overview

High-quality documentation is a development goal of mlpack. mlpack’s documentation is split into two parts: documentation for the bindings, and documentation for the C++ library. Generally, working with the bindings is a good choice for simple machine learning and data science tasks, and writing C++ is a good idea when complex or custom functionality is desired.

All interfaces are heavily documented, and if you find a documentation issue, please report it.

quickstart

Just getting started with mlpack? Try these quickstart tutorials for the bindings to other languages.

Quickstart Tutorials

These tutorials give very quick “getting started” examples that you can use to get started with mlpack in different languages.

Once you’re comfortable with the quickstart guides for the language of your choice, full documentation for every binding can be found below. Quick links are in the left sidebar.

The C++ interfaces of mlpack are carefully documented and doxygen is used to provide automatically-generated searchable documentation.

tutorials

A number of tutorials are available covering individual algorithms and functionality inside of mlpack, both for bindings to other languages and for the C++ interface.

Introductory Tutorials

These tutorials introduce the basic concepts of working with mlpack, aimed at developers who want to use and contribute to mlpack but are not sure where to start.

Method-specific Tutorials

These tutorials introduce the various methods mlpack offers, aimed at users who want to get started quickly. These tutorials start with simple examples and progress to complex, extensible uses.

Advanced Tutorials

These tutorials discuss some of the more advanced functionality contained in mlpack.

Policy Class Documentation

mlpack uses templates to achieve its genericity and flexibility. Some of the template types used by mlpack are common across multiple machine learning algorithms. The links below provide documentation for some of these common types.

mlpack git binding documentation

data formats

mlpack bindings for Go take and return a restricted set of types, for simplicity. These include primitive types, matrix/vector types, categorical matrix types, and model types. Each type is detailed below.

  • int: An integer (i.e., 1).
  • float64: A floating-point number (i.e., 0.5).
  • bool: A boolean flag option (true or false).
  • string: A character string (i.e., "hello").
  • array of ints: An array of integers; i.e., []int{0, 1, 2}.
  • array of strings: An array of strings; i.e., []string{"hello", "goodbye"}.
  • *mat.Dense: A 2-d gonum Matrix. If the type is not already float64, it will be converted.
  • *mat.Dense (with ints): A 2-d gonum Matrix. If the type is not already int64, it will be converted.
  • *mat.Dense (1d): A 1-d gonum Matrix (that is, a Matrix where either the number of rows or number of columns is 1).
  • *mat.Dense (1d with ints): A 1-d gonum Matrix (that is, a Matrix where either the number of rows or number of columns is 1).
  • matrixWithInfo: A Tuple(matrixWithInfo) containing float64 data (Data) along with a boolean array (Categoricals) indicating which dimensions are categorical (represented by true) and which are numeric (represented by false). The number of elements in the boolean array should be the same as the dimensionality of the data matrix. It is expected that each row of the matrix corresponds to a single data point when calling mlpack bindings.
  • mlpackModel: An mlpack model pointer. This type holds a pointer to C++ memory containing the mlpack model. Note that this means the mlpack model itself cannot be easily inspected in Go. However, the pointer can be passed to subsequent calls to mlpack functions.

Adaboost()

AdaBoost

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Adaboost().
param := mlpack.AdaboostOptions()
param.InputModel = nil
param.Iterations = 1000
param.Labels = mat.NewDense(1, 1, nil)
param.Test = mat.NewDense(1, 1, nil)
param.Tolerance = 1e-10
param.Training = mat.NewDense(1, 1, nil)
param.WeakLearner = "decision_stump"

output, output_model, predictions, probabilities := mlpack.Adaboost(param)

An implementation of the AdaBoost.MH (Adaptive Boosting) algorithm for classification. This can be used to train an AdaBoost model on labeled data or use an existing AdaBoost model to predict the classes of new points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
InputModel adaBoostModel Input AdaBoost model. nil
Iterations int The maximum number of boosting iterations to be run (0 will run until convergence.) 1000
Labels *mat.Dense (1d with ints) Labels for the training set. mat.NewDense(1, 1, nil)
Test *mat.Dense Test dataset. mat.NewDense(1, 1, nil)
Tolerance float64 The tolerance for change in values of the weighted error during training. 1e-10
Training *mat.Dense Dataset for training AdaBoost. mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false
WeakLearner string The type of weak learner to use: ‘decision_stump’, or ‘perceptron’. "decision_stump"

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense (1d with ints) Predicted labels for the test set.
outputModel adaBoostModel Output trained AdaBoost model.
predictions *mat.Dense (1d with ints) Predicted labels for the test set.
probabilities *mat.Dense Predicted class probabilities for each point in the test set.

Detailed documentation

This program implements the AdaBoost (or Adaptive Boosting) algorithm. The variant of AdaBoost implemented here is AdaBoost.MH. It uses a weak learner, either decision stumps or perceptrons, and over many iterations, creates a strong learner that is a weighted ensemble of weak learners. It runs these iterations until a tolerance value is crossed for change in the value of the weighted training error.

For more information about the algorithm, see the paper “Improved Boosting Algorithms Using Confidence-Rated Predictions”, by R.E. Schapire and Y. Singer.

This program allows training of an AdaBoost model, and then application of that model to a test dataset. To train a model, a dataset must be passed with the Training option. Labels can be given with the Labels option; if no labels are specified, the labels will be assumed to be the last column of the input dataset. Alternately, an AdaBoost model may be loaded with the InputModel option.

