Quantizing for GNA Device

This example demonstrates the use of the Post-training Optimization Tool API for the task of quantizing a speech model for GNA device. Quantization for GNA is different from CPU quantization due to device specific: GNA supports quantized inputs in INT16 and INT32 (for activations) precision and quantized weights in INT8 and INT16 precision.

This example contains pre-selected quantization options based on the DefaultQuantization algorithm and created for models from Kaldi framework, and its data format. A custom ArkDataLoader is created to load the dataset from files with .ark extension for speech analysis task.

How to prepare the data

To run this example, you will need to use the .ark files for each model input from your <DATA_FOLDER>. For generating data from original formats to .ark, please, follow the Kaldi data preparation tutorial.

How to Run the example

  1. Launch Model Optimizer with the necessary options (for details follow the instructions for Kaldi to generate Intermediate Representation (IR) files for the model:

  2. Launch the example script:

    python3 <POT_DIR>/api/examples/speech/gna_example.py -m <PATH_TO_IR_XML> -w <PATH_TO_IR_BIN> -d <DATA_FOLDER> --input_names [LIST_OF_MODEL_INPUTS] --files_for_input [LIST_OF_INPUT_FILES]

    Required parameters:

    • -i, --input_names option. Defines list of model inputs;

    • -f, --files_for_input option. Defines list of filenames (.ark) mapped with input names. You should define names without extension, for example: FILENAME_1, FILENAME_2 maps with INPUT_1, INPUT_2.

    Optional parameters:

    • -p, --preset option. Defines preset for quantization: performance for INT8 weights, accuracy for INT16 weights;

    • -s, --subset_size option. Defines subset size for calibration;

    • -o, --output option. Defines output folder for quantized model.

  3. Validate your INT8 model using ./speech_example from the Inference Engine examples. Follow the speech example description link for details.