scparadise.scnoah.report_reg

Contents

scparadise.scnoah.report_reg#

scparadise.scnoah.report_reg(adata_prot, adata_pred_prot, multioutput='uniform_average', save_report=False, report_name='report_regression.csv', save_path='', ndigits=3)[source]#

Returns multiple metrics of cell surface proteins prediction. Root mean squared error (RMSE), mean absolute error (MeanAE), median absolute error (MedianAE) : lower value - better prediction. Coefficient of determination (R² score), explained variance score (EVS) : higher value - better prediction

Parameters:

  • adata_prot (AnnData) – Annotated data matrix with proteins. Test dataset not used for model tuning or training.

  • adata_pred_prot (AnnData) – Annotated data matrix with predicted proteins.

  • multioutput ({‘raw_values’, ‘uniform_average’} or array-like of shape (n_outputs,), (default=’uniform_average’)) – Defines aggregating of multiple output values. Array-like value defines weights used to average errors. ‘raw_values’ : Returns a full set of errors in case of multioutput input. ‘uniform_average’ : Errors of all outputs are averaged with uniform weight.

  • save_report (bool, (default: False)) – Save report as csv file or not.

  • report_name (str, (default: 'report_sens_spec.csv')) – Name of a file to save report.

  • save_path (path object) – Path to a folder to save report.

  • ndigits (int (default: 3)) – Round a number to a given precision in decimal digits.

Contents