Once a model is trained or loaded, it may be used to provide class predictions for a given test dataset. A test dataset may be specified with the Test parameter. The predicted classes for each point in the test dataset are output to the Predictions output parameter. The AdaBoost model itself is output to the OutputModel output parameter.

Note: the following parameter is deprecated and will be removed in mlpack 4.0.0: Output. Use Predictions instead of Output.

Example

For example, to run AdaBoost on an input dataset data with labels labelsand perceptrons as the weak learner type, storing the trained model in model, one could use the following command:

// Initialize optional parameters for Adaboost().
param := mlpack.AdaboostOptions()
param.Training = data
param.Labels = labels
param.WeakLearner = "perceptron"

_, model, _, _ := mlpack.Adaboost(param)

Similarly, an already-trained model in model can be used to provide class predictions from test data test_data and store the output in predictions with the following command:

// Initialize optional parameters for Adaboost().
param := mlpack.AdaboostOptions()
param.InputModel = &model
param.Test = test_data

_, _, predictions, _ := mlpack.Adaboost(param)

See also

ApproxKfn()

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for ApproxKfn().
param := mlpack.ApproxKfnOptions()
param.Algorithm = "ds"
param.CalculateError = false
param.ExactDistances = mat.NewDense(1, 1, nil)
param.InputModel = nil
param.K = 0
param.NumProjections = 5
param.NumTables = 5
param.Query = mat.NewDense(1, 1, nil)
param.Reference = mat.NewDense(1, 1, nil)

distances, neighbors, output_model := mlpack.ApproxKfn(param)

An implementation of two strategies for furthest neighbor search. This can be used to compute the furthest neighbor of query point(s) from a set of points; furthest neighbor models can be saved and reused with future query point(s). Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Algorithm string Algorithm to use: ‘ds’ or ‘qdafn’. "ds"
CalculateError bool If set, calculate the average distance error for the first furthest neighbor only. false
ExactDistances *mat.Dense Matrix containing exact distances to furthest neighbors; this can be used to avoid explicit calculation when –calculate_error is set. mat.NewDense(1, 1, nil)
InputModel approxkfnModel File containing input model. nil
K int Number of furthest neighbors to search for. 0
NumProjections int Number of projections to use in each hash table. 5
NumTables int Number of hash tables to use. 5
Query *mat.Dense Matrix containing query points. mat.NewDense(1, 1, nil)
Reference *mat.Dense Matrix containing the reference dataset. mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
distances *mat.Dense Matrix to save furthest neighbor distances to.
neighbors *mat.Dense (with ints) Matrix to save neighbor indices to.
outputModel approxkfnModel File to save output model to.

Detailed documentation

This program implements two strategies for furthest neighbor search. These strategies are:

  • The ‘qdafn’ algorithm from “Approximate Furthest Neighbor in High Dimensions” by R. Pagh, F. Silvestri, J. Sivertsen, and M. Skala, in Similarity Search and Applications 2015 (SISAP).
  • The ‘DrusillaSelect’ algorithm from “Fast approximate furthest neighbors with data-dependent candidate selection”, by R.R. Curtin and A.B. Gardner, in Similarity Search and Applications 2016 (SISAP).

These two strategies give approximate results for the furthest neighbor search problem and can be used as fast replacements for other furthest neighbor techniques such as those found in the mlpack_kfn program. Note that typically, the ‘ds’ algorithm requires far fewer tables and projections than the ‘qdafn’ algorithm.

Specify a reference set (set to search in) with Reference, specify a query set with Query, and specify algorithm parameters with NumTables and NumProjections (or don’t and defaults will be used). The algorithm to be used (either ‘ds’—the default—or ‘qdafn’) may be specified with Algorithm. Also specify the number of neighbors to search for with K.

Note that for ‘qdafn’ in lower dimensions, NumProjections may need to be set to a high value in order to return results for each query point.

If no query set is specified, the reference set will be used as the query set. The OutputModel output parameter may be used to store the built model, and an input model may be loaded instead of specifying a reference set with the InputModel option.

Results for each query point can be stored with the Neighbors and Distances output parameters. Each row of these output matrices holds the k distances or neighbor indices for each query point.

Example

For example, to find the 5 approximate furthest neighbors with reference_set as the reference set and query_set as the query set using DrusillaSelect, storing the furthest neighbor indices to neighbors and the furthest neighbor distances to distances, one could call

// Initialize optional parameters for ApproxKfn().
param := mlpack.ApproxKfnOptions()
param.Query = query_set
param.Reference = reference_set
param.K = 5
param.Algorithm = "ds"

distances, neighbors, _ := mlpack.ApproxKfn(param)

and to perform approximate all-furthest-neighbors search with k=1 on the set data storing only the furthest neighbor distances to distances, one could call

// Initialize optional parameters for ApproxKfn().
param := mlpack.ApproxKfnOptions()
param.Reference = reference_set
param.K = 1

distances, _, _ := mlpack.ApproxKfn(param)

A trained model can be re-used. If a model has been previously saved to model, then we may find 3 approximate furthest neighbors on a query set new_query_set using that model and store the furthest neighbor indices into neighbors by calling

// Initialize optional parameters for ApproxKfn().
param := mlpack.ApproxKfnOptions()
param.InputModel = &model
param.Query = new_query_set
param.K = 3

_, neighbors, _ := mlpack.ApproxKfn(param)

See also

BayesianLinearRegression()

