Relax number of points required wrt regression spec (#244)

* Relax required nb of points wrt regression spec

* Refactor type to specify the number of clusters

* Linting

* Fix docstring

crates/ego/examples/mopta08.rs

@@ -1,6 +1,6 @@
 use clap::Parser;
 use egobox_ego::{EgorBuilder, GroupFunc, InfillOptimizer};
-use egobox_moe::{CorrelationSpec, RegressionSpec};
+use egobox_moe::{CorrelationSpec, NbClusters, RegressionSpec};
 use ndarray::{s, Array1, Array2, ArrayView1, ArrayView2};
 use std::fs::{remove_file, File};
 use std::io::prelude::*;
@@ -267,7 +267,7 @@ fn main() -> anyhow::Result<()> {
             config
                 .n_cstr(N_CSTR)
                 .cstr_tol(cstr_tol.clone())
-                .n_clusters(1)
+                .n_clusters(NbClusters::fixed(1))
                 .n_start(50)
                 .n_doe(n_doe)
                 .max_iters(max_iters)

crates/ego/src/egor.rs

@@ -361,6 +361,7 @@ mod tests {
     use approx::assert_abs_diff_eq;
     use argmin_testfunctions::rosenbrock;
     use egobox_doe::{Lhs, SamplingMethod};
+    use egobox_moe::NbClusters;
     use ndarray::{array, s, Array1, Array2, ArrayView2, Ix1, Zip};
 
     use ndarray_npy::read_npy;
@@ -513,7 +514,7 @@ mod tests {
     #[serial]
     fn test_xsinx_auto_clustering_egor_builder() {
         let res = EgorBuilder::optimize(xsinx)
-            .configure(|config| config.n_clusters(0).max_iters(20))
+            .configure(|config| config.n_clusters(NbClusters::auto()).max_iters(20))
             .min_within(&array![[0.0, 25.0]])
             .run()
             .expect("Egor with auto clustering should minimize xsinx");

crates/ego/src/gpmix/mixint.rs

@@ -10,7 +10,7 @@ use egobox_gp::metrics::CrossValScore;
 use egobox_gp::ThetaTuning;
 use egobox_moe::{
     Clustered, Clustering, CorrelationSpec, FullGpSurrogate, GpMixture, GpMixtureParams,
-    GpSurrogate, GpSurrogateExt, MixtureGpSurrogate, RegressionSpec,
+    GpSurrogate, GpSurrogateExt, MixtureGpSurrogate, NbClusters, RegressionSpec,
 };
 use linfa::traits::{Fit, PredictInplace};
 use linfa::{DatasetBase, Float, ParamGuard};
@@ -358,8 +358,7 @@ impl MixintGpMixtureValidParams {
                 .surrogate_builder
                 .clone()
                 .check()?
-                .train(&xcast, &yt.to_owned())
-                .unwrap(),
+                .train(&xcast, &yt.to_owned())?,
             xtypes: self.xtypes.clone(),
             work_in_folded_space: self.work_in_folded_space,
             training_data: (xt.to_owned(), yt.to_owned()),
@@ -437,7 +436,7 @@ impl SurrogateBuilder for MixintGpMixtureParams {
     }
 
     /// Sets the number of clusters used by the mixture of surrogate experts.
-    fn set_n_clusters(&mut self, n_clusters: usize) {
+    fn set_n_clusters(&mut self, n_clusters: NbClusters) {
         self.0 = MixintGpMixtureValidParams {
             surrogate_builder: self.0.surrogate_builder.clone().n_clusters(n_clusters),
             xtypes: self.0.xtypes.clone(),

crates/ego/src/gpmix/mod.rs

@@ -5,7 +5,7 @@ pub mod spec;
 
 use egobox_gp::ThetaTuning;
 use egobox_moe::{
-    Clustering, CorrelationSpec, GpMixtureParams, MixtureGpSurrogate, RegressionSpec,
+    Clustering, CorrelationSpec, GpMixtureParams, MixtureGpSurrogate, NbClusters, RegressionSpec,
 };
 use ndarray::{ArrayView1, ArrayView2};
 
@@ -40,7 +40,7 @@ impl SurrogateBuilder for GpMixtureParams<f64> {
     }
 
     /// Sets the number of clusters used by the mixture of surrogate experts.
-    fn set_n_clusters(&mut self, n_clusters: usize) {
+    fn set_n_clusters(&mut self, n_clusters: NbClusters) {
         *self = self.clone().n_clusters(n_clusters);
     }
 

crates/ego/src/solver/egor_config.rs

@@ -2,6 +2,7 @@
 use crate::criteria::*;
 use crate::types::*;
 use crate::HotStartMode;
+use egobox_moe::NbClusters;
 use egobox_moe::{CorrelationSpec, RegressionSpec};
 use ndarray::Array1;
 use ndarray::Array2;
@@ -72,9 +73,9 @@ pub struct EgorConfig {
     /// Optional dimension reduction (see [egobox_moe])
     pub(crate) kpls_dim: Option<usize>,
     /// Number of clusters used by mixture of experts (see [egobox_moe])
-    /// When set to 0 the clusters are computes automatically and refreshed
+    /// When set to Auto the clusters are computes automatically and refreshed
     /// every 10-points (tentative) additions
-    pub(crate) n_clusters: usize,
+    pub(crate) n_clusters: NbClusters,
     /// Specification of a target objective value which is used to stop the algorithm once reached
     pub(crate) target: f64,
     /// Directory to save intermediate results: inital doe + evalutions at each iteration
@@ -108,7 +109,7 @@ impl Default for EgorConfig {
             regression_spec: RegressionSpec::CONSTANT,
             correlation_spec: CorrelationSpec::SQUAREDEXPONENTIAL,
             kpls_dim: None,
-            n_clusters: 1,
+            n_clusters: NbClusters::default(),
             target: f64::NEG_INFINITY,
             outdir: None,
             warm_start: false,
@@ -239,9 +240,8 @@ impl EgorConfig {
 
