[Good First Issue][TF FE]: Support complex tensors for ExpandDims operation

[rkazants]

Context

OpenVINO component responsible for support of TensorFlow models is called as TensorFlow Frontend (TF FE). TF FE converts a model represented in TensorFlow opset to a model in OpenVINO opset.
Some audio models use tensors of complex type. Complex type tensor is a tensor that has elements of complex type. For example, 1D tensor with three elements x = [1+2j, 2, -2j].

For supporting ExpandDims operation on complex type tensor, you need to extend the corresponding loader for ExpandDims.

What needs to be done?

The existing loader for ExpandDims needs to be extended by propagating ComplexTypeMark from input to output and to represent output complex type tensor as a floating-point type tensor with auxiliary dimension that concatenates real and imaginary parts of complex tensor.
To validate the extension, the corresponding layer test needs to be updated with complex tensor cases.

Here is an example of how to extend Reshape loader to support complex type tensors:

OutputVector translate_reshape_op(const NodeContext& node) {
    default_op_checks(node, 2, {"Reshape"}, true);
    auto tensor = node.get_input(0);
    auto complex_type_mark = as_type_ptr<ComplexTypeMark>(tensor.get_node_shared_ptr());
    auto shape = node.get_input(1);
    if (complex_type_mark) {
        element::Type complex_part_type = complex_type_mark->get_complex_part_type();
        tensor = complex_type_mark->input_value(0);

        OutputVector concat_inputs;
        concat_inputs.push_back(shape);
        concat_inputs.push_back(make_shared<v0::Constant>(shape.get_element_type(), Shape{1}, 2));

        auto concat = make_shared<v0::Concat>(concat_inputs, 0);
        auto reshape = make_shared<v1::Reshape>(tensor, concat, false);
        set_node_name(node.get_name(), reshape);
        auto complex_reshape = make_shared<ComplexTypeMark>(reshape, complex_part_type);
        return {complex_reshape->output(0)};
    }

    auto reshape = make_shared<v1::Reshape>(tensor, shape, false);
    set_node_name(node.get_name(), reshape);
    return {reshape};
}

Since OpenVINO does not have native support of complex tensors, we handle complex type in intermediate layers by representing them as a floating-point type with additional dimension (specially created) to store real and imaginary parts of the original complex tensor so slicing by the last dimension will give either real or imaginary parts: x[...,0] - real and x[...,1] - imaginary parts.

On the first step, we update default_op_checks with true flag to indicate that loader for Reshape operation now handles complex tensors:

default_op_checks(node, 2, {"Reshape"}, true);

Secondly, we check if complex type mark exists by anticipated inputs. This mark indicates that input tensor of complex type:

auto complex_type_mark = as_type_ptr<ComplexTypeMark>(tensor.get_node_shared_ptr());

Thirdly, we retrieve a floating-point tensor (with additional dimension to store real and imaginary parts) simulating complex tensor:

tensor = complex_type_mark->input_value(0);

After that, we implement conversion for Reshape for this particular case. Since a floating-point tensor simulating complex tensor has additional dimension equal to 2,
we update input target shape by appending 2 value and perform reshape on a floating-point tensor simulating complex tensor.

Finally, since Reshape should produce complex tensor by output we insert a new mark ComplexTypeMark into the output.

To validate support of complex tensors for Reshape, the new layer test TestComplexReshape was added.

Example how to run the layer test:

export TEST_DEVICE=CPU
cd openvino/tests/layer_tests/tensorflow_tests
pytest test_tf_Reshape.py

Example Pull Requests

• Extensions for Shape, Mul, Reshape in [TF FE] Complex type support extended for Separate Bass model. #21477
• Extension for Roll operation in [TF FE] Support Complex Tensors #20860

Resources

• What is OpenVINO?
• How to Build OpenVINO
• Developer documentation for TensorFlow Frontend
• Intel DevHub Discord channel - engage in discussions, ask questions and talk to OpenVINO developers

Contact points

• @openvinotoolkit/openvino-tf-frontend-maintainers
• rkazants in Discord

Ticket

No response

[AtharvaPore01]

.take

[github-actions]

Thank you for looking into this issue! Please let us know if you have any questions or require any help.

[AtharvaPore01]

I am having question regarding the pytest. To run the desired python code to test the C++ code changes, should I need to add the test cases to check for the desired results?

[rkazants]

Hi @AtharvaPore01,

Sorry for delay in response, I missed your message. For validation of C++ part required in this task, we use pytest based validation.
BTW, any update on this task?

Best regards,
Roman

[AtharvaPore01]

Hello @rkazants,

I apologize for the delay. I'm currently working on the task and endeavoring to submit it on time. Due to some prior commitments, I wasn't able to progress as quickly as I would have liked, but rest assured, I am actively working on it.

Best regards,
Atharva

[p-wysocki]

Hello @AtharvaPore01, are you still working on this? Is there anything we could help you with?

[rghvsh]

.take

[github-actions]

Thank you for looking into this issue! Please let us know if you have any questions or require any help.

[rghvsh]

Hi! @rkazants, Is this the correct way to do this

  if (complex_type_mark) {

    element::Type complex_part_type = complex_type_mark->get_complex_part_type();
    input = complex_type_mark->input_value(0);

    auto unsqueeze = make_shared<v0::Unsqueeze>(input, axis);
    set_node_name(node.get_name(), unsqueeze);
    return {unsqueeze};
}

[p-wysocki]

cc @rkazants

[hub-bla]

.take

[github-actions]

Thank you for looking into this issue! Please let us know if you have any questions or require any help.