BayesianLinearRegression

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for BayesianLinearRegression().
param := mlpack.BayesianLinearRegressionOptions()
param.Center = false
param.Input = mat.NewDense(1, 1, nil)
param.InputModel = nil
param.Responses = mat.NewDense(1, 1, nil)
param.Scale = false
param.Test = mat.NewDense(1, 1, nil)

output_model, predictions, stds := mlpack.BayesianLinearRegression(param)

An implementation of the bayesian linear regression. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Center bool Center the data and fit the intercept if enabled. false
Input *mat.Dense Matrix of covariates (X). mat.NewDense(1, 1, nil)
InputModel bayesianLinearRegression Trained BayesianLinearRegression model to use. nil
Responses *mat.Dense (1d) Matrix of responses/observations (y). mat.NewDense(1, 1, nil)
Scale bool Scale each feature by their standard deviations if enabled. false
Test *mat.Dense Matrix containing points to regress on (test points). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel bayesianLinearRegression Output BayesianLinearRegression model.
predictions *mat.Dense If –test_file is specified, this file is where the predicted responses will be saved.
stds *mat.Dense If specified, this is where the standard deviations of the predictive distribution will be saved.

Detailed documentation

An implementation of the bayesian linear regression. This model is a probabilistic view and implementation of the linear regression. The final solution is obtained by computing a posterior distribution from gaussian likelihood and a zero mean gaussian isotropic prior distribution on the solution. Optimization is AUTOMATIC and does not require cross validation. The optimization is performed by maximization of the evidence function. Parameters are tuned during the maximization of the marginal likelihood. This procedure includes the Ockham’s razor that penalizes over complex solutions.

This program is able to train a Bayesian linear regression model or load a model from file, output regression predictions for a test set, and save the trained model to a file.

To train a BayesianLinearRegression model, the Input and Responsesparameters must be given. The Centerand Scale parameters control the centering and the normalizing options. A trained model can be saved with the OutputModel. If no training is desired at all, a model can be passed via the InputModel parameter.

The program can also provide predictions for test data using either the trained model or the given input model. Test points can be specified with the Test parameter. Predicted responses to the test points can be saved with the Predictions output parameter. The corresponding standard deviation can be save by precising the Stds parameter.

Example

For example, the following command trains a model on the data data and responses responseswith center set to true and scale set to false (so, Bayesian linear regression is being solved, and then the model is saved to blr_model:

// Initialize optional parameters for BayesianLinearRegression().
param := mlpack.BayesianLinearRegressionOptions()
param.Input = data
param.Responses = responses
param.Center = 1
param.Scale = 0

blr_model, _, _ := mlpack.BayesianLinearRegression(param)

The following command uses the blr_model to provide predicted responses for the data test and save those responses to test_predictions:

// Initialize optional parameters for BayesianLinearRegression().
param := mlpack.BayesianLinearRegressionOptions()
param.InputModel = &blr_model
param.Test = test

_, test_predictions, _ := mlpack.BayesianLinearRegression(param)

Because the estimator computes a predictive distribution instead of a simple point estimate, the Stds parameter allows one to save the prediction uncertainties:

// Initialize optional parameters for BayesianLinearRegression().
param := mlpack.BayesianLinearRegressionOptions()
param.InputModel = &blr_model
param.Test = test

_, test_predictions, stds := mlpack.BayesianLinearRegression(param)

See also

Cf()

Collaborative Filtering

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Cf().
param := mlpack.CfOptions()
param.Algorithm = "NMF"
param.AllUserRecommendations = false
param.InputModel = nil
param.Interpolation = "average"
param.IterationOnlyTermination = false
param.MaxIterations = 1000
param.MinResidue = 1e-05
param.NeighborSearch = "euclidean"
param.Neighborhood = 5
param.Normalization = "none"
param.Query = mat.NewDense(1, 1, nil)
param.Rank = 0
param.Recommendations = 5
param.Seed = 0
param.Test = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)

output, output_model := mlpack.Cf(param)

An implementation of several collaborative filtering (CF) techniques for recommender systems. This can be used to train a new CF model, or use an existing CF model to compute recommendations. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Algorithm string Algorithm used for matrix factorization. "NMF"
AllUserRecommendations bool Generate recommendations for all users. false
InputModel cfModel Trained CF model to load. nil
Interpolation string Algorithm used for weight interpolation. "average"
IterationOnlyTermination bool Terminate only when the maximum number of iterations is reached. false
MaxIterations int Maximum number of iterations. If set to zero, there is no limit on the number of iterations. 1000
MinResidue float64 Residue required to terminate the factorization (lower values generally mean better fits). 1e-05
NeighborSearch string Algorithm used for neighbor search. "euclidean"
Neighborhood int Size of the neighborhood of similar users to consider for each query user. 5
Normalization string Normalization performed on the ratings. "none"
Query *mat.Dense (with ints) List of query users for which recommendations should be generated. mat.NewDense(1, 1, nil)
Rank int Rank of decomposed matrices (if 0, a heuristic is used to estimate the rank). 0
Recommendations int Number of recommendations to generate for each query user. 5
Seed int Set the random seed (0 uses std::time(NULL)). 0
Test *mat.Dense Test set to calculate RMSE on. mat.NewDense(1, 1, nil)
Training *mat.Dense Input dataset to perform CF on. mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense (with ints) Matrix that will store output recommendations.
outputModel cfModel Output for trained CF model.

