deepbox 0.1.0 → 0.2.0

package/dist/datasets/index.cjs

@@ -1,134 +1,134 @@
 'use strict';
 
-var chunkXEG44RF6_cjs = require('../chunk-XEG44RF6.cjs');
-require('../chunk-6AE5FKKQ.cjs');
-require('../chunk-JSCDE774.cjs');
+var chunkG7KXZHG6_cjs = require('../chunk-G7KXZHG6.cjs');
+require('../chunk-724CXHFH.cjs');
+require('../chunk-Z6BGACIH.cjs');
 
 
 
 Object.defineProperty(exports, "DataLoader", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.DataLoader; }
+  get: function () { return chunkG7KXZHG6_cjs.DataLoader; }
 });
 Object.defineProperty(exports, "loadBreastCancer", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadBreastCancer; }
+  get: function () { return chunkG7KXZHG6_cjs.loadBreastCancer; }
 });
 Object.defineProperty(exports, "loadConcentricRings", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadConcentricRings; }
+  get: function () { return chunkG7KXZHG6_cjs.loadConcentricRings; }
 });
 Object.defineProperty(exports, "loadCropYield", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadCropYield; }
+  get: function () { return chunkG7KXZHG6_cjs.loadCropYield; }
 });
 Object.defineProperty(exports, "loadCustomerSegments", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadCustomerSegments; }
+  get: function () { return chunkG7KXZHG6_cjs.loadCustomerSegments; }
 });
 Object.defineProperty(exports, "loadDiabetes", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadDiabetes; }
+  get: function () { return chunkG7KXZHG6_cjs.loadDiabetes; }
 });
 Object.defineProperty(exports, "loadDigits", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadDigits; }
+  get: function () { return chunkG7KXZHG6_cjs.loadDigits; }
 });
 Object.defineProperty(exports, "loadEnergyEfficiency", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadEnergyEfficiency; }
+  get: function () { return chunkG7KXZHG6_cjs.loadEnergyEfficiency; }
 });
 Object.defineProperty(exports, "loadFitnessScores", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadFitnessScores; }
+  get: function () { return chunkG7KXZHG6_cjs.loadFitnessScores; }
 });
 Object.defineProperty(exports, "loadFlowersExtended", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadFlowersExtended; }
+  get: function () { return chunkG7KXZHG6_cjs.loadFlowersExtended; }
 });
 Object.defineProperty(exports, "loadFruitQuality", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadFruitQuality; }
+  get: function () { return chunkG7KXZHG6_cjs.loadFruitQuality; }
 });
 Object.defineProperty(exports, "loadGaussianIslands", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadGaussianIslands; }
+  get: function () { return chunkG7KXZHG6_cjs.loadGaussianIslands; }
 });
 Object.defineProperty(exports, "loadHousingMini", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadHousingMini; }
+  get: function () { return chunkG7KXZHG6_cjs.loadHousingMini; }
 });
 Object.defineProperty(exports, "loadIris", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadIris; }
+  get: function () { return chunkG7KXZHG6_cjs.loadIris; }
 });
 Object.defineProperty(exports, "loadLeafShapes", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadLeafShapes; }
+  get: function () { return chunkG7KXZHG6_cjs.loadLeafShapes; }
 });
 Object.defineProperty(exports, "loadLinnerud", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadLinnerud; }
+  get: function () { return chunkG7KXZHG6_cjs.loadLinnerud; }
 });
 Object.defineProperty(exports, "loadMoonsMulti", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadMoonsMulti; }
+  get: function () { return chunkG7KXZHG6_cjs.loadMoonsMulti; }
 });
 Object.defineProperty(exports, "loadPerfectlySeparable", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadPerfectlySeparable; }
+  get: function () { return chunkG7KXZHG6_cjs.loadPerfectlySeparable; }
 });
 Object.defineProperty(exports, "loadPlantGrowth", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadPlantGrowth; }
