ftag.fraction_optimization

Functions

convert_dict(→ numpy.ndarray | dict)	Convert the fraction values from dict to array or vice versa.
get_bkg_norm_dict(→ dict)	Get the normalisation dict for the background flavours.
calculate_rejection_sum(→ float)	Calculate the sum of the normalised rejections.
calculate_best_fraction_values(→ dict)	Calculate the best fraction values for a given tagger and working point.

Module Contents

ftag.fraction_optimization.convert_dict(fraction_values: dict | numpy.ndarray, backgrounds: ftag.labels.LabelContainer) → numpy.ndarray | dict

Convert the fraction values from dict to array or vice versa.

Parameters:

• fraction_values (dict | np.ndarray) – Dict of array with the fraction values

• backgrounds (LabelContainer) – LabelContainer with the background flavours

Returns:

Array or dict with the fraction values

Return type:

np.ndarray | dict

Raises:

TypeError – If the type of the input was wrong

ftag.fraction_optimization.get_bkg_norm_dict(jets: numpy.ndarray, tagger: str, signal: ftag.labels.Label, flavours: ftag.labels.LabelContainer, working_point: float) → dict

Get the normalisation dict for the background flavours.

Parameters:

• jets (np.ndarray) – Loaded jets

• tagger (str) – Name of the tagger

• signal (Label) – Label instance of the signal

• flavours (LabelContainer) – LabelContainer instance with all flavours used

• working_point (float) – Working point that is to be used

Returns:

Background normalisation dict

Return type:

dict

ftag.fraction_optimization.calculate_rejection_sum(fraction_dict: dict | numpy.ndarray, jets: numpy.ndarray, tagger: str, signal: ftag.labels.Label, flavours: ftag.labels.LabelContainer, working_point: float, bkg_norm_dict: dict, rejection_weights: dict) → float

Calculate the sum of the normalised rejections.

Parameters:

• fraction_dict (dict | np.ndarray) – Dict/Array with the fraction values

• jets (np.ndarray) – Loaded jets

• tagger (str) – Name of the tagger

• signal (Label) – Label instance of the signal

• flavours (LabelContainer) – LabelContainer with all flavours

• working_point (float) – Working point that is used

• bkg_norm_dict (dict) – Backgroud normalisation dict

• rejection_weights (dict) – Weights for the rejections

Returns:

Sum of the normalised rejections

Return type:

float

ftag.fraction_optimization.calculate_best_fraction_values(jets: numpy.ndarray, tagger: str, signal: ftag.labels.Label, flavours: ftag.labels.LabelContainer, working_point: float, rejection_weights: dict | None = None, optimizer_method: str = 'Powell') → dict

Calculate the best fraction values for a given tagger and working point.

Parameters:

• jets (np.ndarray) – Loaded jets

• tagger (str) – Name of the tagger

• signal (Label) – Label instance of the signal

• flavours (LabelContainer) – LabelContainer with all flavours

• working_point (float) – Working point that is used

• rejection_weights (dict | None, optional) – Rejection weights for the background classes, by default None

• optimizer_method (str, optional) – Optimizer method for the minimization, by default “Powell”

Returns:

Dict with the best fraction values

Return type:

dict