Detailed documentation

This program performs collaborative filtering (CF) on the given dataset. Given a list of user, item and preferences (the Training parameter), the program will perform a matrix decomposition and then can perform a series of actions related to collaborative filtering. Alternately, the program can load an existing saved CF model with the InputModel parameter and then use that model to provide recommendations or predict values.

The input matrix should be a 3-dimensional matrix of ratings, where the first dimension is the user, the second dimension is the item, and the third dimension is that user’s rating of that item. Both the users and items should be numeric indices, not names. The indices are assumed to start from 0.

A set of query users for which recommendations can be generated may be specified with the Query parameter; alternately, recommendations may be generated for every user in the dataset by specifying the AllUserRecommendations parameter. In addition, the number of recommendations per user to generate can be specified with the Recommendations parameter, and the number of similar users (the size of the neighborhood) to be considered when generating recommendations can be specified with the Neighborhood parameter.

For performing the matrix decomposition, the following optimization algorithms can be specified via the Algorithm parameter:

  • ‘RegSVD’ – Regularized SVD using a SGD optimizer
  • ‘NMF’ – Non-negative matrix factorization with alternating least squares update rules
  • ‘BatchSVD’ – SVD batch learning
  • ‘SVDIncompleteIncremental’ – SVD incomplete incremental learning
  • ‘SVDCompleteIncremental’ – SVD complete incremental learning
  • ‘BiasSVD’ – Bias SVD using a SGD optimizer
  • ‘SVDPP’ – SVD++ using a SGD optimizer

The following neighbor search algorithms can be specified via the NeighborSearch parameter:

  • ‘cosine’ – Cosine Search Algorithm
  • ‘euclidean’ – Euclidean Search Algorithm
  • ‘pearson’ – Pearson Search Algorithm

The following weight interpolation algorithms can be specified via the Interpolation parameter:

  • ‘average’ – Average Interpolation Algorithm
  • ‘regression’ – Regression Interpolation Algorithm
  • ‘similarity’ – Similarity Interpolation Algorithm

The following ranking normalization algorithms can be specified via the Normalization parameter:

  • ‘none’ – No Normalization
  • ‘item_mean’ – Item Mean Normalization
  • ‘overall_mean’ – Overall Mean Normalization
  • ‘user_mean’ – User Mean Normalization
  • ‘z_score’ – Z-Score Normalization

A trained model may be saved to with the OutputModel output parameter.

Example

To train a CF model on a dataset training_set using NMF for decomposition and saving the trained model to model, one could call:

// Initialize optional parameters for Cf().
param := mlpack.CfOptions()
param.Training = training_set
param.Algorithm = "NMF"

_, model := mlpack.Cf(param)

Then, to use this model to generate recommendations for the list of users in the query set users, storing 5 recommendations in recommendations, one could call

// Initialize optional parameters for Cf().
param := mlpack.CfOptions()
param.InputModel = &model
param.Query = users
param.Recommendations = 5

recommendations, _ := mlpack.Cf(param)

See also

Dbscan()

DBSCAN clustering

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Dbscan().
param := mlpack.DbscanOptions()
param.Epsilon = 1
param.MinSize = 5
param.Naive = false
param.SelectionType = "ordered"
param.SingleMode = false
param.TreeType = "kd"

assignments, centroids := mlpack.Dbscan(input, param)

An implementation of DBSCAN clustering. Given a dataset, this can compute and return a clustering of that dataset. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Epsilon float64 Radius of each range search. 1
input *mat.Dense Input dataset to cluster. required
MinSize int Minimum number of points for a cluster. 5
Naive bool If set, brute-force range search (not tree-based) will be used. false
SelectionType string If using point selection policy, the type of selection to use (‘ordered’, ‘random’). "ordered"
SingleMode bool If set, single-tree range search (not dual-tree) will be used. false
TreeType string If using single-tree or dual-tree search, the type of tree to use (‘kd’, ‘r’, ‘r-star’, ‘x’, ‘hilbert-r’, ‘r-plus’, ‘r-plus-plus’, ‘cover’, ‘ball’). "kd"
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
assignments *mat.Dense (1d with ints) Output matrix for assignments of each point.
centroids *mat.Dense Matrix to save output centroids to.

Detailed documentation

This program implements the DBSCAN algorithm for clustering using accelerated tree-based range search. The type of tree that is used may be parameterized, or brute-force range search may also be used.

The input dataset to be clustered may be specified with the Input parameter; the radius of each range search may be specified with the Epsilon parameters, and the minimum number of points in a cluster may be specified with the MinSize parameter.

The Assignments and Centroids output parameters may be used to save the output of the clustering. Assignments contains the cluster assignments of each point, and Centroids contains the centroids of each cluster.

The range search may be controlled with the TreeType, SingleMode, and Naive parameters. TreeType can control the type of tree used for range search; this can take a variety of values: ‘kd’, ‘r’, ‘r-star’, ‘x’, ‘hilbert-r’, ‘r-plus’, ‘r-plus-plus’, ‘cover’, ‘ball’. The SingleMode parameter will force single-tree search (as opposed to the default dual-tree search), and ‘Naive will force brute-force range search.