+  get: function () { return chunkG7KXZHG6_cjs.loadPlantGrowth; }
 });
 Object.defineProperty(exports, "loadSeedMorphology", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadSeedMorphology; }
+  get: function () { return chunkG7KXZHG6_cjs.loadSeedMorphology; }
 });
 Object.defineProperty(exports, "loadSensorStates", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadSensorStates; }
+  get: function () { return chunkG7KXZHG6_cjs.loadSensorStates; }
 });
 Object.defineProperty(exports, "loadSpiralArms", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadSpiralArms; }
+  get: function () { return chunkG7KXZHG6_cjs.loadSpiralArms; }
 });
 Object.defineProperty(exports, "loadStudentPerformance", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadStudentPerformance; }
+  get: function () { return chunkG7KXZHG6_cjs.loadStudentPerformance; }
 });
 Object.defineProperty(exports, "loadTrafficConditions", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadTrafficConditions; }
+  get: function () { return chunkG7KXZHG6_cjs.loadTrafficConditions; }
 });
 Object.defineProperty(exports, "loadWeatherOutcomes", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.loadWeatherOutcomes; }
+  get: function () { return chunkG7KXZHG6_cjs.loadWeatherOutcomes; }
 });
 Object.defineProperty(exports, "makeBlobs", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.makeBlobs; }
+  get: function () { return chunkG7KXZHG6_cjs.makeBlobs; }
 });
 Object.defineProperty(exports, "makeCircles", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.makeCircles; }
+  get: function () { return chunkG7KXZHG6_cjs.makeCircles; }
 });
 Object.defineProperty(exports, "makeClassification", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.makeClassification; }
+  get: function () { return chunkG7KXZHG6_cjs.makeClassification; }
 });
 Object.defineProperty(exports, "makeGaussianQuantiles", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.makeGaussianQuantiles; }
+  get: function () { return chunkG7KXZHG6_cjs.makeGaussianQuantiles; }
 });
 Object.defineProperty(exports, "makeMoons", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.makeMoons; }
+  get: function () { return chunkG7KXZHG6_cjs.makeMoons; }
 });
 Object.defineProperty(exports, "makeRegression", {
   enumerable: true,
-  get: function () { return chunkXEG44RF6_cjs.makeRegression; }
+  get: function () { return chunkG7KXZHG6_cjs.makeRegression; }
 });
 //# sourceMappingURL=index.cjs.map
 //# sourceMappingURL=index.cjs.map

package/dist/datasets/index.d.cts

@@ -1,3 +1,3 @@
-export { a as DataLoader, D as DataLoaderOptions, g as Dataset, l as loadBreastCancer, h as loadConcentricRings, j as loadCropYield, k as loadCustomerSegments, n as loadDiabetes, o as loadDigits, p as loadEnergyEfficiency, q as loadFitnessScores, r as loadFlowersExtended, s as loadFruitQuality, t as loadGaussianIslands, u as loadHousingMini, v as loadIris, w as loadLeafShapes, x as loadLinnerud, y as loadMoonsMulti, z as loadPerfectlySeparable, A as loadPlantGrowth, B as loadSeedMorphology, C as loadSensorStates, E as loadSpiralArms, F as loadStudentPerformance, G as loadTrafficConditions, H as loadWeatherOutcomes, m as makeBlobs, b as makeCircles, c as makeClassification, d as makeGaussianQuantiles, e as makeMoons, f as makeRegression } from '../index-D61yaSMY.cjs';
-import '../Tensor-BQLk1ltW.cjs';
-import '../tensor-B96jjJLQ.cjs';
+export { a as DataLoader, D as DataLoaderOptions, g as Dataset, l as loadBreastCancer, h as loadConcentricRings, j as loadCropYield, k as loadCustomerSegments, n as loadDiabetes, o as loadDigits, p as loadEnergyEfficiency, q as loadFitnessScores, r as loadFlowersExtended, s as loadFruitQuality, t as loadGaussianIslands, u as loadHousingMini, v as loadIris, w as loadLeafShapes, x as loadLinnerud, y as loadMoonsMulti, z as loadPerfectlySeparable, A as loadPlantGrowth, B as loadSeedMorphology, C as loadSensorStates, E as loadSpiralArms, F as loadStudentPerformance, G as loadTrafficConditions, H as loadWeatherOutcomes, m as makeBlobs, b as makeCircles, c as makeClassification, d as makeGaussianQuantiles, e as makeMoons, f as makeRegression } from '../index-BL8jLf3K.cjs';
