.. _Build NeurEco Compression model with the command line interface: Build NeurEco Compression model with the command line interface ======================================================================== To build a NeurEco Compression model, run the following command in the terminal: .. code-block:: shell neurecoDNN build path/to/build/configuration/file/build.conf .. _Tabular Build Conf: The skeleton of a configuration file required to build NeurEco Compression model, here build.conf, looks as follows. Its fields should be filled according to the problem at hand. .. code-block:: javascript :linenos: { "neurecoDNN_build": { "DevSettings": { "valid_percentage": 33.33, "initial_beta_reg": 0.1, "validation_indices": "", "final_learning": true, "disconnect_inputs_if_possible": true }, "input_normalization": { "shift_type": "auto", "scale_type": "auto", "normalize_per_feature": true }, "output_normalization": { "shift_type": "none", "scale_type": "none", "normalize_per_feature": false }, "UserSettings": { "gpu_id": 0, "use_gpu": false }, "classification": false, "exc_filenames": [], "output_filenames": [], "validation_exc_filenames": [], "validation_output_filenames": [], "write_model_to": "model.ednn", "write_compression_model_to": "CompModel.ednn", "write_decompression_model_to": "DecompModel.ednn", "minimum_compression_coefficient": 1, "compress_tolerance": 0.02, "build_compress": true, "starting_from_checkpoint_address": "", "checkpoint_address": "ckpt.checkpoint", "resume": false, } } .. _Build parameters Compression conf: Building parameters --------------------- | The available building parameters in the configuration file are described in the following table. .. csv-table:: NeurEco building parameters in python API :file: csv_tables/ConfCompressionNeurEcoBuildingParameters.csv :header-rows: 1 :class: longtable :widths: 3, 3, 8 :delim: ; :align: center .. _Normalizing the data compression conf: Data normalization for Tabular Compression -------------------------------------------- .. include:: ../../CommonParts/NormalizationTabular.rst