build_model

Command-line usage

usage: cell build_model [-h] [-ccf ccalc.pickle] [-ssf sset.json]
                        [-cpf cpool.json] [-mof model.pickle] [-st identity]
                        [-so {'fit_intercept': True}]
                        [-et skl_LinearRegression]
                        [-eo {'fit_intercept': True}] [-plotovsd None]
                        [-wf None]
                        property_name [report_errors]

positional arguments:
  property_name         Property to be modelled.
  report_errors         [True]

options:
  -h, --help            show this help message and exit
  -ccf, --ccalc-filepath ccalc.pickle
                        Path to CorrelationsCalculator object.
  -ssf, --sset-filepath sset.json
                        Path to StructuresSet object.
  -cpf, --cpool-filepath cpool.json
                        Path to ClustersPool object.
  -mof, --model-filepath model.pickle
                        Output Model file path.
  -st, --selector-type identity
                        Selector type.
  -so, --selector-opts {'fit_intercept': True}
                        Selector options.
  -et, --estimator-type skl_LinearRegression
                        Estimator type.
  -eo, --estimator-opts {'fit_intercept': True}
                        Estimator options.
  -plotovsd, --plot-optimization-vs-sparsity None
                        Plot config.
  -wf, --weights-filepath None
                        Path to npz file containing weights.

Arguments and behavior

clusterx.cli.build_model.build_model(property_name, ccalc_filepath='ccalc.pickle', sset_filepath='sset.json', cpool_filepath='cpool.json', model_filepath='model.pickle', selector_type='identity', selector_opts={'fit_intercept': True}, estimator_type='skl_LinearRegression', estimator_opts={'fit_intercept': True}, report_errors=True, plot_optimization_vs_sparsity=None, weights_filepath=None)

Compute CE model