RunnerConfig

class pyprophet._config.RunnerConfig(classifier: ~typing.Literal['LDA', 'SVM', 'XGBoost', 'HistGradientBoosting'] = 'LDA', autotune: bool = False, ss_main_score: str = 'auto', main_score_selection_report: bool = False, xgb_params: dict = <factory>, xeval_fraction: float = 0.5, xeval_num_iter: int = 10, ss_initial_fdr: float = 0.15, ss_iteration_fdr: float = 0.05, ss_num_iter: int = 10, ss_score_filter: bool = False, ss_scale_features: bool = False, group_id: str = 'group_id', error_estimation_config: ~pyprophet._config.ErrorEstimationConfig = <factory>, ipf_max_peakgroup_rank: int = 1, ipf_max_peakgroup_pep: float = 0.7, ipf_max_transition_isotope_overlap: float = 0.5, ipf_min_transition_sn: float = 0.0, glyco: bool = False, density_estimator: str = 'gmm', grid_size: int = 256, add_alignment_features: bool = False, tric_chromprob: bool = False, threads: int = 1, test: bool = False, color_palette: str = 'normal')[source]

Bases: object

Configuration for scoring, classifier setup, learning parameters, and optional features.

classifier

Classifier type used for semi-supervised learning (‘LDA’, ‘SVM’, ‘XGBoost’ or ‘HistGradientBoosting’).

Type:: str

autotune

Whether to autotune hyperparameters for the classifier (XGBoost / SVM / HistGradientBoosting)

Type:: bool

ss_main_score

Starting main score for semi-supervised learning (can be ‘auto’).

Type:: str

main_score_selection_report

Whether to generate a report for main score selection.

Type:: bool

xgb_params

Default parameters for XGBoost/HistGradientBoosting training.

Type:: dict

xeval_fraction

Fraction of data used in each cross-validation iteration.

Type:: float

xeval_num_iter

Number of cross-validation iterations.

Type:: int

ss_initial_fdr

Initial FDR threshold for target selection.

Type:: float

ss_iteration_fdr

FDR threshold used in subsequent learning iterations.

Type:: float

ss_num_iter

Number of semi-supervised training iterations.

Type:: int

ss_score_filter

Whether to filter features based on score set or profile.

Type:: bool

ss_scale_features

Whether to scale features before training.

Type:: bool

ss_use_dynamic_main_score

Automatically determined during __post_init__.

Type:: bool

group_id

Column used to group PSMs for learning and statistics.

Type:: str

error_estimation_config

Settings for global and local error estimation.

Type:: ErrorEstimationConfig

ipf_max_peakgroup_rank

Max rank of peak groups considered in IPF.

Type:: int

ipf_max_peakgroup_pep

Max PEP for peak group consideration in IPF.

Type:: float

ipf_max_transition_isotope_overlap

Max isotope overlap for transition selection in IPF.

Type:: float

ipf_min_transition_sn

Min log S/N for transition selection in IPF.

Type:: float

glyco

Whether glycopeptide-specific scoring is enabled.

Type:: bool

density_estimator

Score density estimation method (‘kde’ or ‘gmm’).

Type:: str

grid_size

Number of grid cutoffs used for local FDR calculation.

Type:: int

add_alignment_features

Whether to add chromatographic alignment features.

Type:: bool

tric_chromprob

Whether to compute chromatogram probabilities (for TRIC).

Type:: bool

threads

Number of CPU threads to use; -1 means all CPUs.

Type:: int

test

Whether to enable test mode with deterministic behavior.

Type:: bool

color_palette

Color palette used in PDF report rendering.

Type:: str

__eq__(other): Return self==value.

__hash__ = None

__init__(classifier: ~typing.Literal['LDA', 'SVM', 'XGBoost', 'HistGradientBoosting'] = 'LDA', autotune: bool = False, ss_main_score: str = 'auto', main_score_selection_report: bool = False, xgb_params: dict = <factory>, xeval_fraction: float = 0.5, xeval_num_iter: int = 10, ss_initial_fdr: float = 0.15, ss_iteration_fdr: float = 0.05, ss_num_iter: int = 10, ss_score_filter: bool = False, ss_scale_features: bool = False, group_id: str = 'group_id', error_estimation_config: ~pyprophet._config.ErrorEstimationConfig = <factory>, ipf_max_peakgroup_rank: int = 1, ipf_max_peakgroup_pep: float = 0.7, ipf_max_transition_isotope_overlap: float = 0.5, ipf_min_transition_sn: float = 0.0, glyco: bool = False, density_estimator: str = 'gmm', grid_size: int = 256, add_alignment_features: bool = False, tric_chromprob: bool = False, threads: int = 1, test: bool = False, color_palette: str = 'normal') → None

__repr__()[source]: Return repr(self).

__str__()[source]: Return str(self).

__weakref__: list of weak references to the object (if defined)