titanicprediction.data package

Submodules

titanicprediction.data.analysis module

class titanicprediction.data.analysis.BackendType(*values)

Bases: Enum

MATPLOTLIB = 'matplotlib'

PLOTLY = 'plotly'

SEABORN = 'seaborn'

class titanicprediction.data.analysis.CorrelationVisualizer

Bases: object

_apply_common_styling(ax: Axes, config: PlotConfig) → None

_create_correlation_heatmap(ax: Axes, corr_matrix: DataFrame, config: PlotConfig) → None

create_plot(data: DataFrame, plot_config: PlotConfig) → PlotResult

class titanicprediction.data.analysis.DistributionVisualizer

Bases: object

_apply_common_styling(ax: Axes, config: PlotConfig) → None

_create_histogram(ax: Axes, data: Series, config: PlotConfig) → None

_create_kde(ax: Axes, data: Series, config: PlotConfig) → None

_create_violin(ax: Axes, data: Series, config: PlotConfig) → None

create_plot(data: Any, plot_config: PlotConfig) → PlotResult

class titanicprediction.data.analysis.EDAVisualizer(output_dir: str = 'data/plots')

Bases: object

__init__(output_dir: str = 'data/plots')

_save_plot(plot_result: PlotResult, filename: str) → Path

create_correlation_analysis(dataset: Dataset) → dict[str, Path]

create_demographic_plots(dataset: Dataset) → dict[str, Path]

create_feature_importance_plot(model: TrainedModel) → dict[str, Path]

create_survival_analysis_plots(dataset: Dataset) → dict[str, Path]

create_training_plots(loss_history: list[float]) → dict[str, Path]

get_supported_formats() → list[str]

class titanicprediction.data.analysis.FeatureAnalysisVisualizer

Bases: object

_apply_common_styling(ax: Axes, config: PlotConfig) → None

_create_box_plot(ax: Axes, dataset: Dataset, config: PlotConfig) → None

_create_count_plot(ax: Axes, dataset: Dataset, config: PlotConfig) → None

_create_feature_distribution(ax: Axes, dataset: Dataset, config: PlotConfig) → None

_create_pie_chart(ax: Axes, dataset: Dataset, config: PlotConfig) → None

_create_scatter_plot(ax: Axes, dataset: Dataset, config: PlotConfig) → None

_create_survival_by_feature(ax: Axes, dataset: Dataset, config: PlotConfig) → None

create_plot(dataset: Dataset, plot_config: PlotConfig) → PlotResult

class titanicprediction.data.analysis.IVisualizer(*args, **kwargs)

Bases: Protocol

__init__(*args, **kwargs)

_abc_impl = <_abc._abc_data object>

_is_protocol = True

create_plot(data: Any, plot_config: PlotConfig) → PlotResult

get_supported_formats() → list[str]

save_plot(plot: PlotResult, filename: str) → Path

class titanicprediction.data.analysis.PlotConfig(plot_type: titanicprediction.data.analysis.PlotType, title: str, x_label: str, y_label: str, figsize: tuple[int, int] = (10, 6), style: str = 'seaborn-v0_8', colors: list[str] = None, save_format: str = 'png', dpi: int = 300, font_size: int = 12, legend: bool = True, grid: bool = True)

Bases: object

__init__(plot_type: PlotType, title: str, x_label: str, y_label: str, figsize: tuple[int, int] = (10, 6), style: str = 'seaborn-v0_8', colors: list[str] = None, save_format: str = 'png', dpi: int = 300, font_size: int = 12, legend: bool = True, grid: bool = True) → None

colors: list[str] = None

dpi: int = 300

figsize: tuple[int, int] = (10, 6)

font_size: int = 12

grid: bool = True

legend: bool = True

plot_type: PlotType

save_format: str = 'png'

style: str = 'seaborn-v0_8'

title: str

x_label: str

y_label: str

class titanicprediction.data.analysis.PlotResult(figure: matplotlib.figure.Figure, axes: matplotlib.axes._axes.Axes, config: titanicprediction.data.analysis.PlotConfig, metadata: dict[str, Any])

Bases: object

__init__(figure: Figure, axes: Axes, config: PlotConfig, metadata: dict[str, Any]) → None

axes: Axes

config: PlotConfig

figure: Figure

metadata: dict[str, Any]

class titanicprediction.data.analysis.PlotType(*values)

Bases: Enum

BAR = 'bar'

BOX = 'box'

COUNT = 'count'

HEATMAP = 'heatmap'

HISTOGRAM = 'histogram'

KDE = 'kde'

LINE = 'line'

PIE = 'pie'

SCATTER = 'scatter'

VIOLIN = 'violin'

class titanicprediction.data.analysis.TrainingVisualizer

Bases: object

_create_training_curve(ax: Axes, loss_history: list[float], config: PlotConfig) → None

create_plot(loss_history: list[float], plot_config: PlotConfig) → PlotResult

titanicprediction.data.preprocessing module

class titanicprediction.data.preprocessing.AgeImputer(strategy: Literal['mean', 'median', 'mode', 'constant'] = 'median', fill_value: float | None = None, _imputer: Any = None)

Bases: object

__init__(strategy: Literal['mean', 'median', 'mode', 'constant'] = 'median', fill_value: float | None = None, _imputer: Any = None) → None

