riantr/moonbit_dowhy/overrule

README
- ::causal_graphCausalModel::causal_graph
- ::dataCausalModel::data
- ::estimand_typeCausalModel::estimand_type
- ::estimate_effectCausalModel::estimate_effect
- ::get_common_causesCausalModel::get_common_causes
- ::get_effect_modifiersCausalModel::get_effect_modifiers
- ::get_estimatorCausalModel::get_estimator
- ::get_instrumentsCausalModel::get_instruments
- ::graph_with_rolesCausalModel::graph_with_roles
- ::identified_estimandCausalModel::identified_estimand
- ::identifies_variablesCausalModel::identifies_variables
- ::identify_effectCausalModel::identify_effect
- ::newCausalModel::new
- ::new_with_graph_stringCausalModel::new_with_graph_string
- ::proceeds_when_unidentifiableCausalModel::proceeds_when_unidentifiable
- ::refute_assess_overlapCausalModel::refute_assess_overlap
- ::refute_estimateCausalModel::refute_estimate
- ::summaryCausalModel::summary
- ::unobserved_graph_variablesCausalModel::unobserved_graph_variables
- ::unused_data_variablesCausalModel::unused_data_variables
V
is_supported_graph_format

riantr/moonbit_dowhy/overrule does not have a README file

Rule

type Rule = Array[RuleAtom]

A rule of a ruleset: the literals that have to hold at the same time.

RuleSet

source

type RuleSet = Array[Array[RuleAtom]]

The rules of a ruleset estimator.

BeamSearchResult

source

pub struct BeamSearchResult {
  values : Array[Double]
  z : Array[Array[Bool]]
}

The result of the beam search.

BinarizedData

source

pub struct BinarizedData {
  literals : Array[Literal]
  columns : Array[Array[Double]]
  feature_of : Array[Int]
  n_rows : Int
}

The binarised features of a data set.

BooleanRule

source

pub struct BooleanRule {
  alpha : Double
  lambda0 : Double
  lambda1 : Double
  k : Int
  d : Int
  b : Int
  iter_max : Int
  eps : Double
  rounding : RoundingStrategy
  w : Array[Double]
  w_raw : Array[Double]
  z : Array[Array[Bool]]
  lp_objective : Double
}

The estimator of a boolean rule set, the port of the OverlapBooleanRule of the upstream.

The rules are learned with a column generation: a program is solved, the conjunctions that are worth adding are searched for with a beam search, and the program is solved again with the new conjunctions. The coefficients of the program are rounded into rules at the end.

BooleanRule::covers

source

fn BooleanRule::covers(self : BooleanRule, coverage : Array[Double]) -> Bool

Whether a rule covers the given row.

BooleanRule::fit

source

fn BooleanRule::fit(self : BooleanRule, data : BinarizedData, y : Array[Double]) -> Unit

Learn the rules of the given rows.

The labels of the rows are one for the rows that have to be covered, zero for the rows that have to be avoided and minus one for the background, the way the upstream labels them.

BooleanRule::lp_value

source

fn BooleanRule::lp_value(self : BooleanRule) -> Double

The value of the program of the rules.

BooleanRule::new

source

fn BooleanRule::new(alpha : Double, lambda0 : Double, lambda1 : Double, k : Int, d : Int, b : Int, iter_max : Int, eps : Double, rounding : RoundingStrategy) -> BooleanRule

The estimator of a boolean rule set.

BooleanRule::predict

source

fn BooleanRule::predict(self : BooleanRule, data : BinarizedData) -> Array[Bool]

Whether the rules cover every row of the given covariates.

BooleanRule::rules

source

fn BooleanRule::rules(self : BooleanRule, literals : Array[Literal]) -> Array[Array[RuleAtom]]

The rules of the estimator.

FeatureBinarizer

source

pub struct FeatureBinarizer {
  names : Array[String]
  negations : Bool
  num_thresholds : Int
  kind : Array[FeatureKind]
  levels : Array[Array[Double]]
  thresholds : Array[Array[Double]]
}

The binariser of the features, the port of the FeatureBinarizer of the upstream.

Every feature is turned into a set of literals, and every literal holds for the rows whose value of the feature satisfies it. A feature with at most two distinct values becomes one literal and, with the negations, its negation. A feature with more distinct values becomes one literal per threshold, and the thresholds are the quantiles of the values.

FeatureBinarizer::fit

source

fn FeatureBinarizer::fit(self : FeatureBinarizer, columns : Array[Array[Double]]) -> Unit

Learn the literals of the features from the given rows.

