Skip to main content

​
DecisionTreeClassifier

A non-parametric supervised learning method that learns simple decision rules inferred from the data features. Uses CART (Classification and Regression Trees) algorithm with Gini impurity for classification.

​
Constructor

new DecisionTreeClassifier(options?: {
  maxDepth?: number;
  minSamplesSplit?: number;
  minSamplesLeaf?: number;
  maxFeatures?: number;
  randomState?: number;
})
​
options.maxDepth
number
default:"10"
Maximum depth of the tree. Controls model complexity.
​
options.minSamplesSplit
number
default:"2"
Minimum number of samples required to split an internal node.
​
options.minSamplesLeaf
number
default:"1"
Minimum number of samples required to be at a leaf node.
​
options.maxFeatures
number
Number of features to consider when looking for the best split. If undefined, uses all features.
​
options.randomState
number
Random seed for reproducibility.

​
Methods

​
fit

fit(X: Tensor, y: Tensor): this
Build a decision tree classifier from the training set.

​
predict

predict(X: Tensor): Tensor
Predict class labels for samples in X.

​
predictProba

predictProba(X: Tensor): Tensor
Predict class probabilities for samples in X. Returns: Class probability matrix of shape (n_samples, n_classes)

​
score

score(X: Tensor, y: Tensor): number
Return the mean accuracy on the given test data and labels.

​
Properties

​
classes
Tensor | undefined
Unique class labels discovered during fitting.

​
Example

import { DecisionTreeClassifier } from 'deepbox/ml';
import { tensor } from 'deepbox/ndarray';

const X = tensor([[1, 2], [3, 4], [5, 6], [7, 8]]);
const y = tensor([0, 0, 1, 1]);

const clf = new DecisionTreeClassifier({ maxDepth: 3 });
clf.fit(X, y);
const predictions = clf.predict(X);

​
DecisionTreeRegressor

Decision Tree Regressor using MSE reduction to find optimal splits.

​
Constructor

new DecisionTreeRegressor(options?: {
  maxDepth?: number;
  minSamplesSplit?: number;
  minSamplesLeaf?: number;
  maxFeatures?: number;
  randomState?: number;
})
​
options.maxDepth
number
default:"10"
Maximum depth of the tree.
​
options.minSamplesSplit
number
default:"2"
Minimum samples required to split an internal node.
​
options.minSamplesLeaf
number
default:"1"
Minimum samples required at a leaf node.
​
options.maxFeatures
number
Number of features to consider for best split.
​
options.randomState
number
Random seed for reproducibility.

​
Methods

​
fit

fit(X: Tensor, y: Tensor): this
Build a decision tree regressor from the training set.

​
predict

predict(X: Tensor): Tensor
Predict target values for samples in X.

​
score

score(X: Tensor, y: Tensor): number
Return the R² score on the given test data. Returns: R² = 1 - SS_res / SS_tot

​
Example

import { DecisionTreeRegressor } from 'deepbox/ml';
import { tensor } from 'deepbox/ndarray';

const X = tensor([[1], [2], [3], [4]]);
const y = tensor([1.5, 2.5, 3.5, 4.5]);

const reg = new DecisionTreeRegressor({ maxDepth: 5 });
reg.fit(X, y);
const predictions = reg.predict(X);

​
RandomForestClassifier

An ensemble of decision trees trained on random subsets of data and features. Predictions are made by majority voting.

​
Constructor

new RandomForestClassifier(options?: {
  nEstimators?: number;
  maxDepth?: number;
  minSamplesSplit?: number;
  minSamplesLeaf?: number;
  maxFeatures?: "sqrt" | "log2" | number;
  bootstrap?: boolean;
  randomState?: number;
})
​
options.nEstimators
number
default:"100"
Number of trees in the forest.
​
options.maxDepth
number
default:"10"
Maximum depth of each tree.
​
options.minSamplesSplit
number
default:"2"
Minimum samples to split a node.
​
options.minSamplesLeaf
number
default:"1"
Minimum samples at a leaf.
​
options.maxFeatures
string | number
default:"sqrt"
Number of features to consider: ‘sqrt’, ‘log2’, or integer.
​
options.bootstrap
boolean
default:"true"
Whether to use bootstrap samples.
​
options.randomState
number
Random seed for reproducibility.

​
Methods

​
fit

fit(X: Tensor, y: Tensor): this
Fit the random forest classifier. Builds an ensemble of decision trees.

​
predict

predict(X: Tensor): Tensor
Predict class labels via majority voting.

​
predictProba

predictProba(X: Tensor): Tensor
Predict class probabilities by averaging predictions from all trees.

​
score

score(X: Tensor, y: Tensor): number
Return mean accuracy.

​
Properties

​
classes
Tensor | undefined
Unique class labels.

​
Example

import { RandomForestClassifier } from 'deepbox/ml';
import { tensor } from 'deepbox/ndarray';

const clf = new RandomForestClassifier({ nEstimators: 100 });
clf.fit(X_train, y_train);
const predictions = clf.predict(X_test);

​
RandomForestRegressor

An ensemble of decision tree regressors. Predictions are averaged across all trees.

​
Constructor

new RandomForestRegressor(options?: {
  nEstimators?: number;
  maxDepth?: number;
  minSamplesSplit?: number;
  minSamplesLeaf?: number;
  maxFeatures?: "sqrt" | "log2" | number;
  bootstrap?: boolean;
  randomState?: number;
})
​
options.nEstimators
number
default:"100"
Number of trees in the forest.
​
options.maxDepth
number
default:"10"
Maximum depth of each tree.
​
options.minSamplesSplit
number
default:"2"
Minimum samples to split a node.
​
options.minSamplesLeaf
number
default:"1"
Minimum samples at a leaf.
​
options.maxFeatures
string | number
default:"1.0"
Number of features to consider: ‘sqrt’, ‘log2’, 1.0 (all), or integer.
​
options.bootstrap
boolean
default:"true"
Whether to use bootstrap samples.
​
options.randomState
number
Random seed.

​
Methods

​
fit

fit(X: Tensor, y: Tensor): this
Fit the random forest regressor.

​
predict

predict(X: Tensor): Tensor
Predict target values by averaging predictions from all trees.

​
score

score(X: Tensor, y: Tensor): number
Return R² score.

​
Example

import { RandomForestRegressor } from 'deepbox/ml';
import { tensor } from 'deepbox/ndarray';

const reg = new RandomForestRegressor({ nEstimators: 100 });
reg.fit(X_train, y_train);
const predictions = reg.predict(X_test);

Build docs developers (and LLMs) love

Get started for freeTalk to us