Keyboard shortcuts

Press ← or → to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

KfoldSplit

Utility sklearn-compatible ⚙️ Model Selection

K-Fold split — returns train/test index arrays for k-fold cross-validation. / Division K-Fold — retourne les tableaux d'indices train/test pour la validation croisée k-fold.

⚡ Rust-native ✓ sklearn parity
Quick start — Python 
import seraplot as sp, json, numpy as np
X = np.random.randn(100, 4)
res = json.loads(sp.ml_kfold_split(json.dumps({"X_train": X.tolist(), "n_splits": 5})))
print(len(res["folds"]))
💡
EN — Drop-in replacement: sp.KfoldSplit has the same API as sklearn.
FR — Remplacement direct : même API que sklearn, changez l'import.

API Reference

JSON function name

ml_kfold_split — aliases: kfold_split

Python class
sp.KfoldSplit(n_splits=5, stratified=false, seed=42)
Constructor Parameters
ParameterTypeDefaultDescription
n_splitsint5Number of folds.
stratifiedboolfalseUse stratified splits (requires y_train).
seedint42Random seed.
Returns

JSON with folds array, each fold has train_indices, test_indices.

Example
import seraplot as sp, json, numpy as np
X = np.random.randn(100, 4)
res = json.loads(sp.ml_kfold_split(json.dumps({"X_train": X.tolist(), "n_splits": 5})))
print(len(res["folds"]))

Référence API

Nom de fonction JSON

ml_kfold_split — alias : kfold_split

Classe Python
sp.KfoldSplit(n_splits=5, stratified=false, seed=42)
Paramètres du constructeur
ParamètreTypeDéfautDescription
n_splitsint5Nombre de plis.
stratifiedboolfalseUtiliser des splits stratifiés (nécessite y_train).
seedint42Graine aléatoire.
Retourne

JSON avec tableau folds, chaque pli a train_indices, test_indices.

Exemple
import seraplot as sp, json, numpy as np
X = np.random.randn(100, 4)
res = json.loads(sp.ml_kfold_split(json.dumps({"X_train": X.tolist(), "n_splits": 5})))
print(len(res["folds"]))

CrossValScore

Utility sklearn-compatible ⚙️ Model Selection

Cross-validation score — evaluates model performance across k folds. / Score de validation croisée — évalue la performance du modèle sur k plis.

⚡ Rust-native ✓ sklearn parity
Quick start — Python 
import seraplot as sp, json, numpy as np
X = np.random.randn(200, 5)
y = X @ [1, -1, 0.5, 0, 1] + np.random.randn(200) * 0.3
payload = {"X_train": X.tolist(), "y_train": y.tolist(), "model": "ridge", "n_splits": 5}
res = json.loads(sp.ml_cross_val_score(json.dumps(payload)))
print(res["scores"], res["mean_score"])
💡
EN — Drop-in replacement: sp.CrossValScore has the same API as sklearn.
FR — Remplacement direct : même API que sklearn, changez l'import.

API Reference

JSON function name

ml_cross_val_score — aliases: cross_val_score

Python class
sp.CrossValScore(model=ridge, n_splits=5, scoring=auto)
Constructor Parameters
ParameterTypeDefaultDescription
modelstrridgeModel name for evaluation.
n_splitsint5Number of folds.
scoringstrautoScore metric.
Returns

JSON with scores array and mean_score.

Algorithm

$$\text{CV}(k) = \frac{1}{k}\sum_{i=1}^{k} s_i$$

Example
import seraplot as sp, json, numpy as np
X = np.random.randn(200, 5)
y = X @ [1, -1, 0.5, 0, 1] + np.random.randn(200) * 0.3
payload = {"X_train": X.tolist(), "y_train": y.tolist(), "model": "ridge", "n_splits": 5}
res = json.loads(sp.ml_cross_val_score(json.dumps(payload)))
print(res["scores"], res["mean_score"])

Référence API

Nom de fonction JSON

ml_cross_val_score — alias : cross_val_score

Classe Python
sp.CrossValScore(model=ridge, n_splits=5, scoring=auto)
Paramètres du constructeur
ParamètreTypeDéfautDescription
modelstrridgeNom du modèle à évaluer.
n_splitsint5Nombre de plis.
scoringstrautoMétrique de score.
Retourne

JSON avec tableau scores et mean_score.

Algorithme

$$\text{CV}(k) = \frac{1}{k}\sum_{i=1}^{k} s_i$$

Exemple
import seraplot as sp, json, numpy as np
X = np.random.randn(200, 5)
y = X @ [1, -1, 0.5, 0, 1] + np.random.randn(200) * 0.3
payload = {"X_train": X.tolist(), "y_train": y.tolist(), "model": "ridge", "n_splits": 5}
res = json.loads(sp.ml_cross_val_score(json.dumps(payload)))
print(res["scores"], res["mean_score"])