The binariser keeps the distinct values of the binary features and the thresholds of the ordinal ones, so that it can transform other rows with the same literals.

FeatureBinarizer::new

source

fn FeatureBinarizer::new(names : Array[String], num_thresholds : Int, negations : Bool) -> FeatureBinarizer

A binariser of the features that is not fitted yet.

FeatureBinarizer::transform

source

fn FeatureBinarizer::transform(self : FeatureBinarizer, columns : Array[Array[Double]]) -> BinarizedData

Turn the given rows into the literals of the features.

FeatureKind

source

pub enum FeatureKind {
  Binary
  Ordinal
}

The kind of a feature of the binariser.

Literal

source

pub struct Literal {
  feature : String
  operation : LiteralOperation
  value : Double
}

A literal of a binarised feature: a condition on the value of a feature.

LiteralOperation

source

pub enum LiteralOperation {
  Binary
  Negation
  LessEqual
  Greater
}

The operation of a literal of a binarised feature.

LiteralOperation::name

source

fn LiteralOperation::name(self : LiteralOperation) -> String

The name of the operation, the way the upstream reports it in a rule.

OverlapAnalyzer

source

pub struct OverlapAnalyzer {
  backdoor_vars : Array[String]
  treatment_name : String
  support_config : SupportConfig
  overlap_config : OverlapConfig
  overlap_eps : Double
  support_only : Bool
  overlap_only : Bool
  support_estimator : RulesetEstimator?
  overlap_estimator : RulesetEstimator?
  support_indicator : Array[Bool]
  overlap_indicator : Array[Bool]
  raw_overlap : Array[Bool]
  fitted : Bool
}

The analyzer of the support and the overlap rules, the port of the OverruleAnalyzer of the upstream.

The support rules describe the region of the covariates the data cover, and the overlap rules describe the region where the treatment and the control overlap, which is found with the propensity scores.

OverlapAnalyzer::fit

source

fn OverlapAnalyzer::fit(self : OverlapAnalyzer, columns : Array[Array[Double]], treatment : Array[Double]) -> Unit

Learn the rules of the given covariates.

The columns are the backdoor variables and the treatment is the column of the treatment, the way the upstream takes them.

OverlapAnalyzer::new

source

fn OverlapAnalyzer::new(backdoor_vars : Array[String], treatment_name : String, support_config : SupportConfig, overlap_config : OverlapConfig, overlap_eps? : Double, support_only? : Bool, overlap_only? : Bool) -> OverlapAnalyzer

An analyzer of the support and the overlap rules.

The range of the propensity scores that describes the overlap region is (overlap_eps, 1 - overlap_eps), and only the support or only the overlap rules are learned when the analyzer is asked for one of them.

OverlapAnalyzer::overlap_indicator

source

fn OverlapAnalyzer::overlap_indicator(self : OverlapAnalyzer) -> Array[Bool]

The rows of the overlap region.

OverlapAnalyzer::overlap_objective

source

fn OverlapAnalyzer::overlap_objective(self : OverlapAnalyzer) -> Double

The value of the program of the overlap rules.

OverlapAnalyzer::overlap_rules

source

fn OverlapAnalyzer::overlap_rules(self : OverlapAnalyzer) -> Array[Array[RuleAtom]]

The rules of the overlap region.

OverlapAnalyzer::predict_overlap_support

source

fn OverlapAnalyzer::predict_overlap_support(self : OverlapAnalyzer, columns : Array[Array[Double]]) -> Array[Bool]

The rows of the covariates that the support rules cover and the overlap rules cover.

OverlapAnalyzer::support_indicator

source

fn OverlapAnalyzer::support_indicator(self : OverlapAnalyzer) -> Array[Bool]

The rows of the support.

OverlapAnalyzer::support_objective

source

fn OverlapAnalyzer::support_objective(self : OverlapAnalyzer) -> Double

The value of the program of the support rules.

OverlapAnalyzer::support_rules

source

fn OverlapAnalyzer::support_rules(self : OverlapAnalyzer) -> Array[Array[RuleAtom]]

The rules of the support.

OverlapConfig

source

pub struct OverlapConfig {
  alpha : Double
  lambda0 : Double
  lambda1 : Double
  k : Int
  d : Int
  b : Int
  iter_max : Int
  num_thresholds : Int
  rounding : RoundingStrategy
}

The configuration of the overlap rules, the port of the OverlapConfig of the upstream.

OverlapConfig::new

source

fn OverlapConfig::new() -> OverlapConfig

The configuration of the overlap rules, with the defaults of the upstream.

