From 7ccf6160ece5504b75a7597a95073833274ba08b Mon Sep 17 00:00:00 2001
From: Mahmoud Bentriou <mahmoud.bentriou@centralesupelec.fr>
Date: Thu, 20 May 2021 23:19:10 +0200
Subject: [PATCH] change doc of abc methods
---
algorithms/abc_model_choice.jl | 33 ++++++++++++++++-----------------
algorithms/abc_smc.jl | 17 ++++++++++++-----
2 files changed, 28 insertions(+), 22 deletions(-)
diff --git a/algorithms/abc_model_choice.jl b/algorithms/abc_model_choice.jl
index a1ffede..5760591 100644
--- a/algorithms/abc_model_choice.jl
+++ b/algorithms/abc_model_choice.jl
@@ -23,10 +23,10 @@ function getproperty(dataset::AbcModelChoiceDataset, sym::Symbol)
end
"""
- `abc_model_choice_dataset(models, models_prior,
- summary_stats_observations,
- summary_stats_func::Function, distance_func::Function,
- k::Int, N_ref::Int; dir_results::Union{Nothing,String} = nothing)`
+ abc_model_choice_dataset(models, models_prior,
+ summary_stats_observations,
+ summary_stats_func::Function, distance_func::Function,
+ k::Int, N_ref::Int; dir_results::Union{Nothing,String} = nothing)
Creates a reference table for ABC model choice.
@@ -56,10 +56,10 @@ function abc_model_choice_dataset(models::Vector{<:Union{Model,ParametricModel}}
end
"""
- `abc_model_choice_dataset(models, models_prior,
- summary_stats_observations,
- summary_stats_func::Function, distance_func::Function,
- k::Int, N_ref::Int; dir_results::Union{Nothing,String} = nothing)`
+ abc_model_choice_dataset(models,
+ summary_stats_observations,
+ summary_stats_func::Function, distance_func::Function,
+ k::Int, N_ref::Int; dir_results::Union{Nothing,String} = nothing)
Creates a reference table for ABC model choice with discrete uniform prior distribution over the models.
"""
@@ -161,10 +161,10 @@ function _distributed_abc_model_choice_dataset(models::Vector{<:Union{Model,Para
end
"""
- `rf_abc_model_choice(models, summary_stats_observations,
- summary_stats_func::Function, N_ref::Int;
- k::Int = N_ref, distance_func::Function = (x,y) -> 1,
- hyperparameters_range::Dict)`
+ rf_abc_model_choice(models, summary_stats_observations,
+ summary_stats_func::Function, N_ref::Int;
+ k::Int = N_ref, distance_func::Function = (x,y) -> 1,
+ hyperparameters_range::Dict)
Run the Random Forest Approximate Bayesian Computation model choice method.
@@ -179,8 +179,8 @@ The optional arguments are:
* `k`: the k nearest samples from the observations to keep in the reference table (by default: k = N_ref)
* `distance_func`: the distance function, has to be defined if k < N_ref
* `hyperparameters_range`: a dict with the hyperparameters range values for the cross validation
-fit of the Random Forest (by default: `Dict(:n_estimators => [200], :min_samples_leaf => [1], :min_samples_split => [2])`).
-See scikit-learn documentation of RandomForestClassifier for the hyperparameters name.
+ fit of the Random Forest (by default: `Dict(:n_estimators => [200], :min_samples_leaf => [1], :min_samples_split => [2])`).
+ See scikit-learn documentation of RandomForestClassifier for the hyperparameters name.
The result is a `RandomForestABC` object with fields:
* `reference_table` an AbcModelChoiceDataset that corresponds to the reference table of the algorithm,
@@ -205,11 +205,10 @@ function rf_abc_model_choice(models::Vector{<:Union{Model,ParametricModel}},
end
"""
- `posterior_proba_model(rf_abc::RandomForestABC)`
+ posterior_proba_model(rf_abc::RandomForestABC)
-Estimates the posterior probability of the model with the Random Forest ABC method.
+Estimates the posterior probability of the model ``P(M = \\widehat{M}(s_{obs}) | s_{obs})`` with the Random Forest ABC method.
"""
-# P(m = m^(ss_obs) | ss_obs) estimate
function posterior_proba_model(rf_abc::RandomForestABC)
oob_votes = rf_abc.clf.oob_decision_function_
y_pred_oob = argmax.([oob_votes[i,:] for i = 1:size(oob_votes)[1]])
diff --git a/algorithms/abc_smc.jl b/algorithms/abc_smc.jl
index 23f5dd3..fe74b14 100644
--- a/algorithms/abc_smc.jl
+++ b/algorithms/abc_smc.jl
@@ -24,6 +24,12 @@ struct ResultAbc
weights::Vector{Float64}
l_ess::Vector{Float64}
end
+"""
+ automaton_abc(pm::ParametricModel, l_obs, func_dist; nbr_particles, alpha, kernel_type, NT
+ duration_time, bound_sim, sym_var_aut, verbose)
+
+Run the Automaton-ABC-SMC algorithm with the pm parametric model. `pm.m` has to be a `SynchronizedModel`.
+"""
function automaton_abc(pm::ParametricModel;
nbr_particles::Int = 100, tolerance::Float64 = 0.0, alpha::Float64 = 0.75, kernel_type = "mvnormal",
@@ -57,18 +63,19 @@ function automaton_abc(pm::ParametricModel;
end
"""
- `abc_smc(pm::ParametricModel, l_obs, func_dist; nbr_particles, alpha, kernel_type, NT
- duration_tiùe, bound_sim, sym_var_aut, verbose)`
+ abc_smc(pm::ParametricModel, l_obs, func_dist; nbr_particles, alpha, kernel_type, NT
+ duration_time, bound_sim, sym_var_aut, verbose)
Run the ABC-SMC algorithm with the pm parametric model.
+
`func_dist(l_sim, l_obs)` is the distance function between simulations and observation,
-it corresponds to \$\rho(\eta(y_sim), \eta(y_exp))\$.
+it corresponds to ``\\rho(\\eta(y_sim), \\eta(y_exp))\\``.
`l_obs::Vector{<:T2}` is a collection of observations.
`dist` must have a signature of the form `func_dist(l_sim::Vector{T1}, l_obs::Vector{T2})`.
-If pm is defined on a ContinuousTimeModel, then `T1` should verify `T1 <: Trajectory`.
-!!! Distance function and distributed ABC
+If `pm` is defined on a `ContinuousTimeModel`, then `T1` should verify `T1 <: Trajectory`.
+!!! Distance function and distributed ABC
If you use `abc_smc` with multiple workers, `dist` has to be defined
on every workers by using @everywhere.
--
GitLab