immuneML.hyperparameter_optimization.states package

Submodules

immuneML.hyperparameter_optimization.states.HPAssessmentState module

class immuneML.hyperparameter_optimization.states.HPAssessmentState.HPAssessmentState(split_index: int, train_val_dataset, test_dataset, path: Path, label_configuration: LabelConfiguration)[source]

Bases: object

immuneML.hyperparameter_optimization.states.HPItem module

class immuneML.hyperparameter_optimization.states.HPItem.HPItem(method: immuneML.ml_methods.classifiers.MLMethod.MLMethod = None, encoder: immuneML.encodings.DatasetEncoder.DatasetEncoder = None, performance: dict = None, hp_setting: immuneML.hyperparameter_optimization.HPSetting.HPSetting = None, train_predictions_path: pathlib.Path = None, test_predictions_path: pathlib.Path = None, ml_settings_export_path: pathlib.Path = None, train_dataset: immuneML.data_model.datasets.Dataset.Dataset = None, test_dataset: immuneML.data_model.datasets.Dataset.Dataset = None, split_index: int = None, model_report_results: List[immuneML.reports.ReportResult.ReportResult] = <factory>, encoding_train_results: List[immuneML.reports.ReportResult.ReportResult] = <factory>, encoding_test_results: List[immuneML.reports.ReportResult.ReportResult] = <factory>)[source]

Bases: object

encoder: DatasetEncoder = None
encoding_test_results: List[ReportResult]
encoding_train_results: List[ReportResult]
hp_setting: HPSetting = None
method: MLMethod = None
ml_settings_export_path: Path = None
model_report_results: List[ReportResult]
performance: dict = None
split_index: int = None
test_dataset: Dataset = None
test_predictions_path: Path = None
train_dataset: Dataset = None
train_predictions_path: Path = None

immuneML.hyperparameter_optimization.states.HPLabelState module

class immuneML.hyperparameter_optimization.states.HPLabelState.HPLabelState(label, auxiliary_labels)[source]

Bases: object

property optimal_assessment_item
property optimal_hp_setting

immuneML.hyperparameter_optimization.states.HPSelectionState module

class immuneML.hyperparameter_optimization.states.HPSelectionState.HPSelectionState(train_datasets, val_datasets, path: Path, hp_strategy: HPOptimizationStrategy)[source]

Bases: object

property optimal_hp_setting

immuneML.hyperparameter_optimization.states.TrainMLModelState module

class immuneML.hyperparameter_optimization.states.TrainMLModelState.TrainMLModelState(dataset: immuneML.data_model.datasets.Dataset.Dataset, hp_strategy: immuneML.hyperparameter_optimization.strategy.HPOptimizationStrategy.HPOptimizationStrategy, hp_settings: List[immuneML.hyperparameter_optimization.HPSetting.HPSetting], assessment: immuneML.hyperparameter_optimization.config.SplitConfig.SplitConfig, selection: immuneML.hyperparameter_optimization.config.SplitConfig.SplitConfig, metrics: Set[immuneML.ml_metrics.ClassificationMetric.ClassificationMetric], optimization_metric: immuneML.ml_metrics.ClassificationMetric.ClassificationMetric, label_configuration: immuneML.environment.LabelConfiguration.LabelConfiguration, path: pathlib.Path = None, context: dict = None, number_of_processes: int = 1, reports: dict = <factory>, name: str = None, refit_optimal_model: bool = None, export_all_ml_settings: bool = None, example_weighting: immuneML.example_weighting.ExampleWeightingStrategy.ExampleWeightingStrategy = None, sequence_type: immuneML.environment.SequenceType.SequenceType = <SequenceType.AMINO_ACID: 'sequence_aa'>, region_type: immuneML.data_model.SequenceParams.RegionType = <RegionType.IMGT_CDR3: 'cdr3'>, optimal_hp_items: Dict[str, immuneML.hyperparameter_optimization.states.HPItem.HPItem] = <factory>, optimal_hp_item_paths: Dict[str, str] = <factory>, assessment_states: List[immuneML.hyperparameter_optimization.states.HPAssessmentState.HPAssessmentState] = <factory>, report_results: List[immuneML.reports.ReportResult.ReportResult] = <factory>)[source]

Bases: object

assessment: SplitConfig
assessment_states: List[HPAssessmentState]
context: dict = None
dataset: Dataset
example_weighting: ExampleWeightingStrategy = None
export_all_ml_settings: bool = None
hp_settings: List[HPSetting]
hp_strategy: HPOptimizationStrategy
label_configuration: LabelConfiguration
metrics: Set[ClassificationMetric]
name: str = None
number_of_processes: int = 1
optimal_hp_item_paths: Dict[str, str]
optimal_hp_items: Dict[str, HPItem]
optimization_metric: ClassificationMetric
path: Path = None
refit_optimal_model: bool = None
region_type: RegionType = 'cdr3'
report_results: List[ReportResult]
reports: dict
selection: SplitConfig
sequence_type: SequenceType = 'sequence_aa'

Module contents