rail.evaluation.single_evaluator module
” Abstract base class defining an Evaluator
The key feature is that the evaluate method.
- class rail.evaluation.single_evaluator.SingleEvaluator(args, comm=None)[source]
Bases:
Evaluator
Evaluate the performance of a photo-Z estimator
Configuration Parameters: output_mode [str]: What to do with the outputs (default=default) metrics [list]: The metrics you want to evaluate. (default=[]) exclude_metrics [list]: List of metrics to exclude (default=[]) metric_config [dict]: configuration of individual_metrics (default={}) chunk_size [int]: The default number of PDFs to evaluate per loop. (default=10000) _random_state [float]: Random seed value to use for reproducible results. (default=None) force_exact [bool]: Force the exact calculation. This will not allow parallelization (default=False) point_estimates [list]: List of point estimates to use (default=[]) truth_point_estimates [list]: List of true point values to use (default=[]) hdf5_groupname [str]: HDF5 Groupname for truth table. (default=photometry)
- config_options = {'_random_state': <ceci.config.StageParameter object>, 'chunk_size': <ceci.config.StageParameter object>, 'exclude_metrics': <ceci.config.StageParameter object>, 'force_exact': <ceci.config.StageParameter object>, 'hdf5_groupname': <ceci.config.StageParameter object>, 'metric_config': <ceci.config.StageParameter object>, 'metrics': <ceci.config.StageParameter object>, 'output_mode': <ceci.config.StageParameter object>, 'point_estimates': <ceci.config.StageParameter object>, 'truth_point_estimates': <ceci.config.StageParameter object>}
- inputs = [('input', <class 'rail.core.data.QPOrTableHandle'>), ('truth', <class 'rail.core.data.QPOrTableHandle'>)]
- metric_base_class
alias of
BaseMetric
- name = 'SingleEvaluator'