var.name_hint: var})
 
         # Now we can construct the relay if statement and return.
-        return _expr.If(cond, then_expr, else_expr)
+        ret = _expr.If(cond, then_expr, else_expr)
+        if len(then_branch.output) > 1:
+            ret = _expr.TupleWrapper(ret, len(then_branch.output))
+        return ret
 
 
 class NonMaxSuppression(OnnxOpConverter):

tests/python/frontend/onnx/test_forward.py

@@ -4108,29 +4108,41 @@ def test_loop():
     verify_tensor_loop()
 
 
-def verify_if(cond_array):
+def verify_if(cond_array, num_outputs):
     # Given a bool scalar input cond.
     # return constant tensor x if cond is True, otherwise return constant tensor y.
-    then_out = onnx.helper.make_tensor_value_info("then_out", onnx.TensorProto.FLOAT, [5])
-    else_out = onnx.helper.make_tensor_value_info("else_out", onnx.TensorProto.FLOAT, [5])
 
-    x = np.array([1, 2, 3, 4, 5]).astype(np.float32)
-    y = np.array([5, 4, 3, 2, 1]).astype(np.float32)
+    def append_constant_nodes(nodes, outputs, expected, name):
+        outputs.append(onnx.helper.make_tensor_value_info(name, onnx.TensorProto.FLOAT, [5]))
 
-    then_const_node = onnx.helper.make_node(
-        "Constant", inputs=[], outputs=["then_out"], value=numpy_helper.from_array(x)
-    )
+        expected.append(np.random.randn(5).astype("float32"))
 
-    else_const_node = onnx.helper.make_node(
-        "Constant", inputs=[], outputs=["else_out"], value=numpy_helper.from_array(y)
-    )
+        nodes.append(
+            onnx.helper.make_node(
+                "Constant", inputs=[], outputs=[name], value=numpy_helper.from_array(expected[-1])
+            )
+        )
+
+    if_outputs = []
+    graph_outputs = []
 
-    then_body = onnx.helper.make_graph([then_const_node], "then_body", [], [then_out])
+    then_nodes, then_outs, then_expected = [], [], []
+    else_nodes, else_outs, else_expected = [], [], []
 
-    else_body = onnx.helper.make_graph([else_const_node], "else_body", [], [else_out])
+    for i in range(num_outputs):
+        append_constant_nodes(then_nodes, then_outs, then_expected, "then_out{}".format(i))
+        append_constant_nodes(else_nodes, else_outs, else_expected, "else_out{}".format(i))
+
+        if_outputs.append("res{}".format(i))
+        graph_outputs.append(
+            onnx.helper.make_tensor_value_info("res{}".format(i), onnx.TensorProto.FLOAT, [5]),
+        )
+
+    then_body = onnx.helper.make_graph(then_nodes, "then_body", [], then_outs)
+    else_body = onnx.helper.make_graph(else_nodes, "else_body", [], else_outs)
 
     if_node = onnx.helper.make_node(
-        "If", inputs=["cond"], outputs=["res"], then_branch=then_body, else_branch=else_body
+        "If", inputs=["cond"], outputs=if_outputs, then_branch=then_body, else_branch=else_body
     )
 
     if_graph = onnx.helper.make_graph(
@@ -4139,31 +4151,33 @@ def verify_if(cond_array):
         inputs=[
             onnx.helper.make_tensor_value_info("cond", onnx.TensorProto.BOOL, []),
         ],
-        outputs=[
-            onnx.helper.make_tensor_value_info("res", onnx.TensorProto.FLOAT, [5]),
-        ],
+        outputs=graph_outputs,
     )
 
     if_model = onnx.helper.make_model(if_graph)
     if cond_array:
         cond = np.array([1]).astype("bool")
     else:
         cond = np.array(1).astype("bool")
-    correct_out = x if cond else y
+    correct_out = then_expected if cond else else_expected
 
     # TODO(jwfromm): Onnxruntime 1.0.0 is buggy with If statements. Replace this with
     # verify_with_ort once we update versions.
     for target, dev in tvm.testing.enabled_targets():
         tvm_out = get_tvm_output_with_vm(if_model, [cond], target, dev, freeze_params=True)
+        if not isinstance(tvm_out, list):
+            tvm_out = [tvm_out]
         for i in range(len(tvm_out)):
             tvm.testing.assert_allclose(correct_out[i], tvm_out[i], rtol=1e-05, atol=1e-05)
 
 
 @tvm.testing.uses_gpu
 def test_if():
     # Confirm that if works with cond as an array or scalar.
-    verify_if(cond_array=False)
-    verify_if(cond_array=True)
+    verify_if(cond_array=False, num_outputs=1)
+    verify_if(cond_array=False, num_outputs=2)
+    verify_if(cond_array=True, num_outputs=1)
+    verify_if(cond_array=True, num_outputs=2)
 
 
 @tvm.testing.uses_gpu