cenreg package

Subpackages

• cenreg.distribution package
  • Submodules
  • cenreg.distribution.empirical module
    • EmpiricalCDF
      • EmpiricalCDF.average_cdf()
      • EmpiricalCDF.cdf()
      • EmpiricalCDF.icdf()
      • EmpiricalCDF.set()
  • cenreg.distribution.linear module
    • LinearCDF
      • LinearCDF.average_cdf()
      • LinearCDF.cdf()
      • LinearCDF.get_boundary_lengths()
      • LinearCDF.icdf()
      • LinearCDF.set_knot_values()
    • LinearQuantileFunction
      • LinearQuantileFunction.average_cdf()
      • LinearQuantileFunction.cdf()
      • LinearQuantileFunction.get_qk_lengths()
      • LinearQuantileFunction.icdf()
      • LinearQuantileFunction.set_knot_values()
  • Module contents
• cenreg.metric package
  • Submodules
  • cenreg.metric.cdf module
    • brier()
    • km_calibration()
    • negative_loglikelihood()
    • nll_sc()
    • ranked_probability_score()
  • cenreg.metric.cjd module
    • brier()
    • kolmogorov_smirnov_calibration_error()
    • negative_loglikelihood()
    • ranked_probability_score()
  • cenreg.metric.quantile module
    • d_calibration()
  • Module contents
• cenreg.model package
  • Submodules
  • cenreg.model.F2cjd_np module
    • convert()
  • cenreg.model.cjd2F_np module
    • cjd2surv()
  • cenreg.model.copula_graphic module
    • CopulaGraphicDistribution
      • CopulaGraphicDistribution.average_cdf()
      • CopulaGraphicDistribution.cdf()
      • CopulaGraphicDistribution.icdf()
      • CopulaGraphicDistribution.survival_function()
    • estimate()
  • cenreg.model.copula_np module
    • FrankCopula
      • FrankCopula.cdf()
    • IndependenceCopula
      • IndependenceCopula.cdf()
    • SurvivalCopula
      • SurvivalCopula.cdf()
    • create()
  • cenreg.model.kaplan_meier module
    • KaplanMeierDistribution
      • KaplanMeierDistribution.average_cdf()
      • KaplanMeierDistribution.cdf()
      • KaplanMeierDistribution.fit()
      • KaplanMeierDistribution.icdf()
      • KaplanMeierDistribution.survival_function()
    • createCdfLinear()
    • createQuantilesLinear()
  • cenreg.model.nonparametric module
  • Module contents
    • cjd2surv()
• cenreg.pytorch package
  • Submodules
  • cenreg.pytorch.cjd2F module
    • MseModel
      • MseModel.configure_optimizers()
      • MseModel.forward()
      • MseModel.loss()
    • minimize_mse()
  • cenreg.pytorch.copula_torch module
    • ClaytonCopula
      • ClaytonCopula.cdf()
    • FrankCopula
      • FrankCopula.cdf()
    • GumbelCopula
      • GumbelCopula.cdf()
    • IndependenceCopula
      • IndependenceCopula.cdf()
    • SurvivalCopula
      • SurvivalCopula.cdf()
    • create()
  • cenreg.pytorch.datamodule module
    • ProbDataModule
      • ProbDataModule.test_dataloader()
      • ProbDataModule.train_dataloader()
      • ProbDataModule.val_dataloader()
    • ProbDataset
      • ProbDataset.len_feature()
    • SurvDataModule
      • SurvDataModule.test_dataloader()
      • SurvDataModule.train_dataloader()
      • SurvDataModule.val_dataloader()
    • SurvDataset
      • SurvDataset.len_feature()
  • cenreg.pytorch.distribution module
    • LinearCDF
      • LinearCDF.average_cdf()
      • LinearCDF.cdf()
      • LinearCDF.get_boundary_lengths()
      • LinearCDF.icdf()
      • LinearCDF.set_knot_values()
    • LinearQuantileFunction
      • LinearQuantileFunction.average_cdf()
      • LinearQuantileFunction.cdf()
      • LinearQuantileFunction.get_qk_lengths()
      • LinearQuantileFunction.icdf()
      • LinearQuantileFunction.set_knot_values()
  • cenreg.pytorch.loss_cdf module
    • Brier
      • Brier.loss()
    • CNLL_CR
      • CNLL_CR.loss()
    • NegativeLogLikelihood
      • NegativeLogLikelihood.loss()
    • RankedProbabilityScore
      • RankedProbabilityScore.loss()
    • brier()
    • negative_log_likelihood()
    • ranked_probability_score()
  • cenreg.pytorch.loss_cjd module
    • Brier
      • Brier.loss()
    • NegativeLogLikelihood
      • NegativeLogLikelihood.loss()
    • RankedProbabilityScore
      • RankedProbabilityScore.loss()
  • cenreg.pytorch.loss_cont module
    • CopulaNegativeLogLikelihood
      • CopulaNegativeLogLikelihood.loss()
    • NegativeLogLikelihood
      • NegativeLogLikelihood.loss()
  • cenreg.pytorch.mlp module
    • Linear
      • Linear.forward()
    • MLP
      • MLP.forward()
    • MLP_MultiHead
      • MLP_MultiHead.forward()
    • SMM
      • SMM.forward()
    • SMM_MultiHead
      • SMM_MultiHead.forward()
  • cenreg.pytorch.utils module
    • denormalize_pred()
    • normalize_y()
  • Module contents

Submodules

cenreg.utils module

cenreg.utils.create_bins(max_y: float, min_y: float = 0.0, num_bins=10, algorithm: str = 'even') → ndarray

Create bins for discretizing a continuous variable. This function generates evenly spaced bins between min_y and max_y with num_bins-2 intervals, and generates an additional bin at the end to include the value exceeding max_y.

Parameters:

• max_y (float) – Maximum value of y.

• min_y (float) – Minimum value of y.

• num_bins (int) – Number of bins.

Returns:

bins – Array of bin edges.

Return type:

np.ndarray

cenreg.utils.create_discretized_labels(bins: ndarray, num_risks: int, t: ndarray, e: ndarray) → ndarray

Create discretized labels for survival analysis.

Parameters:

• bins (np.ndarray) – Array of bin edges.

• num_risks (int) – Number of risks (or categories) to be assigned to each bin.

• t (np.ndarray) – Array of time values.

• e (np.ndarray) – Array of event indicators (1 for event, 0 for censored).

Returns:

Array of discretized labels, where each label corresponds to a bin and a risk category.

Return type:

np.ndarray

Module contents