Forums - snpe-tensorflow-to-dlc conversion failed in tf with UDO

I have a quantized tensorflow model file, and also define a UDO (json format file). The model defines an OwnOnednnMul and FsmnForward operation, but when it is converted to dlc format through snpe-tensorflow-to-dlc (command line: snpe -tensorflow-to-dlc --input_network model.pb --input_dim xxxx --out_node output --output_path model.dlc --udo udo.json --show_unconsumed_nodes), the following error will appear:


qti.aisw.converters.tensorflow.util.ConverterError: ERROR_TF_INPUT_OPERATION_NOT_FOUND: Input operation not found for Inference/split
2021-06-17 15:52:34,624-183-ERROR-Conversion FAILED: build_layer failed with Exception ERROR_TF_INPUT_OPERATION_NOT_FOUND: Input operation not found for Inference/split in layer SliceLayerBuilder
Traceback (most recent call last):
  File "/opt/snpe-1.50.0.2622/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 548, in _create_layer
    layer_builder.build_layer(self.graph, self._context, descriptor, inputs, outputs)
  File "/opt/snpe-1.50.0.2622/lib/python/qti/aisw/converters/tensorflow/layers/slice.py", line 75, in build_layer
    input_shape = converter_context.get_input_layer_output_shape_for(descriptor.child_ops[0])
  File "/opt/snpe-1.50.0.2622/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 200, in get_input_layer_output_shape_for
    output_op = self._get_input_layer_output_op_for(operation, idx)
  File "/opt/snpe-1.50.0.2622/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 261, in _get_input_layer_output_op_for
    raise ConverterError(code_to_message.get_error_message('ERROR_TF_INPUT_OPERATION_NOT_FOUND')(operation.name))
qti.aisw.converters.tensorflow.util.ConverterError: ERROR_TF_INPUT_OPERATION_NOT_FOUND: Input operation not found for Inference/split

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/opt/snpe-1.50.0.2622/bin/x86_64-linux-clang/snpe-tensorflow-to-dlc", line 52, in main
    ir_graph = converter.convert()
  File "/opt/snpe-1.50.0.2622/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 339, in convert
    self._convert_layers()
  File "/opt/snpe-1.50.0.2622/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 390, in _convert_layers
    self._create_layers(descriptors)
  File "/opt/snpe-1.50.0.2622/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 440, in _create_layers
    self._create_layer(layer_builder, descriptor)
  File "/opt/snpe-1.50.0.2622/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 553, in _create_layer
    (e, layer_builder.__class__.__name__))
qti.aisw.converters.tensorflow.util.ConverterError: build_layer failed with Exception ERROR_TF_INPUT_OPERATION_NOT_FOUND: Input operation not found for Inference/split in layer SliceLayerBuilder


The content of the UDO file is as follows:
{
  "UdoPackage_0":
  {
    "Operators": [
      {
        "type": "OwnOnednnMul",
        "inputs":[
          {"name": "input", "data_type": "INT_32", "quantization_mode":"TF"}
        ],
        "outputs":[
          {"name": "output", "data_type": "INT_32"}
        ],
        "core_types": ["DSP"],
        "dsp_arch_types": ["v68"]
      },
{
        "type": "FsmnForward",
        "inputs":[
          {"name": "input", "data_type": "INT_32", "quantization_mode":"TF"}
        ],
        "outputs":[
          {"name": "out", "data_type": "INT_32"}
        ],
        "core_types": ["DSP"],
        "dsp_arch_types": ["v68"]
      }
    ],
    "UDO_PACKAGE_NAME": "XXUDOPackage"
  }
}

How can I solve this problem? This problem has been blocked for several days. Any help in solving this issue will be greatly appreciated.