+import '../Tensor-fxBg-TFZ.cjs';
+import '../tensor-IlVTF0bz.cjs';

package/dist/datasets/index.d.ts

@@ -1,3 +1,3 @@
-export { a as DataLoader, D as DataLoaderOptions, g as Dataset, l as loadBreastCancer, h as loadConcentricRings, j as loadCropYield, k as loadCustomerSegments, n as loadDiabetes, o as loadDigits, p as loadEnergyEfficiency, q as loadFitnessScores, r as loadFlowersExtended, s as loadFruitQuality, t as loadGaussianIslands, u as loadHousingMini, v as loadIris, w as loadLeafShapes, x as loadLinnerud, y as loadMoonsMulti, z as loadPerfectlySeparable, A as loadPlantGrowth, B as loadSeedMorphology, C as loadSensorStates, E as loadSpiralArms, F as loadStudentPerformance, G as loadTrafficConditions, H as loadWeatherOutcomes, m as makeBlobs, b as makeCircles, c as makeClassification, d as makeGaussianQuantiles, e as makeMoons, f as makeRegression } from '../index-BJY2SI4i.js';
-import '../Tensor-g8mUClel.js';
-import '../tensor-B96jjJLQ.js';
+export { a as DataLoader, D as DataLoaderOptions, g as Dataset, l as loadBreastCancer, h as loadConcentricRings, j as loadCropYield, k as loadCustomerSegments, n as loadDiabetes, o as loadDigits, p as loadEnergyEfficiency, q as loadFitnessScores, r as loadFlowersExtended, s as loadFruitQuality, t as loadGaussianIslands, u as loadHousingMini, v as loadIris, w as loadLeafShapes, x as loadLinnerud, y as loadMoonsMulti, z as loadPerfectlySeparable, A as loadPlantGrowth, B as loadSeedMorphology, C as loadSensorStates, E as loadSpiralArms, F as loadStudentPerformance, G as loadTrafficConditions, H as loadWeatherOutcomes, m as makeBlobs, b as makeCircles, c as makeClassification, d as makeGaussianQuantiles, e as makeMoons, f as makeRegression } from '../index-CiTd61a5.js';
+import '../Tensor-BORFp_zt.js';
+import '../tensor-IlVTF0bz.js';

package/dist/datasets/index.js

@@ -1,5 +1,5 @@
-export { DataLoader, loadBreastCancer, loadConcentricRings, loadCropYield, loadCustomerSegments, loadDiabetes, loadDigits, loadEnergyEfficiency, loadFitnessScores, loadFlowersExtended, loadFruitQuality, loadGaussianIslands, loadHousingMini, loadIris, loadLeafShapes, loadLinnerud, loadMoonsMulti, loadPerfectlySeparable, loadPlantGrowth, loadSeedMorphology, loadSensorStates, loadSpiralArms, loadStudentPerformance, loadTrafficConditions, loadWeatherOutcomes, makeBlobs, makeCircles, makeClassification, makeGaussianQuantiles, makeMoons, makeRegression } from '../chunk-B5TNKUEY.js';
-import '../chunk-BCR7G3A6.js';
-import '../chunk-4S73VUBD.js';
+export { DataLoader, loadBreastCancer, loadConcentricRings, loadCropYield, loadCustomerSegments, loadDiabetes, loadDigits, loadEnergyEfficiency, loadFitnessScores, loadFlowersExtended, loadFruitQuality, loadGaussianIslands, loadHousingMini, loadIris, loadLeafShapes, loadLinnerud, loadMoonsMulti, loadPerfectlySeparable, loadPlantGrowth, loadSeedMorphology, loadSensorStates, loadSpiralArms, loadStudentPerformance, loadTrafficConditions, loadWeatherOutcomes, makeBlobs, makeCircles, makeClassification, makeGaussianQuantiles, makeMoons, makeRegression } from '../chunk-NDDTUFKK.js';
+import '../chunk-KCF6P34A.js';
+import '../chunk-CZOMBUI7.js';
 //# sourceMappingURL=index.js.map
 //# sourceMappingURL=index.js.map

package/dist/{index-C1mfVYoo.d.ts → index-B18dHc8q.d.ts}

@@ -1,14 +1,15 @@
-import { T as Tensor } from './Tensor-g8mUClel.js';
+import { T as Tensor } from './Tensor-BORFp_zt.js';
+import { S as Shape, c as ScalarDType } from './tensor-IlVTF0bz.js';
 
 /**
  * Base type for all estimators (models) in Deepbox.