Example

An example usage to run DBSCAN on the dataset in input with a radius of 0.5 and a minimum cluster size of 5 is given below:

// Initialize optional parameters for Dbscan().
param := mlpack.DbscanOptions()
param.Epsilon = 0.5
param.MinSize = 5

_, _ := mlpack.Dbscan(input, param)

See also

DecisionStump()

Decision Stump

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for DecisionStump().
param := mlpack.DecisionStumpOptions()
param.BucketSize = 6
param.InputModel = nil
param.Labels = mat.NewDense(1, 1, nil)
param.Test = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)

output_model, predictions := mlpack.DecisionStump(param)

An implementation of a decision stump, which is a single-level decision tree. Given labeled data, a new decision stump can be trained; or, an existing decision stump can be used to classify points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
BucketSize int The minimum number of training points in each decision stump bucket. 6
InputModel dsModel Decision stump model to load. nil
Labels *mat.Dense (1d with ints) Labels for the training set. If not specified, the labels are assumed to be the last row of the training data. mat.NewDense(1, 1, nil)
Test *mat.Dense A dataset to calculate predictions for. mat.NewDense(1, 1, nil)
Training *mat.Dense The dataset to train on. mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel dsModel Output decision stump model to save.
predictions *mat.Dense (1d with ints) The output matrix that will hold the predicted labels for the test set.

Detailed documentation

This program implements a decision stump, which is a single-level decision tree. The decision stump will split on one dimension of the input data, and will split into multiple buckets. The dimension and bins are selected by maximizing the information gain of the split. Optionally, the minimum number of training points in each bin can be specified with the BucketSize parameter.

The decision stump is parameterized by a splitting dimension and a vector of values that denote the splitting values of each bin.

This program enables several applications: a decision tree may be trained or loaded, and then that decision tree may be used to classify a given set of test points. The decision tree may also be saved to a file for later usage.

To train a decision stump, training data should be passed with the Training parameter, and their corresponding labels should be passed with the Labels option. Optionally, if Labels is not specified, the labels are assumed to be the last dimension of the training dataset. The BucketSize parameter controls the minimum number of training points in each decision stump bucket.

For classifying a test set, a decision stump may be loaded with the InputModel parameter (useful for the situation where a stump has already been trained), and a test set may be specified with the Test parameter. The predicted labels can be saved with the Predictions output parameter.

Because decision stumps are trained in batch, retraining does not make sense and thus it is not possible to pass both Training and InputModel; instead, simply build a new decision stump with the training data.

After training, a decision stump can be saved with the OutputModel output parameter. That stump may later be re-used in subsequent calls to this program (or others).

See also

DecisionTree()

Decision tree

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for DecisionTree().
param := mlpack.DecisionTreeOptions()
param.InputModel = nil
param.Labels = mat.NewDense(1, 1, nil)
param.MaximumDepth = 0
param.MinimumGainSplit = 1e-07
param.MinimumLeafSize = 20
param.PrintTrainingAccuracy = false
param.PrintTrainingError = false
param.Test = mat.NewDense(1, 1, nil)
param.TestLabels = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)
param.Weights = mat.NewDense(1, 1, nil)

output_model, predictions, probabilities := mlpack.DecisionTree(param)

An implementation of an ID3-style decision tree for classification, which supports categorical data. Given labeled data with numeric or categorical features, a decision tree can be trained and saved; or, an existing decision tree can be used for classification on new points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
InputModel decisionTreeModel Pre-trained decision tree, to be used with test points. nil
Labels *mat.Dense (1d with ints) Training labels. mat.NewDense(1, 1, nil)
MaximumDepth int Maximum depth of the tree (0 means no limit). 0
MinimumGainSplit float64 Minimum gain for node splitting. 1e-07
MinimumLeafSize int Minimum number of points in a leaf. 20
PrintTrainingAccuracy bool Print the training accuracy. false
PrintTrainingError bool Print the training error (deprecated; will be removed in mlpack 4.0.0). false
Test matrixWithInfo Testing dataset (may be categorical). mat.NewDense(1, 1, nil)
TestLabels *mat.Dense (1d with ints) Test point labels, if accuracy calculation is desired. mat.NewDense(1, 1, nil)
Training matrixWithInfo Training dataset (may be categorical). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false
Weights *mat.Dense The weight of labels mat.NewDense(1, 1, nil)

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel decisionTreeModel Output for trained decision tree.
predictions *mat.Dense (1d with ints) Class predictions for each test point.
probabilities *mat.Dense Class probabilities for each test point.

Detailed documentation

Train and evaluate using a decision tree. Given a dataset containing numeric or categorical features, and associated labels for each point in the dataset, this program can train a decision tree on that data.

The training set and associated labels are specified with the Training and Labels parameters, respectively. The labels should be in the range [0, num_classes - 1]. Optionally, if Labels is not specified, the labels are assumed to be the last dimension of the training dataset.

When a model is trained, the OutputModel output parameter may be used to save the trained model. A model may be loaded for predictions with the InputModel parameter. The InputModel parameter may not be specified when the Training parameter is specified. The MinimumLeafSize parameter specifies the minimum number of training points that must fall into each leaf for it to be split. The MinimumGainSplit parameter specifies the minimum gain that is needed for the node to split. The MaximumDepth parameter specifies the maximum depth of the tree. If PrintTrainingError is specified, the training error will be printed.

Test data may be specified with the Test parameter, and if performance numbers are desired for that test set, labels may be specified with the TestLabels parameter. Predictions for each test point may be saved via the Predictions output parameter. Class probabilities for each prediction may be saved with the Probabilities output parameter.