RoundingStrategy

source

pub enum RoundingStrategy {
  Greedy
  GreedySweep
}

The strategy of the rounding of the coefficients of the rules.

RoundingStrategy::greedy

source

fn RoundingStrategy::greedy() -> RoundingStrategy

The rounding with a fixed reward for the positive samples.

RuleAtom

source

pub struct RuleAtom {
  feature : String
  operation : String
  value : String
}

An atom of a rule, the way the upstream reports it.

RulesetConfig

source

pub struct RulesetConfig {
  n_ref_multiplier : Double
  alpha : Double
  lambda0 : Double
  lambda1 : Double
  k : Int
  d : Int
  b : Int
  iter_max : Int
  eps : Double
  num_thresholds : Int
  negations : Bool
  rounding : RoundingStrategy
  seed : Int
}

The configuration of a ruleset estimator, the port of the parameters of the BCSRulesetEstimator of the upstream.

RulesetEstimator

source

pub struct RulesetEstimator {
  config : RulesetConfig
  binarizer : FeatureBinarizer
  literals : Array[Literal]
  rule : BooleanRule
  fitted : Bool
}

The estimator of a boolean rule set, the port of the BCSRulesetEstimator of the upstream.

The estimator learns the rules of a region of the covariates: the support of the data or the region where the treatment and the control overlap.

RulesetEstimator::fit

source

fn RulesetEstimator::fit(self : RulesetEstimator, columns : Array[Array[Double]], names : Array[String], labels : Array[Double]) -> Unit

Learn the rules of the given rows.

The labels of the rows are one for the rows the rules have to cover and zero for the rows they have to avoid. The reference samples are the background the rules are learned against.

RulesetEstimator::new

source

fn RulesetEstimator::new(config : RulesetConfig) -> RulesetEstimator

An estimator of a boolean rule set.

RulesetEstimator::objective

source

fn RulesetEstimator::objective(self : RulesetEstimator) -> Double

The value of the program of the rules.

RulesetEstimator::predict

source

fn RulesetEstimator::predict(self : RulesetEstimator, columns : Array[Array[Double]]) -> Array[Bool]

Whether the rules cover every row of the given covariates.

RulesetEstimator::rules

source

fn RulesetEstimator::rules(self : RulesetEstimator) -> Array[Array[RuleAtom]]

The rules of the estimator.

SupportConfig

source

pub struct SupportConfig {
  n_ref_multiplier : Double
  seed : Int
  alpha : Double
  lambda0 : Double
  lambda1 : Double
  k : Int
  d : Int
  b : Int
  iter_max : Int
  num_thresholds : Int
  rounding : RoundingStrategy
}

The configuration of the support rules, the port of the SupportConfig of the upstream.

SupportConfig::new

source

fn SupportConfig::new() -> SupportConfig

The configuration of the support rules, with the defaults of the upstream.

beam_search

source

fn beam_search(r : Array[Double], data : BinarizedData, lambda0 : Double, lambda1 : Double, k : Int, ub : Double, eps : Double, b : Int, d : Int, w_lb : Double) -> BeamSearchResult

Search the conjunctions with a negative reduced cost, the port of the beam search of the upstream.

The search starts from the empty conjunction and adds one literal at a time, keeping the best candidates of every degree. The value of a conjunction is the sum of the residuals of the rows it covers and the regularisation of its literals, so the candidates are the literals that make the value the smallest.

distinct_values

source

fn distinct_values(values : Array[Double]) -> Array[Double]

The distinct values of a column, in the increasing order.

refute_assess_overlap

source

fn refute_assess_overlap(columns : Array[Array[Double]], names : Array[String], backdoor_vars : Array[String], treatment_name : String, support_config : SupportConfig, overlap_config : OverlapConfig, overlap_eps : Double, support_only : Bool, overlap_only : Bool) -> OverlapAnalyzer

Learn the rules of the support and of the overlap of the covariates.

The rules are learned with respect to the backdoor variables only, and the treatment is the column of the treatment, the way the upstream takes them. The columns that are not named are left out of the rules.

sample_reference

source

fn sample_reference(columns : Array[Array[Double]], n : Int, random : @riantr/moonbit_dowhy/utils.Random
Random, seed : Int) -> Array[Array[Double]]

Draw the reference samples of the upstream, the port of sample_reference.

The reference samples are drawn uniformly over the range of every column, and they are the background the rules are learned against. A column with two distinct values is drawn from its two values, and a column with more distinct values is drawn uniformly between its smallest and its largest value, the way the upstream does it with a generator of its own for every column.

