Skip to content
This repository has been archived by the owner on Oct 8, 2019. It is now read-only.

List of Supported Algorithms

Makoto YUI edited this page Jun 15, 2015 · 1 revision

Here is a list of supported algorithms since Hivemall v0.2.

Type Algorithm Supported version
Classfication Perceptron v0.2
Classfication Passive Aggressive (PA, PA1, PA2) v0.2
Classfication Confidence Weighted (CW) v0.2
Classfication Adaptive Regularization of Weight Vectors (AROW) v0.2
Classfication Soft Confidence Weighted (SCW1, SCW2) v0.2
Classfication AdaGradRDA (with hinge loss) v0.3 (beta 2)
Regression Logistic Regression using Stochastic Gradient Descent v0.2
Regression AdaGrad / AdaDelta (with Logistic Loss) v0.3 (beta 2)
Regression Passive Aggressive Regression (PA1, PA2) v0.2
Regression AROW regression v0.2
Regression Polynomial Regression v0.3.2
Recommendation Matrix Factorization (sgd, adagrad) v0.3
Recommendation Minhash (LSH with jaccard index) v0.2
Nearest Neighbor Minhash (LSH with jaccard index) v0.2
Nearest Neighbor b-Bit minhash v0.2
Nearest Neighbor Brute-force search using Cosine similarity v0.2
Anomaly Detection Local Outlier Factor (LOF) v0.3.2
Feature engineering Feature Hashing (MurmurHash, SHA1) v0.2
Feature engineering Feature scaling (Min-Max Normalization, Z-Score) v0.2
Feature engineering Polynomial Features v0.3.2
Feature engineering Feature instances amplifier that reduces iterations on training v0.2
Feature engineering TF-IDF vectorizer v0.3 (beta 3)
Feature engineering Bias clause v0.2
Feature engineering Data generator for one-vs-the-rest classifiers v0.2
Clone this wiki locally