  *
- * This follows the scikit-learn estimator type design.
+ * Base estimator type for all ML models.
  *
  * @template FitParams - Type of parameters passed to fit method
  *
  * References:
- * - scikit-learn API: https://scikit-learn.org/stable/developers/develop.html
+ * - Deepbox ML: https://deepbox.dev/docs/ml-linear
  */
 type Estimator<FitParams = void> = {
     /**
@@ -238,7 +239,7 @@ type OutlierDetector = Estimator<void> & {
  * // labels: [0, 0, 0, 1, 1, -1]  (-1 = noise)
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.cluster.DBSCAN.html | scikit-learn DBSCAN}
+ * @see {@link https://deepbox.dev/docs/ml-clustering | Deepbox Clustering}
  */
 declare class DBSCAN implements Clusterer {
     private eps;
@@ -295,6 +296,13 @@ declare class DBSCAN implements Clusterer {
      * @throws {NotFittedError} If the model has not been fitted
      */
     get labels(): Tensor;
+    /**
+     * Number of clusters found (excluding noise).
+     *
+     * @returns Number of distinct clusters (labels >= 0)
+     * @throws {NotFittedError} If the model has not been fitted
+     */
+    get nClusters(): number;
     /**
      * Get indices of core samples discovered during fitting.
      *
@@ -352,8 +360,8 @@ declare class DBSCAN implements Clusterer {
  * console.log('Centroids:', kmeans.clusterCenters);
  * ```
  *
- * @see {@link https://en.wikipedia.org/wiki/K-means_clustering | Wikipedia: K-means}
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html | scikit-learn KMeans}
+ * @see {@link https://deepbox.dev/docs/ml-clustering | Deepbox Clustering}
+ * @see {@link https://deepbox.dev/docs/ml-clustering | Deepbox Clustering}
  */
 declare class KMeans implements Clusterer {
     private nClusters;
@@ -482,8 +490,8 @@ declare class KMeans implements Clusterer {
  * console.log('Explained variance ratio:', pca.explainedVarianceRatio);
  * ```
  *
- * @see {@link https://en.wikipedia.org/wiki/Principal_component_analysis | Wikipedia: PCA}
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.PCA.html | scikit-learn PCA}
+ * @see {@link https://deepbox.dev/docs/ml-decomposition | Deepbox Dimensionality Reduction}
+ * @see {@link https://deepbox.dev/docs/ml-decomposition | Deepbox Dimensionality Reduction}
  */
 declare class PCA implements Transformer {
     private readonly nComponents?;
@@ -590,7 +598,7 @@ declare class PCA implements Transformer {
  * const predictions = gbr.predict(X);
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingRegressor.html | scikit-learn GradientBoostingRegressor}
+ * @see {@link https://deepbox.dev/docs/ml-ensemble | Deepbox Ensemble Methods}
  */
 declare class GradientBoostingRegressor implements Regressor {
     /** Number of boosting stages (trees) */
@@ -669,8 +677,10 @@ declare class GradientBoostingRegressor implements Regressor {
 /**
  * Gradient Boosting Classifier.
  *
- * Uses gradient boosting with shallow regression trees for binary classification.
- * Optimizes log loss (cross-entropy) using sigmoid function.
+ * Uses gradient boosting with shallow regression trees for classification.
+ * Supports both binary and multiclass classification.
+ * - Binary: optimizes log loss using sigmoid function.
+ * - Multiclass: uses One-vs-Rest (OvR) strategy, training one binary model per class.