Example

For example, to train a decision tree with a minimum leaf size of 20 on the dataset contained in data with labels labels, saving the output model to tree and printing the training error, one could call

// Initialize optional parameters for DecisionTree().
param := mlpack.DecisionTreeOptions()
param.Training = data
param.Labels = labels
param.MinimumLeafSize = 20
param.MinimumGainSplit = 0.001
param.PrintTrainingAccuracy = true

tree, _, _ := mlpack.DecisionTree(param)

Then, to use that model to classify points in test_set and print the test error given the labels test_labels using that model, while saving the predictions for each point to predictions, one could call

// Initialize optional parameters for DecisionTree().
param := mlpack.DecisionTreeOptions()
param.InputModel = &tree
param.Test = test_set
param.TestLabels = test_labels

_, predictions, _ := mlpack.DecisionTree(param)

See also

Det()

Density Estimation With Density Estimation Trees

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Det().
param := mlpack.DetOptions()
param.Folds = 10
param.InputModel = nil
param.MaxLeafSize = 10
param.MinLeafSize = 5
param.PathFormat = "lr"
param.SkipPruning = false
param.Test = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)

output_model, tag_counters_file, tag_file, test_set_estimates,
    training_set_estimates, vi := mlpack.Det(param)

An implementation of density estimation trees for the density estimation task. Density estimation trees can be trained or used to predict the density at locations given by query points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Folds int The number of folds of cross-validation to perform for the estimation (0 is LOOCV) 10
InputModel dTree Trained density estimation tree to load. nil
MaxLeafSize int The maximum size of a leaf in the unpruned, fully grown DET. 10
MinLeafSize int The minimum size of a leaf in the unpruned, fully grown DET. 5
PathFormat string The format of path printing: ‘lr’, ‘id-lr’, or ‘lr-id’. "lr"
SkipPruning bool Whether to bypass the pruning process and output the unpruned tree only. false
Test *mat.Dense A set of test points to estimate the density of. mat.NewDense(1, 1, nil)
Training *mat.Dense The data set on which to build a density estimation tree. mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel dTree Output to save trained density estimation tree to.
tagCountersFile string The file to output the number of points that went to each leaf.
tagFile string The file to output the tags (and possibly paths) for each sample in the test set.
testSetEstimates *mat.Dense The output estimates on the test set from the final optimally pruned tree.
trainingSetEstimates *mat.Dense The output density estimates on the training set from the final optimally pruned tree.
vi *mat.Dense The output variable importance values for each feature.

Detailed documentation

This program performs a number of functions related to Density Estimation Trees. The optimal Density Estimation Tree (DET) can be trained on a set of data (specified by Training) using cross-validation (with number of folds specified with the Folds parameter). This trained density estimation tree may then be saved with the OutputModel output parameter.

The variable importances (that is, the feature importance values for each dimension) may be saved with the Vi output parameter, and the density estimates for each training point may be saved with the TrainingSetEstimates output parameter.

Enabling path printing for each node outputs the path from the root node to a leaf for each entry in the test set, or training set (if a test set is not provided). Strings like ‘LRLRLR’ (indicating that traversal went to the left child, then the right child, then the left child, and so forth) will be output. If ‘lr-id’ or ‘id-lr’ are given as the PathFormat parameter, then the ID (tag) of every node along the path will be printed after or before the L or R character indicating the direction of traversal, respectively.

This program also can provide density estimates for a set of test points, specified in the Test parameter. The density estimation tree used for this task will be the tree that was trained on the given training points, or a tree given as the parameter InputModel. The density estimates for the test points may be saved using the TestSetEstimates output parameter.

See also

Emst()

Fast Euclidean Minimum Spanning Tree

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Emst().
param := mlpack.EmstOptions()
param.LeafSize = 1
param.Naive = false

output := mlpack.Emst(input, param)

An implementation of the Dual-Tree Boruvka algorithm for computing the Euclidean minimum spanning tree of a set of input points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
input *mat.Dense Input data matrix. required
LeafSize int Leaf size in the kd-tree. One-element leaves give the empirically best performance, but at the cost of greater memory requirements. 1
Naive bool Compute the MST using O(n^2) naive algorithm. false
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Output data. Stored as an edge list.

Detailed documentation

This program can compute the Euclidean minimum spanning tree of a set of input points using the dual-tree Boruvka algorithm.

The set to calculate the minimum spanning tree of is specified with the Input parameter, and the output may be saved with the Output output parameter.

The LeafSize parameter controls the leaf size of the kd-tree that is used to calculate the minimum spanning tree, and if the Naive option is given, then brute-force search is used (this is typically much slower in low dimensions). The leaf size does not affect the results, but it may have some effect on the runtime of the algorithm.

Example

For example, the minimum spanning tree of the input dataset data can be calculated with a leaf size of 20 and stored as spanning_tree using the following command:

// Initialize optional parameters for Emst().
param := mlpack.EmstOptions()
param.LeafSize = 20

spanning_tree := mlpack.Emst(data, param)

The output matrix is a three-dimensional matrix, where each row indicates an edge. The first dimension corresponds to the lesser index of the edge; the second dimension corresponds to the greater index of the edge; and the third column corresponds to the distance between the two points.

