API

This section provides comprehensive details about the exposed ReLife API. ReLife is structured into different modules, each with a clear and specific role. We divided the API documentation close to the same logic.

ParametricModel	Base class of every parametric models in ReLife.
MaximumLikelihoodOptimizer	Abstract maximum likelihood optimizer.

Lifetime models

Parametric lifetime models

Parametric lifetime distributions

Exponential	Exponential lifetime distribution.
Weibull	Weibull lifetime distribution.
Gompertz	Gompertz lifetime distribution.
Gamma	Gamma lifetime distribution.
LogLogistic	Log-logistic probability distribution.

Parametric lifetime regressions

ParametricProportionalHazard	Proportional Hazard regression.
ParametricAcceleratedFailureTime	Accelerated failure time regression.

Semiparametric lifetime regression

SemiParametricProportionalHazard

Class for Cox, semi-parametric, Proportional Hazards, model

Non parametric lifetime models

KaplanMeier	Kaplan-Meier estimator.
NelsonAalen	Nelson-Aalen estimator.
ECDF	Empirical Cumulative Distribution Function.

Conditional lifetime models

Conditional lifetime models

LeftTruncatedModel	Left truncated model.
AgeReplacementModel	Age replacement model.

Stochastic processes

RenewalProcess	Renewal process.
RenewalRewardProcess	Renewal reward process.
NonHomogeneousPoissonProcess	Non-homogeneous Poisson process.

Maintenance policies

Convenient functions to instanciate a policy based on the given model and cost structure.

age_replacement_policy	Creates a preventive age replacement policy.
run_to_failure_policy	Creates a run-to-failure policy.

Lower-level policy classes that can be used in addition to the above functions.

AgeReplacementPolicy	Age replacement policy.
OneCycleAgeReplacementPolicy	One-cyle age replacement policy.
RunToFailurePolicy	Run-to-failure renewal policy.
OneCycleRunToFailurePolicy	One cyle run-to-failure policy.
NonHomogeneousPoissonAgeReplacementPolicy	Age replacement policy for non-Homogeneous Poisson process.

Rewards

RunToFailureReward	Run-to-failure reward.
AgeReplacementReward	Age replacement reward.
ExponentialDiscounting	Exponential discounting.

Built-in datasets

load_circuit_breaker	Load example dataset of circuit breaker lifetimes.
load_insulator_string	Load example dataset of insulator string lifetimes with covariates.
load_power_transformer	Load example dataset of power transformers lifetimes.

Likelihoods

LifetimeLikelihood	Maximum likelihood estimator from lifetime data.
CoxPartialLifetimeLikelihood
EfronPartialLifetimeLikelihood
BreslowPartialLifetimeLikelihood

Quadratures

Quadratures are used a many computations. We don’t use Scipy quadrature implementations as, to our knowledge, they don’t support automatic broadcasting of 2D bounds.

legendre_quadrature	Numerical integration of \(f(x)\) over the interval \([a,b]\)
laguerre_quadrature	Numerical integration of \(f(x) * exp(-x)\) over the interval \([a, \infty]\).
unweighted_laguerre_quadrature	Numerical integration of \(f(x)\) over the interval \([a, \infty]\)

Utils

Various utilities to help with development.

to_numpy_float64	Convert the input to np.float64 if it is a scalar or an array of np.float64 otherwise.
to_column_2d_if_1d	Reshapes to column array.
flatten_if_at_least_2d	Flatten array-like object when possible.
get_nb_assets	Gets the number of assets encoded in args.