     /// Sets the number of clusters used by the mixture of surrogate experts.
     ///
-    /// When set to Some(0), the number of clusters is determined automatically
-    /// When set None, default to 1
-    pub fn n_clusters(mut self, n_clusters: usize) -> Self {
+    /// When set to Auto, the number of clusters is determined automatically
+    pub fn n_clusters(mut self, n_clusters: NbClusters) -> Self {
         self.n_clusters = n_clusters;
         self
     }

crates/ego/src/solver/egor_impl.rs

@@ -101,7 +101,8 @@ where
     C: CstrFn,
 {
     pub fn have_to_recluster(&self, added: usize, prev_added: usize) -> bool {
-        self.config.n_clusters == 0 && (added != 0 && added % 10 == 0 && added - prev_added > 0)
+        self.config.n_clusters.is_auto()
+            && (added != 0 && added % 10 == 0 && added - prev_added > 0)
     }
 
     /// Build surrogate given training data and surrogate builder
@@ -124,7 +125,7 @@ where
         builder.set_kpls_dim(self.config.kpls_dim);
         builder.set_regression_spec(self.config.regression_spec);
         builder.set_correlation_spec(self.config.correlation_spec);
-        builder.set_n_clusters(self.config.n_clusters);
+        builder.set_n_clusters(self.config.n_clusters.clone());
 
         if make_clustering
         /* init || recluster */

crates/ego/src/types.rs

@@ -1,7 +1,7 @@
 use crate::gpmix::spec::*;
 use crate::{errors::Result, EgorState};
 use argmin::core::CostFunction;
-use egobox_moe::{Clustering, MixtureGpSurrogate, ThetaTuning};
+use egobox_moe::{Clustering, MixtureGpSurrogate, NbClusters, ThetaTuning};
 use linfa::Float;
 use ndarray::{Array1, Array2, ArrayView1, ArrayView2};
 use serde::{Deserialize, Serialize};
@@ -140,7 +140,7 @@ pub trait SurrogateBuilder: Clone + Serialize + Sync {
     fn set_kpls_dim(&mut self, kpls_dim: Option<usize>);
 
     /// Sets the number of clusters used by the mixture of surrogate experts.
-    fn set_n_clusters(&mut self, n_clusters: usize);
+    fn set_n_clusters(&mut self, n_clusters: NbClusters);
 
     /// Sets the hyperparameters tuning strategy
     fn set_theta_tunings(&mut self, theta_tunings: &[ThetaTuning<f64>]);

crates/moe/examples/clustering.rs

@@ -1,5 +1,5 @@
 use egobox_doe::{Lhs, SamplingMethod};
-use egobox_moe::{GpMixture, Recombination};
+use egobox_moe::{GpMixture, NbClusters, Recombination};
 use linfa::prelude::{Dataset, Fit};
 use ndarray::{arr2, Array, Array2, Axis, Zip};
 use std::error::Error;
@@ -24,7 +24,7 @@ fn main() -> Result<(), Box<dyn Error>> {
     let ds = Dataset::new(xtrain, ytrain.remove_axis(Axis(1)));
     let moe1 = GpMixture::params().fit(&ds)?;
     let moe3 = GpMixture::params()
-        .n_clusters(3)
+        .n_clusters(NbClusters::fixed(3))
         .recombination(Recombination::Hard)
         .fit(&ds)?;
 

crates/moe/examples/moe_norm1.rs

@@ -1,6 +1,6 @@
 use csv::ReaderBuilder;
 use egobox_doe::{FullFactorial, SamplingMethod};
-use egobox_moe::GpMixture;
+use egobox_moe::{GpMixture, NbClusters};
 use linfa::{traits::Fit, Dataset};
 use ndarray::{arr2, s, Array2, Axis};
 use ndarray_csv::Array2Reader;
@@ -24,7 +24,9 @@ fn main() -> Result<(), Box<dyn Error>> {
     let xtrain = data_train.slice(s![.., ..2_usize]).to_owned();
     let ytrain = data_train.slice(s![.., 2_usize..]).to_owned();
     let ds = Dataset::new(xtrain, ytrain.remove_axis(Axis(1)));
-    let moe = GpMixture::params().n_clusters(4).fit(&ds)?;
+    let moe = GpMixture::params()
+        .n_clusters(NbClusters::fixed(4))
+        .fit(&ds)?;
 
     let xlimits = arr2(&[[-1., 1.], [-1., 1.]]);
     let xtest = FullFactorial::new(&xlimits).sample(100);

crates/moe/examples/norm1.rs

@@ -1,5 +1,5 @@
 use egobox_doe::{Lhs, SamplingMethod};
-use egobox_moe::{GpMixture, Recombination};
+use egobox_moe::{GpMixture, NbClusters, Recombination};
 use linfa::{traits::Fit, Dataset};
 use ndarray::{arr2, Array2, Axis};
 use std::error::Error;
@@ -14,7 +14,7 @@ fn main() -> Result<(), Box<dyn Error>> {
     let ds = Dataset::new(xtrain, ytrain.remove_axis(Axis(1)));
     let moe1 = GpMixture::params().fit(&ds)?;
     let moe5 = GpMixture::params()
-        .n_clusters(6)
+        .n_clusters(NbClusters::fixed(6))
         .recombination(Recombination::Hard)
         .fit(&ds)?;