_imputer: Any = None

fill_value: float | None = None

fit(dataset: Dataset) → None

fit_transform(dataset: Dataset) → Dataset

get_params() → dict[str, Any]

strategy: Literal['mean', 'median', 'mode', 'constant'] = 'median'

transform(dataset: Dataset) → Dataset

class titanicprediction.data.preprocessing.CategoricalEncoder(encoding_type: Literal['onehot', 'label'] = 'onehot', columns: list[str] = None, handle_unknown: Literal['error', 'ignore', 'use_encoded_value'] = 'error', _encoders: dict[str, Any] = None, _feature_names: list[str] = None)

Bases: object

__init__(encoding_type: Literal['onehot', 'label'] = 'onehot', columns: list[str] = None, handle_unknown: Literal['error', 'ignore', 'use_encoded_value'] = 'error', _encoders: dict[str, Any] = None, _feature_names: list[str] = None) → None

_encoders: dict[str, Any] = None

_feature_names: list[str] = None

_update_feature_names(original_col: str, new_cols: list[str]) → None

columns: list[str] = None

encoding_type: Literal['onehot', 'label'] = 'onehot'

fit(dataset: Dataset) → None

fit_transform(dataset: Dataset) → Dataset

get_params() → dict[str, Any]

handle_unknown: Literal['error', 'ignore', 'use_encoded_value'] = 'error'

transform(dataset: Dataset) → Dataset

class titanicprediction.data.preprocessing.ColumnDropper(columns: list[str] = None)

Bases: object

__init__(columns: list[str] = None) → None

columns: list[str] = None

fit(dataset: Dataset) → None

fit_transform(dataset: Dataset) → Dataset

get_params() → dict[str, Any]

transform(dataset: Dataset) → Dataset

class titanicprediction.data.preprocessing.DataPreprocessor

Bases: object

__init__()

add_step(name: str, transformer: IDataTransformer) → None

fit(dataset: Dataset) → None

fit_transform(dataset: Dataset) → Dataset

fitted: bool

get_params() → dict[str, Any]

preprocessing_steps: list[tuple[str, IDataTransformer]]

transform(dataset: Dataset) → Dataset

transform_features(features: DataFrame) → DataFrame

class titanicprediction.data.preprocessing.FeatureScaler(method: Literal['standard', 'minmax', 'robust'] = 'standard', columns: list[str] = None, with_mean: bool = True, with_std: bool = True, _scalers: dict[str, Any] = None)

Bases: object

__init__(method: Literal['standard', 'minmax', 'robust'] = 'standard', columns: list[str] = None, with_mean: bool = True, with_std: bool = True, _scalers: dict[str, Any] = None) → None

_scalers: dict[str, Any] = None

columns: list[str] = None

fit(dataset: Dataset) → None

fit_transform(dataset: Dataset) → Dataset

get_params() → dict[str, Any]

method: Literal['standard', 'minmax', 'robust'] = 'standard'

transform(dataset: Dataset) → Dataset

with_mean: bool = True

with_std: bool = True

class titanicprediction.data.preprocessing.IDataTransformer(*args, **kwargs)

Bases: Protocol

__init__(*args, **kwargs)

_abc_impl = <_abc._abc_data object>

_is_protocol = True

fit(dataset: Dataset) → None

fit_transform(dataset: Dataset) → Dataset

get_params() → dict[str, Any]

transform(dataset: Dataset) → Dataset

class titanicprediction.data.preprocessing.PreprocessorFactory

Bases: object

static create_titanic_preprocessor() → DataPreprocessor

class titanicprediction.data.preprocessing.TitleExtractor(name_column: str = 'Name', title_column: str = 'Title', custom_mappings: dict[str, str] = None, _title_patterns: dict[str, str] = None)

Bases: object

__init__(name_column: str = 'Name', title_column: str = 'Title', custom_mappings: dict[str, str] = None, _title_patterns: dict[str, str] = None) → None

_title_patterns: dict[str, str] = None

custom_mappings: dict[str, str] = None

fit(dataset: Dataset) → None

fit_transform(dataset: Dataset) → Dataset

get_params() → dict[str, Any]

name_column: str = 'Name'

title_column: str = 'Title'

transform(dataset: Dataset) → Dataset

titanicprediction.data.repositories module

class titanicprediction.data.repositories.CSVDataRepository(file_path: str, target_column: str = 'Survived')

Bases: object

__init__(file_path: str, target_column: str = 'Survived')

get_metadata() → dict[str, Any]

load_data() → Dataset

save_data(dataset: Dataset) → bool

class titanicprediction.data.repositories.FileModelRepository(models_dir: str = 'models')

Bases: object

__init__(models_dir: str = 'models')

delete_model(name: str) → bool

get_model_info(name: str) → dict[str, Any] | None

list_models() → list[str]

load_model(name: str) → TrainedModel | None

save_model(model: TrainedModel, name: str) → bool

class titanicprediction.data.repositories.IDataRepository(*args, **kwargs)

Bases: Protocol

__init__(*args, **kwargs)

_abc_impl = <_abc._abc_data object>

_is_protocol = True

get_metadata() → dict[str, Any]

load_data() → Dataset

save_data(dataset: Dataset) → bool

class titanicprediction.data.repositories.IModelRepository(*args, **kwargs)

Bases: Protocol

__init__(*args, **kwargs)

_abc_impl = <_abc._abc_data object>

_is_protocol = True

list_models() → list[str]

load_model(name: str) → TrainedModel | None

save_model(model: TrainedModel, name: str) → bool

Module contents