  *
  * @example
  * ```ts
@@ -685,7 +695,7 @@ declare class GradientBoostingRegressor implements Regressor {
  * const predictions = gbc.predict(X);
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingClassifier.html | scikit-learn GradientBoostingClassifier}
+ * @see {@link https://deepbox.dev/docs/ml-ensemble | Deepbox Ensemble Methods}
  */
 declare class GradientBoostingClassifier implements Classifier {
     /** Number of boosting stages */
@@ -696,10 +706,10 @@ declare class GradientBoostingClassifier implements Classifier {
     private maxDepth;
     /** Minimum samples to split */
     private minSamplesSplit;
-    /** Array of weak learners */
-    private estimators;
-    /** Initial log-odds prediction */
-    private initPrediction;
+    /** Per-class arrays of weak learners (OvR for multiclass, single for binary) */
+    private estimatorsPerClass;
+    /** Per-class initial log-odds predictions */
+    private initPredictions;
     /** Number of features */
     private nFeatures;
     /** Unique class labels */
@@ -712,19 +722,29 @@ declare class GradientBoostingClassifier implements Classifier {
         readonly maxDepth?: number;
         readonly minSamplesSplit?: number;
     });
+    /**
+     * Fit a single binary boosting ensemble.
+     * Trains nEstimators regression trees to optimize log loss for a binary target.
+     */
+    private fitBinary;
+    /**
+     * Compute raw scores for a single binary ensemble.
+     */
+    private predictRawBinary;
     /**
      * Fit the gradient boosting classifier on training data.
      *
      * Builds an additive model by sequentially fitting regression trees
      * to the pseudo-residuals (gradient of log loss).
+     * Supports binary (2 classes) and multiclass (>2 classes via OvR).
      *
      * @param X - Training data of shape (n_samples, n_features)
-     * @param y - Target class labels of shape (n_samples,). Must contain exactly 2 classes.
+     * @param y - Target class labels of shape (n_samples,). Must contain at least 2 classes.
      * @returns this - The fitted estimator
      * @throws {ShapeError} If X is not 2D or y is not 1D
      * @throws {ShapeError} If X and y have different number of samples
      * @throws {DataValidationError} If X or y contain NaN/Inf values
-     * @throws {InvalidParameterError} If y does not contain exactly 2 classes
+     * @throws {InvalidParameterError} If y does not contain at least 2 classes
      */
     fit(X: Tensor, y: Tensor): this;
     /**
@@ -740,11 +760,10 @@ declare class GradientBoostingClassifier implements Classifier {
     /**
      * Predict class probabilities for samples in X.
      *
-     * Returns a matrix of shape (n_samples, 2) where columns are
-     * [P(class_0), P(class_1)].
+     * Returns a matrix of shape (n_samples, n_classes).
      *
      * @param X - Samples of shape (n_samples, n_features)
-     * @returns Class probability matrix of shape (n_samples, 2)
+     * @returns Class probability matrix of shape (n_samples, n_classes)
      * @throws {NotFittedError} If the model has not been fitted
      * @throws {ShapeError} If X has wrong dimensions or feature count
      * @throws {DataValidationError} If X contains NaN/Inf values
@@ -1041,7 +1060,7 @@ declare class LinearRegression implements Regressor {
      * @throws {ShapeError} If X has wrong dimensions or feature count
      * @throws {DataValidationError} If X contains NaN/Inf values
      */
-    predict(X: Tensor): Tensor;
+    predict(X: Tensor): Tensor<Shape, ScalarDType>;
     /**
      * Return the coefficient of determination R^2 of the prediction.
      *
@@ -1434,7 +1453,7 @@ declare class Ridge implements Regressor {
  * const embedding = tsne.fitTransform(X);
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html | scikit-learn TSNE}
+ * @see {@link https://deepbox.dev/docs/ml-manifold | Deepbox Manifold Learning}
  * @see van der Maaten, L.J.P.; Hinton, G.E. (2008). "Visualizing High-Dimensional Data Using t-SNE"
  */
 declare class TSNE {
@@ -1543,6 +1562,14 @@ declare class TSNE {
      * Fit the model (same as fitTransform for t-SNE).
      */
     fit(X: Tensor): this;
+    /**
+     * Return the fitted embedding. For t-SNE, transform is equivalent to
+     * returning the already-computed embedding (t-SNE is non-parametric).
