Forums - Build SNPE Network failure

I have a problem with building SNPE neural network corresponding to my DLC file.

I took existing PB model file, which worked fine in an application using TensorFlow interface.

public class TensorFlowClassifier {
    static {
        System.loadLibrary("tensorflow_inference");
    }

    private TensorFlowInferenceInterface inferenceInterface;
    private static final String MODEL_FILE = "file:///android_asset/frozen_model.pb";
    private static final String INPUT_NODE = "inputs";
    private static final String[] OUTPUT_NODES = {"y_"};
    private static final String OUTPUT_NODE = "y_";
    private static final long[] INPUT_SIZE = {1, 200, 3};
    private static final int OUTPUT_SIZE = 6;

    public TensorFlowClassifier(final Context context) {
        inferenceInterface = new TensorFlowInferenceInterface(context.getAssets(), MODEL_FILE);
    }

    public float[] processData(float[] data) {
        float[] result = new float[OUTPUT_SIZE];
        inferenceInterface.feed(INPUT_NODE, data, INPUT_SIZE);
        inferenceInterface.run(OUTPUT_NODES);
        inferenceInterface.fetch(OUTPUT_NODE, result);

        return result;
    }
}

I converted the PB model to DLC  by means of the following command:

./snpe-1.17.1/bin/x86_64-linux-clang/snpe-tensorflow-to-dlc --graph ~/ALPHA/FrozenModelLSTM/frozen_har.pb -i "input" 1,200,3 --out_node "y_" --dlc frozen_har.dlc --allow_unconsumed_nodes

I used the created DLC model file in my application:

final SNPE.NeuralNetworkBuilder builder = new SNPE.NeuralNetworkBuilder(mApp)
        .setDebugEnabled(false)
        // Allows selecting a runtime order for the network.
         .setRuntimeOrder(CPU)
        // Loads a model from DLC file
        .setModel(new File(modelFileWithPath));


Log.e(LOG_TAG, "Before builder.build()");

network = builder.build();

Please help.  Will be very grateful for your prompt answer.