Quickstart: Parametric Frequency Sweep with the command line interface#

NeurEcoFNN is the executable used for building, evaluating and exporting Parametric Frequency Sweep models. The executable can be called directly from a terminal / powershell only after a full installation (the portable version does not offer this option).

To build a NeurEco Parametric Frequency Sweep model, run the following command in the terminal:

neurecoFNN build path/to/build/configuration/file/build.conf

The skeleton of a configuration file required to build NeurEco Parametric Frequency Sweep model, here build.conf, looks as follows (for the test case Frequency Selective Surface). Its fields should be filled according to the problem at hand.

 1 {"neurecoFNN_build": {
 2     "AdvancedSettings": {
 3     },
 4     "checkpoint_address": "./fssModel/fss_model.checkpoint",
 5     "input_filenames": [
 6         "./inputs_train.npy"
 7     ],
 8     "output_filenames": [
 9         "./targets_train.npy"
10     ],
11     "resume": false,
12     "settings": {
13         "compressed_space_size": 5,
14         "enrichment_rate": 0.2,
15         "max_number_of_enrichments": 200,
16         "min_number_of_enrichments": 10,
17         "unsuccessful_enrichments": 4,
18         "validation_percentage": 30
19     },
20     "test_input_filenames": [
21         "./inputs_test.npy"
22     ],
23     "test_output_filenames": [
24         "./targets_test.npy"
25     ],
26     "validation_input_filenames": [
27         "./inputs_valid.npy"
28     ],
29     "validation_output_filenames": [
30         "./targets_valid.npy"
31     ],
32     "write_model_to": "./fssModel/fss_model.efnn"
33 }

Note

For detailed documentation on build, see Build NeurEco Parametric Frequency Sweep model with the command line interface. For data preparation, see Data preparation for NeurEco Parametric Frequency Sweep with the command line interface.

To perform an evaluation, run the following command in the terminal:

neurecoFNN evaluate path/to/evaluation/configuration/file/eval.conf

The skeleton of an evaluation configuration file, here eval.conf, looks as follows (for the test case Frequency Selective Surface). Its fields should be filled according to the problem at hand.

{
 "neurecoFNN_evaluate": {
     "input_filenames": ["./inputs_test.npy"],
     "neureco_filename": "./FSS.efnn",
     "write_model_output_to_directory": "./EvaluationResults"
 }
}

Note

For detailed documentation on evaluate, see Evaluate NeurEco Parametric Frequency Sweep model with the command line interface.

To export the model to FMU format, run:

neurecoRNN exportFMU ./FSS.efnn ./FSS.fmu

Export to FMU format requires neureco_embed_pfs license.

Note

For detailed documentation on Parametric Frequency Sweep with the command line interface, see Parametric Frequency Sweep with the command line interface.