+     *
+     * @param _X - Ignored, present for API compatibility
+     * @returns Low-dimensional embedding of shape (n_samples, n_components)
+     */
+    transform(_X?: Tensor): Tensor;
     /**
      * Get the embedding.
      */
@@ -1585,8 +1612,8 @@ declare class TSNE {
  * const predictions = nb.predict(tensor([[2.5, 3.5]]));
  * ```
  *
- * @see {@link https://en.wikipedia.org/wiki/Naive_Bayes_classifier | Wikipedia: Naive Bayes}
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.naive_bayes.GaussianNB.html | scikit-learn GaussianNB}
+ * @see {@link https://deepbox.dev/docs/ml-naive-bayes | Deepbox Naive Bayes}
+ * @see {@link https://deepbox.dev/docs/ml-naive-bayes | Deepbox Naive Bayes}
  */
 declare class GaussianNB implements Classifier {
     private readonly varSmoothing;
@@ -1737,8 +1764,8 @@ declare abstract class KNeighborsBase {
  * const predictions = knn.predict(tensor([[1.5, 1.5]]));
  * ```
  *
- * @see {@link https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm | Wikipedia: KNN}
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsClassifier.html | scikit-learn KNeighborsClassifier}
+ * @see {@link https://deepbox.dev/docs/ml-neighbors | Deepbox Nearest Neighbors}
+ * @see {@link https://deepbox.dev/docs/ml-neighbors | Deepbox Nearest Neighbors}
  */
 declare class KNeighborsClassifier extends KNeighborsBase implements Classifier {
     /**
@@ -1805,7 +1832,7 @@ declare class KNeighborsClassifier extends KNeighborsBase implements Classifier
  * const predictions = knn.predict(tensor([[1.5]]));
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsRegressor.html | scikit-learn KNeighborsRegressor}
+ * @see {@link https://deepbox.dev/docs/ml-neighbors | Deepbox Nearest Neighbors}
  */
 declare class KNeighborsRegressor extends KNeighborsBase implements Regressor {
     /**
@@ -1868,7 +1895,7 @@ declare class KNeighborsRegressor extends KNeighborsBase implements Regressor {
  * const predictions = svm.predict(X);
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.svm.LinearSVC.html | scikit-learn LinearSVC}
+ * @see {@link https://deepbox.dev/docs/ml-svm | Deepbox SVM}
  */
 declare class LinearSVC implements Classifier {
     /** Regularization parameter (inverse of regularization strength) */
@@ -1877,13 +1904,13 @@ declare class LinearSVC implements Classifier {
     private readonly maxIter;
     /** Tolerance for stopping criterion */
     private readonly tol;
-    /** Weight vector of shape (n_features,) */
-    private weights;
-    /** Bias term */
-    private bias;
+    /** Per-class weight vectors (OvR for multiclass, single for binary) */
+    private weightsPerClass;
+    /** Per-class bias terms */
+    private biasPerClass;
     /** Number of features seen during fit */
     private nFeatures;
-    /** Unique class labels [0, 1] mapped from original labels */
+    /** Unique class labels */
     private classLabels;
     /** Whether the model has been fitted */
     private fitted;
@@ -1900,19 +1927,27 @@ declare class LinearSVC implements Classifier {
         readonly maxIter?: number;
         readonly tol?: number;
     });
+    /**
+     * Fit a single binary SVM using sub-gradient descent on hinge loss.
+     * Maps labels to {-1, +1} and returns learned weights + bias.
+     */
+    private fitBinary;
+    /**
+     * Compute decision value for a single binary classifier.
+     */
+    private decisionBinary;
     /**
      * Fit the SVM classifier using sub-gradient descent.
      *
-     * Uses a simplified hinge loss optimization with L2 regularization.
-     * Objective: minimize (1/2)||w||² + C * Σmax(0, 1 - y_i(w · x_i + b))
+     * Supports both binary and multiclass classification (via OvR).
      *
      * @param X - Training data of shape (n_samples, n_features)
-     * @param y - Target labels of shape (n_samples,). Must contain exactly 2 classes.
+     * @param y - Target labels of shape (n_samples,). Must contain at least 2 classes.