See also

Fastmks()

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Fastmks().
param := mlpack.FastmksOptions()
param.Bandwidth = 1
param.Base = 2
param.Degree = 2
param.InputModel = nil
param.K = 0
param.Kernel = "linear"
param.Naive = false
param.Offset = 0
param.Query = mat.NewDense(1, 1, nil)
param.Reference = mat.NewDense(1, 1, nil)
param.Scale = 1
param.Single = false

indices, kernels, output_model := mlpack.Fastmks(param)

An implementation of the single-tree and dual-tree fast max-kernel search (FastMKS) algorithm. Given a set of reference points and a set of query points, this can find the reference point with maximum kernel value for each query point; trained models can be reused for future queries. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Bandwidth float64 Bandwidth (for Gaussian, Epanechnikov, and triangular kernels). 1
Base float64 Base to use during cover tree construction. 2
Degree float64 Degree of polynomial kernel. 2
InputModel fastmksModel Input FastMKS model to use. nil
K int Number of maximum kernels to find. 0
Kernel string Kernel type to use: ‘linear’, ‘polynomial’, ‘cosine’, ‘gaussian’, ‘epanechnikov’, ‘triangular’, ‘hyptan’. "linear"
Naive bool If true, O(n^2) naive mode is used for computation. false
Offset float64 Offset of kernel (for polynomial and hyptan kernels). 0
Query *mat.Dense The query dataset. mat.NewDense(1, 1, nil)
Reference *mat.Dense The reference dataset. mat.NewDense(1, 1, nil)
Scale float64 Scale of kernel (for hyptan kernel). 1
Single bool If true, single-tree search is used (as opposed to dual-tree search. false
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
indices *mat.Dense (with ints) Output matrix of indices.
kernels *mat.Dense Output matrix of kernels.
outputModel fastmksModel Output for FastMKS model.

Detailed documentation

This program will find the k maximum kernels of a set of points, using a query set and a reference set (which can optionally be the same set). More specifically, for each point in the query set, the k points in the reference set with maximum kernel evaluations are found. The kernel function used is specified with the Kernel parameter.

Example

For example, the following command will calculate, for each point in the query set query, the five points in the reference set reference with maximum kernel evaluation using the linear kernel. The kernel evaluations may be saved with the kernels output parameter and the indices may be saved with the indices output parameter.

// Initialize optional parameters for Fastmks().
param := mlpack.FastmksOptions()
param.K = 5
param.Reference = reference
param.Query = query
param.Kernel = "linear"

indices, kernels, _ := mlpack.Fastmks(param)

The output matrices are organized such that row i and column j in the indices matrix corresponds to the index of the point in the reference set that has j’th largest kernel evaluation with the point in the query set with index i. Row i and column j in the kernels matrix corresponds to the kernel evaluation between those two points.

This program performs FastMKS using a cover tree. The base used to build the cover tree can be specified with the Base parameter.

See also

GmmTrain()

Gaussian Mixture Model (GMM) Training

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for GmmTrain().
param := mlpack.GmmTrainOptions()
param.DiagonalCovariance = false
param.InputModel = nil
param.KmeansMaxIterations = 1000
param.MaxIterations = 250
param.NoForcePositive = false
param.Noise = 0
param.Percentage = 0.02
param.RefinedStart = false
param.Samplings = 100
param.Seed = 0
param.Tolerance = 1e-10
param.Trials = 1

output_model := mlpack.GmmTrain(gaussians, input, param)

An implementation of the EM algorithm for training Gaussian mixture models (GMMs). Given a dataset, this can train a GMM for future use with other tools. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
DiagonalCovariance bool Force the covariance of the Gaussians to be diagonal. This can accelerate training time significantly. false
gaussians int Number of Gaussians in the GMM. required
input *mat.Dense The training data on which the model will be fit. required
InputModel gmm Initial input GMM model to start training with. nil
KmeansMaxIterations int Maximum number of iterations for the k-means algorithm (used to initialize EM). 1000
MaxIterations int Maximum number of iterations of EM algorithm (passing 0 will run until convergence). 250
NoForcePositive bool Do not force the covariance matrices to be positive definite. false
Noise float64 Variance of zero-mean Gaussian noise to add to data. 0
Percentage float64 If using –refined_start, specify the percentage of the dataset used for each sampling (should be between 0.0 and 1.0). 0.02
RefinedStart bool During the initialization, use refined initial positions for k-means clustering (Bradley and Fayyad, 1998). false
Samplings int If using –refined_start, specify the number of samplings used for initial points. 100
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
Tolerance float64 Tolerance for convergence of EM. 1e-10
Trials int Number of trials to perform in training GMM. 1
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel gmm Output for trained GMM model.

Detailed documentation

This program takes a parametric estimate of a Gaussian mixture model (GMM) using the EM algorithm to find the maximum likelihood estimate. The model may be saved and reused by other mlpack GMM tools.

The input data to train on must be specified with the Input parameter, and the number of Gaussians in the model must be specified with the Gaussians parameter. Optionally, many trials with different random initializations may be run, and the result with highest log-likelihood on the training data will be taken. The number of trials to run is specified with the Trials parameter. By default, only one trial is run.

The tolerance for convergence and maximum number of iterations of the EM algorithm are specified with the Tolerance and MaxIterations parameters, respectively. The GMM may be initialized for training with another model, specified with the InputModel parameter. Otherwise, the model is initialized by running k-means on the data. The k-means clustering initialization can be controlled with the KmeansMaxIterations, RefinedStart, Samplings, and Percentage parameters. If RefinedStart is specified, then the Bradley-Fayyad refined start initialization will be used. This can often lead to better clustering results.

The ‘diagonal_covariance’ flag will cause the learned covariances to be diagonal matrices. This significantly simplifies the model itself and causes training to be faster, but restricts the ability to fit more complex GMMs.

