Tutorial: using NeurEco command line interface for a Parametric Frequency Sweep problem#

NeurEcoFNN is the executable used for building, evaluating and exporting Frequency Domain models (Parametric Frequency Sweep). The executable can be called directly from a terminal / powershell only after a full installation (the portable version doesn’t allow this option). To call the executable, run the following command:

neurecoFNN

which will output:

Running NeurEco Frequential version 3.0.616.0 compiled with MSVC v1928  on Apr  3 2023 @ 16:53:14
usage: neurecoRNN [-h] [command <parameters>]

Entry point for neurecoRNN network building and evaluation.

Commands:
build <configurationFilename>
      build a neurecoFNN network from a given input solution/excitation set.

evaluate <configurationFilename>
      evaluate a neurecoFNN network from a given input solution/excitation set.

exportFMU <FMU model full path> <serialized network full path> <ORed platform flag (windows=1, linux=2)>
      export a serialized network as an FMU file.

Optional arguments:
-h, --help   show this message and exit

The following section uses the test case Frequency Selective Surface. This test case is delivered with the NeurEco installation package.

To build a Parametric Frequency Sweep model using the executable:

{"neurecoFNN_build": {
    "AdvancedSettings": {
    },
    "checkpoint_address": "./fssModel/fss_model.checkpoint",
    "input_filenames": [
        "./inputs_train.npy"
    ],
    "output_filenames": [
        "./targets_train.npy"
    ],
    "resume": false,
    "settings": {
        "compressed_space_size": 5,
        "enrichment_rate": 0.2,
        "max_number_of_enrichments": 200,
        "min_number_of_enrichments": 10,
        "unsuccessful_enrichments": 4,
        "validation_percentage": 30
    },
    "test_input_filenames": [
        "./inputs_test.npy"
    ],
    "test_output_filenames": [
        "./targets_test.npy"
    ],
    "validation_input_filenames": [
        "./inputs_valid.npy"
    ],
    "validation_output_filenames": [
        "./targets_valid.npy"
    ],
    "write_model_to": "./fssModel/fss_model.efnn"
}

  • Place this configuration file in the same directory as the data of the test case (inputs_train.npy, inputs_valid.npy, inputs_test.npy, targets_train.npy, targets_valid.npy, targets_test.npy), otherwise adjust the relative paths to the data files in the configuration file.

  • To launch the build, run the following command in the terminal (opened in the data directory, otherwise adjust the relative path to the configuration file):

neurecoFNN build ./build_configuration_file.conf

  • The build starts automatically:

00h00m00s info > Running NeurEco Frequential version 3.0.616.0 compiled with MSVC v1928  on Apr  3 2023 @ 16:53:14
00h00m00s info > Reading Dataset...

To evaluate a Parametric Frequency Sweep model using the executable:

{
 "neurecoFNN_evaluate": {
     "input_filenames": ["./inputs_test.npy"],
     "neureco_filename": "./fssModel/fss_model.efnn",
     "write_model_output_to_directory": "./EvalResults"
 }
 }

  • Place this configuration file in the same directory as the data of the test case (inputs_test.npy), otherwise adjust the relative paths to the data files in the configuration file.

  • To launch the evaluation, run the following command in the terminal (opened in the data directory, otherwise adjust the relative path to the configuration file):

neurecoFNN evaluate ./eval_configuration_file.conf

  • The model is evaluated on the testing data in “./inputs_test.npy”, and the results are saved in the directory created by NeurEco: “./EvalResults”.

To export a Parametric Frequency Sweep model to the FMU format using the executable (neureco_embed_pfs license is required):

  • Run the following command (with 1 for ORed platform flag: windows=1, linux=2):

neurecoFNN exportFMU ./fssModel/fss_model.efnn ./fssModel/fss_model.fmu 1