      * @returns this - The fitted estimator
      * @throws {ShapeError} If X is not 2D or y is not 1D
      * @throws {ShapeError} If X and y have different number of samples
      * @throws {DataValidationError} If X or y contain NaN/Inf values
-     * @throws {InvalidParameterError} If y does not contain exactly 2 classes
+     * @throws {InvalidParameterError} If y does not contain at least 2 classes
      */
     fit(X: Tensor, y: Tensor): this;
     /**
@@ -1954,12 +1989,12 @@ declare class LinearSVC implements Classifier {
      */
     get coef(): Tensor;
     /**
-     * Get the bias term.
+     * Get the bias terms.
      *
-     * @returns Bias value
+     * @returns Bias values as tensor
      * @throws {NotFittedError} If the model has not been fitted
      */
-    get intercept(): number;
+    get intercept(): Tensor;
     /**
      * Get hyperparameters for this estimator.
      *
@@ -1992,7 +2027,7 @@ declare class LinearSVC implements Classifier {
  * const predictions = svr.predict(X);
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.svm.LinearSVR.html | scikit-learn LinearSVR}
+ * @see {@link https://deepbox.dev/docs/ml-svm | Deepbox SVM}
  */
 declare class LinearSVR implements Regressor {
     /** Regularization parameter */
@@ -2088,7 +2123,7 @@ declare class LinearSVR implements Regressor {
  * const predictions = clf.predict(X);
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html | scikit-learn DecisionTreeClassifier}
+ * @see {@link https://deepbox.dev/docs/ml-tree | Deepbox Decision Trees}
  */
 declare class DecisionTreeClassifier implements Classifier {
     private maxDepth;
@@ -2182,7 +2217,7 @@ declare class DecisionTreeClassifier implements Classifier {
  *
  * Uses MSE reduction to find optimal splits for regression tasks.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeRegressor.html | scikit-learn DecisionTreeRegressor}
+ * @see {@link https://deepbox.dev/docs/ml-tree | Deepbox Decision Trees}
  */
 declare class DecisionTreeRegressor implements Regressor {
     private maxDepth;
@@ -2275,7 +2310,7 @@ declare class DecisionTreeRegressor implements Regressor {
  * const predictions = clf.predict(X_test);
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html | scikit-learn RandomForestClassifier}
+ * @see {@link https://deepbox.dev/docs/ml-tree | Deepbox Decision Trees}
  */
 declare class RandomForestClassifier implements Classifier {
     private readonly nEstimators;
@@ -2389,7 +2424,7 @@ declare class RandomForestClassifier implements Classifier {
  * const predictions = reg.predict(X_test);
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html | scikit-learn RandomForestRegressor}
+ * @see {@link https://deepbox.dev/docs/ml-tree | Deepbox Decision Trees}
  */
 declare class RandomForestRegressor implements Regressor {
     private readonly nEstimators;

package/dist/{index-GFAVyOWO.d.ts → index-BHHX0qTY.d.cts}

@@ -1,5 +1,5 @@
-import { A as Axis } from './tensor-B96jjJLQ.js';
-import { T as Tensor } from './Tensor-g8mUClel.js';
+import { A as Axis } from './tensor-IlVTF0bz.cjs';
+import { T as Tensor } from './Tensor-fxBg-TFZ.cjs';
 
 /**
  * Data structure for DataFrame initialization.
@@ -15,6 +15,7 @@ import { T as Tensor } from './Tensor-g8mUClel.js';
  * };
  * ```
  */
+type DataValue = number | string | boolean | null | undefined;
 type DataFrameData = Record<string, unknown[]>;
 /**
  * Configuration options for DataFrame construction.
@@ -64,7 +65,7 @@ type AggregateFunction = "sum" | "mean" | "median" | "min" | "max" | "std" | "va
  * - An array of index labels (can be strings or numbers)
  * - An optional name
  *
- * Similar to pandas Series in Python.