If GMM training fails with an error indicating that a covariance matrix could not be inverted, make sure that the NoForcePositive parameter is not specified. Alternately, adding a small amount of Gaussian noise (using the Noise parameter) to the entire dataset may help prevent Gaussians with zero variance in a particular dimension, which is usually the cause of non-invertible covariance matrices.

The NoForcePositive parameter, if set, will avoid the checks after each iteration of the EM algorithm which ensure that the covariance matrices are positive definite. Specifying the flag can cause faster runtime, but may also cause non-positive definite covariance matrices, which will cause the program to crash.

Example

As an example, to train a 6-Gaussian GMM on the data in data with a maximum of 100 iterations of EM and 3 trials, saving the trained GMM to gmm, the following command can be used:

// Initialize optional parameters for GmmTrain().
param := mlpack.GmmTrainOptions()
param.Trials = 3

gmm := mlpack.GmmTrain(data, 6, param)

To re-train that GMM on another set of data data2, the following command may be used:

// Initialize optional parameters for GmmTrain().
param := mlpack.GmmTrainOptions()
param.InputModel = &gmm

new_gmm := mlpack.GmmTrain(data2, 6, param)

See also

GmmGenerate()

GMM Sample Generator

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for GmmGenerate().
param := mlpack.GmmGenerateOptions()
param.Seed = 0

output := mlpack.GmmGenerate(inputModel, samples, param)

A sample generator for pre-trained GMMs. Given a pre-trained GMM, this can sample new points randomly from that distribution. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
inputModel gmm Input GMM model to generate samples from. required
samples int Number of samples to generate. required
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Matrix to save output samples in.

Detailed documentation

This program is able to generate samples from a pre-trained GMM (use gmm_train to train a GMM). The pre-trained GMM must be specified with the InputModel parameter. The number of samples to generate is specified by the Samples parameter. Output samples may be saved with the Output output parameter.

Example

The following command can be used to generate 100 samples from the pre-trained GMM gmm and store those generated samples in samples:

// Initialize optional parameters for GmmGenerate().
param := mlpack.GmmGenerateOptions()

samples := mlpack.GmmGenerate(&gmm, 100, param)

See also

GmmProbability()

GMM Probability Calculator

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for GmmProbability().
param := mlpack.GmmProbabilityOptions()

output := mlpack.GmmProbability(input, inputModel, )

A probability calculator for GMMs. Given a pre-trained GMM and a set of points, this can compute the probability that each point is from the given GMM. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
input *mat.Dense Input matrix to calculate probabilities of. required
inputModel gmm Input GMM to use as model. required
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Matrix to store calculated probabilities in.

Detailed documentation

This program calculates the probability that given points came from a given GMM (that is, P(X | gmm)). The GMM is specified with the InputModel parameter, and the points are specified with the Input parameter. The output probabilities may be saved via the Output output parameter.

Example

So, for example, to calculate the probabilities of each point in points coming from the pre-trained GMM gmm, while storing those probabilities in probs, the following command could be used:

// Initialize optional parameters for GmmProbability().
param := mlpack.GmmProbabilityOptions()

probs := mlpack.GmmProbability(&gmm, points, param)

See also

HmmTrain()

Hidden Markov Model (HMM) Training

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for HmmTrain().
param := mlpack.HmmTrainOptions()
param.Batch = false
param.Gaussians = 0
param.InputModel = nil
param.LabelsFile = ""
param.Seed = 0
param.States = 0
param.Tolerance = 1e-05
param.Type = "gaussian"

output_model := mlpack.HmmTrain(inputFile, param)

An implementation of training algorithms for Hidden Markov Models (HMMs). Given labeled or unlabeled data, an HMM can be trained for further use with other mlpack HMM tools. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Batch bool If true, input_file (and if passed, labels_file) are expected to contain a list of files to use as input observation sequences (and label sequences). false
Gaussians int Number of gaussians in each GMM (necessary when type is ‘gmm’). 0
inputFile string File containing input observations. required
InputModel hmmModel Pre-existing HMM model to initialize training with. nil
LabelsFile string Optional file of hidden states, used for labeled training. ""
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
States int Number of hidden states in HMM (necessary, unless model_file is specified). 0
Tolerance float64 Tolerance of the Baum-Welch algorithm. 1e-05
Type string Type of HMM: discrete | gaussian | diag_gmm | gmm. "gaussian"
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel hmmModel Output for trained HMM.

Detailed documentation

This program allows a Hidden Markov Model to be trained on labeled or unlabeled data. It supports four types of HMMs: Discrete HMMs, Gaussian HMMs, GMM HMMs, or Diagonal GMM HMMs

Either one input sequence can be specified (with InputFile), or, a file containing files in which input sequences can be found (when InputFileandBatch are used together). In addition, labels can be provided in the file specified by LabelsFile, and if Batch is used, the file given to LabelsFile should contain a list of files of labels corresponding to the sequences in the file given to InputFile.

The HMM is trained with the Baum-Welch algorithm if no labels are provided. The tolerance of the Baum-Welch algorithm can be set with the Toleranceoption. By default, the transition matrix is randomly initialized and the emission distributions are initialized to fit the extent of the data.

Optionally, a pre-created HMM model can be used as a guess for the transition matrix and emission probabilities; this is specifiable with OutputModel.

See also

HmmLoglik()

Hidden Markov Model (HMM) Sequence Log-Likelihood

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for HmmLoglik().
param := mlpack.HmmLoglikOptions()

log_likelihood := mlpack.HmmLoglik(input, inputModel, )

A utility for computing the log-likelihood of a sequence for Hidden Markov Models (HMMs). