[TF FE]: Support complex tensors for ScatterNd operation

src/frontends/tensorflow_common/src/op/scatter_nd.cpp

@@ -3,7 +3,9 @@
 //
 
 #include "common_op_table.hpp"
+#include "helper_ops/complex_type_mark.hpp"
 #include "openvino/op/broadcast.hpp"
+#include "openvino/op/concat.hpp"
 #include "openvino/op/scatter_nd_update.hpp"
 #include "utils.hpp"
 
@@ -15,10 +17,29 @@ namespace frontend {
 namespace tensorflow {
 namespace op {
 OutputVector translate_scatter_nd_op(const NodeContext& node) {
-    default_op_checks(node, 3, {"ScatterNd", "SCATTER_ND"});
+    default_op_checks(node, 3, {"ScatterNd", "SCATTER_ND"}, true);
     auto input_indices = node.get_input(0);
     auto updates = node.get_input(1);
     auto shape = node.get_input(2);
+    auto complex_type_mark = as_type_ptr<ComplexTypeMark>(updates.get_node_shared_ptr());
+
+    if (complex_type_mark) {
+        element::Type complex_part_type = complex_type_mark->get_complex_part_type();
+        updates = complex_type_mark->input_value(0);
+
+        auto new_dim = create_same_type_const<int32_t>(shape, vector<int32_t>{2}, Shape{1});
+        auto new_shape = make_shared<v0::Concat>(OutputVector{shape, new_dim}, -1);
+
+        auto const_zero = create_same_type_const<int32_t>(updates, vector<int32_t>{0}, Shape{1});
+        auto broadcast = make_shared<v3::Broadcast>(const_zero, new_shape);
+
+        auto complex_scatter_nd = make_shared<v3::ScatterNDUpdate>(broadcast, input_indices, updates);
+
+        set_node_name(node.get_name(), complex_scatter_nd);
+        auto complex_result = make_shared<ComplexTypeMark>(complex_scatter_nd, complex_part_type);
+
+        return {complex_result};
+    }
 
     auto input_data = create_same_type_const<int32_t>(updates, vector<int32_t>{0}, Shape{1});
     auto broadcast = make_shared<v3::Broadcast>(input_data, shape);

tests/layer_tests/tensorflow_tests/test_tf_ScatterND.py

@@ -3,9 +3,11 @@
 
 import platform
 
+import numpy as np
 import pytest
 from common.tf_layer_test_class import CommonTFLayerTest
 
+rng = np.random.default_rng(475912)
 
 class TestTFScatterND(CommonTFLayerTest):
     def create_tf_scatternd_placeholder_const_net(self, x_shape, indices, updates, ir_version,
@@ -73,3 +75,54 @@ def test_tf_scatter_nd(self, params, ie_device, precision, ir_version, temp_dir,
                                                                    use_legacy_frontend=use_legacy_frontend),
                    ie_device, precision, temp_dir=temp_dir, ir_version=ir_version,
                    use_legacy_frontend=use_legacy_frontend, **params)
+
+class TestTFScatterNDComplex(CommonTFLayerTest):
+    def _prepare_input(self, inputs_info):
+        assert 'param_real:0' in inputs_info, "Test error: inputs_info must contain `param_real`"
+        assert 'param_imag:0' in inputs_info, "Test error: inputs_info must contain `param_imag`"
+        updates_shape = inputs_info['param_real:0']
+        inputs_data = {}
+        inputs_data['param_real:0'] = rng.integers(-10, 10, updates_shape).astype(np.float32)
+        inputs_data['param_imag:0'] = rng.integers(-10, 10, updates_shape).astype(np.float32)
+
+        return inputs_data
+
+    def create_tf_scatternd_complex_placeholder_const_net(self, x_shape, indices, updates_shape, indices_type,
+                                                          ir_version, use_legacy_frontend):
+        import tensorflow as tf
+        tf.compat.v1.reset_default_graph()
+        with tf.compat.v1.Session() as sess:
+            param_real = tf.compat.v1.placeholder(tf.float32, updates_shape, 'param_real')
+            param_imag = tf.compat.v1.placeholder(tf.float32, updates_shape, 'param_imag')
+
+            tf_indices = tf.constant(indices, dtype=indices_type)
+            tf_shape = tf.constant(x_shape, dtype=indices_type)
+
+            complex = tf.raw_ops.Complex(real=param_real, imag=param_imag)
+
+            result = tf.scatter_nd(indices=tf_indices, updates=complex, shape=tf_shape, name="Operation")
+            tf.raw_ops.Real(input=result)
+            tf.raw_ops.Imag(input=result)
+
+            tf.compat.v1.global_variables_initializer()
+
+            tf_net = sess.graph_def
+
+        return tf_net, None
+
+    test_data = [
+        dict(x_shape=[8], indices=[[4], [3], [1], [7]], updates_shape=[4], indices_type=np.int32),
+        dict(x_shape=[10], indices=[[0], [2], [4], [6], [8]], updates_shape=[5], indices_type=np.int64),
+        dict(x_shape=[5, 5], indices=[[0, 0], [1, 1], [2, 2], [3, 3]], updates_shape=[4], indices_type=np.int64),
+        dict(x_shape=[3, 3, 3], indices=[[0, 0, 0], [1, 1, 1], [2, 2, 2]], updates_shape=[3], indices_type=np.int32),
+    ]
+
+    @pytest.mark.parametrize("params", test_data)
+    @pytest.mark.nightly
+    @pytest.mark.precommit
+    def test_tf_scatter_nd_complex(self, params, ie_device, precision, ir_version, temp_dir,
+                           use_legacy_frontend):
+        self._test(*self.create_tf_scatternd_complex_placeholder_const_net(**params, ir_version=ir_version,
+                                                                   use_legacy_frontend=use_legacy_frontend),
+                   ie_device, precision, temp_dir=temp_dir, ir_version=ir_version,
+                   use_legacy_frontend=use_legacy_frontend, **params)