+ * A one-dimensional labeled array. @see { https://deepbox.dev/docs/dataframe-series | Deepbox Series}
  *
  * @template T - The type of data stored in the Series
  *
@@ -80,7 +81,7 @@ type AggregateFunction = "sum" | "mean" | "median" | "min" | "max" | "std" | "va
  * });
  * ```
  *
- * @see {@link https://pandas.pydata.org/docs/reference/api/pandas.Series.html | Pandas Series}
+ * @see {@link https://deepbox.dev/docs/dataframe-series | Deepbox Series}
  */
 declare class Series<T = unknown> {
     private _data;
@@ -435,7 +436,7 @@ declare class Series<T = unknown> {
  * - Columns (variables) identified by column names
  * - Data stored in a columnar format for efficient access
  *
- * Similar to pandas DataFrame in Python.
+ * A tabular data structure with labeled columns. @see { https://deepbox.dev/docs/dataframe-overview | Deepbox DataFrame}
  *
  * @example
  * ```ts
@@ -451,7 +452,7 @@ declare class Series<T = unknown> {
  * console.log(df.columns);  // ['name', 'age', 'score']
  * ```
  *
- * @see {@link https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html | Pandas DataFrame}
+ * @see {@link https://deepbox.dev/docs/dataframe-overview | Deepbox DataFrame}
  */
 declare class DataFrame {
     private _data;
@@ -588,7 +589,7 @@ declare class DataFrame {
      * // DataFrame with Bob and Carol
      * ```
      */
-    filter(predicate: (row: Record<string, unknown>) => boolean): DataFrame;
+    filter(predicate: (row: Record<string, any>) => boolean): DataFrame;
     /**
      * Select a subset of columns.
      *
@@ -687,11 +688,11 @@ declare class DataFrame {
      * // Result: Alice, Bob, Charlie (Charlie has null for product)
      * ```
      *
-     * @see {@link https://en.wikipedia.org/wiki/Hash_join | Hash Join Algorithm}
+     * @see {@link https://deepbox.dev/docs/dataframe-overview | Deepbox DataFrame}
      */
     join(other: DataFrame, on: string, how?: "inner" | "left" | "right" | "outer"): DataFrame;
     /**
-     * Merge with another DataFrame using pandas-style merge.
+     * Merge with another DataFrame using SQL-style merge.
      *
      * More flexible than join() - supports different column names for join keys.
      * Uses hash join algorithm for O(n + m) complexity.
@@ -726,7 +727,7 @@ declare class DataFrame {
      * });
      * ```
      *
-     * @see {@link https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.merge.html | Pandas merge}
+     * @see {@link https://deepbox.dev/docs/dataframe-overview | Deepbox DataFrame}
      */
     merge(other: DataFrame, options?: {
         readonly on?: string;
@@ -1417,11 +1418,12 @@ type index_DataFrameData = DataFrameData;
 type index_DataFrameGroupBy = DataFrameGroupBy;
 declare const index_DataFrameGroupBy: typeof DataFrameGroupBy;
 type index_DataFrameOptions = DataFrameOptions;
+type index_DataValue = DataValue;
 type index_Series<T = unknown> = Series<T>;
 declare const index_Series: typeof Series;
 type index_SeriesOptions = SeriesOptions;
 declare namespace index {
-  export { type index_AggregateFunction as AggregateFunction, index_DataFrame as DataFrame, type index_DataFrameData as DataFrameData, index_DataFrameGroupBy as DataFrameGroupBy, type index_DataFrameOptions as DataFrameOptions, index_Series as Series, type index_SeriesOptions as SeriesOptions };
+  export { type index_AggregateFunction as AggregateFunction, index_DataFrame as DataFrame, type index_DataFrameData as DataFrameData, index_DataFrameGroupBy as DataFrameGroupBy, type index_DataFrameOptions as DataFrameOptions, type index_DataValue as DataValue, index_Series as Series, type index_SeriesOptions as SeriesOptions };
 }
 
-export { type AggregateFunction as A, DataFrame as D, Series as S, DataFrameGroupBy as a, type DataFrameData as b, type DataFrameOptions as c, type SeriesOptions as d, index as i };
+export { type AggregateFunction as A, DataFrame as D, Series as S, DataFrameGroupBy as a, type DataFrameData as b, type DataFrameOptions as c, type DataValue as d, type SeriesOptions as e, index as i };