Source Files

Install

Download zip

MoonBit port of DoWhy 0.14 (PyWhy). Causal graph construction and analysis, effect identification (backdoor / frontdoor / complete adjustment), estimation (linear regression, generalized linear model, propensity score weighting / matching / stratification, doubly robust, instrumental variable, two stage, regression discontinuity), refutation (bootstrap, placebo treatment, data subset, dummy outcome, unobserved common cause, the assess overlap refuter that learns the rules of the support and of the overlap with boolean rules), intervention sampling (do samplers), temporal graphs, causal discovery (the PC algorithm with the Fisher-z test) and the portable core of the graphical causal models (the p-value merging, the bootstrap confidence intervals, the Kullback-Leibler divergences, the density estimators, the Shapley values, the independence tests, the prediction models, the causal mechanisms, the fitting of a causal model, the what-if queries (the interventional samples, the counterfactual samples and the average causal effect), the validation layer (the kernel based conditional independence test, the validation of a causal structure and the validation of an invertible model) and the falsification layer (the local Markov condition, the pairwise dependencies, the parental d-separations, the causal minimality and the permutation test of a graph) and the model evaluation layer (the normalised mean squared error, the continuous ranked probability score, the performance of the causal mechanisms, the test of the invertibility assumption and the report of the evaluation) and the distribution change layer (the test of the change of a mechanism, the attribution of the change of the distribution of a target node, the multiply-robust attribution of the change of the mean or of the variance of a target node and the scores of the change of the distribution of every node). Pure MoonBit with zero external runtime dependencies.

causal-inferencedowhycausal-graphcausal-discoverybackdoorfrontdoorpropensity-scoretreatment-effectrefutationoverlapidentifiersestimatorsshapley-valuesindependence-teststructural-causal-modelcausal-mechanismwhat-ifcounterfactual

riantr

MIT

gitee.com/ren-yongxiang/moonbit_doWhy

Updated 10 hours ago

130 downloads

Site source Report issue Modules Skills Build queue Statistics

#Rule

#RuleSet

#BeamSearchResult

#BinarizedData

#BooleanRule

#BooleanRule::covers

#BooleanRule::fit

#BooleanRule::lp_value

#BooleanRule::new

#BooleanRule::predict

#BooleanRule::rules

#FeatureBinarizer

#FeatureBinarizer::fit

#FeatureBinarizer::new

#FeatureBinarizer::transform

#FeatureKind

#Literal

#LiteralOperation

#LiteralOperation::name

#OverlapAnalyzer

#OverlapAnalyzer::fit

#OverlapAnalyzer::new

#OverlapAnalyzer::overlap_indicator

#OverlapAnalyzer::overlap_objective

#OverlapAnalyzer::overlap_rules

#OverlapAnalyzer::predict_overlap_support

#OverlapAnalyzer::support_indicator

#OverlapAnalyzer::support_objective

#OverlapAnalyzer::support_rules

#OverlapConfig

#OverlapConfig::new

#RoundingStrategy

#RoundingStrategy::greedy

#RuleAtom

#RulesetConfig

#RulesetEstimator

#RulesetEstimator::fit

#RulesetEstimator::new

#RulesetEstimator::objective

#RulesetEstimator::predict

#RulesetEstimator::rules

#SupportConfig

#SupportConfig::new

#beam_search

#distinct_values

#refute_assess_overlap

#sample_reference

Source Files

Rule

RuleSet

BeamSearchResult

BinarizedData

BooleanRule

BooleanRule::covers

BooleanRule::fit

BooleanRule::lp_value

BooleanRule::new

BooleanRule::predict

BooleanRule::rules

FeatureBinarizer

FeatureBinarizer::fit

FeatureBinarizer::new

FeatureBinarizer::transform

FeatureKind

Literal

LiteralOperation

LiteralOperation::name

OverlapAnalyzer

OverlapAnalyzer::fit

OverlapAnalyzer::new

OverlapAnalyzer::overlap_indicator

OverlapAnalyzer::overlap_objective

OverlapAnalyzer::overlap_rules

OverlapAnalyzer::predict_overlap_support

OverlapAnalyzer::support_indicator

OverlapAnalyzer::support_objective

OverlapAnalyzer::support_rules

OverlapConfig

OverlapConfig::new

RoundingStrategy

RoundingStrategy::greedy

RuleAtom

RulesetConfig

RulesetEstimator

RulesetEstimator::fit

RulesetEstimator::new

RulesetEstimator::objective

RulesetEstimator::predict

RulesetEstimator::rules

SupportConfig

SupportConfig::new

beam_search

distinct_values

refute_assess_overlap

sample_reference