return conflictResolver.resolveConflicts(getGraph(), ArtifactScopeEnum.compile);
        } else if (requestType.equals(MetadataResolutionRequestTypeEnum.classpathRuntime)) {
            return conflictResolver.resolveConflicts(getGraph(), ArtifactScopeEnum.runtime);
        } else if (requestType.equals(MetadataResolutionRequestTypeEnum.classpathTest)) {
            return conflictResolver.resolveConflicts(getGraph(), ArtifactScopeEnum.test);

  metagraph_def.ParseFromArray(metagraph->data, metagraph->length);
  TF_DeleteBuffer(metagraph);

  EXPECT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
  CSession csession(session);

  // Retrieve the regression signature from meta graph def.
  const auto signature_def_map = metagraph_def.signature_def();
  const auto signature_def = signature_def_map.at("regress_x_to_y");

  *input = {oper, 0};
  return true;
}

// Copies `src_graph` into `dst_graph`. Any node in `src_graph` with input
// `src_inputs[i]` will have that input replaced with `dst_inputs[i]`.  `prefix`
// will be prepended to copied node names. `control_deps` are nodes in
// `dst_graph` that the copied `src_graph` nodes will have control dependencies

        assumes layers only ever have one positional argument. This assumption
        doesn't hold true before TF 2.4 either, but is more likely to cause
        issues now.
    *   Code that manually enters `keras.backend.get_graph()` before building a
        functional model is no longer needed.
    *   Start enforcing input shape assumptions when calling Functional API
        Keras models. This may potentially break some users, in case there is a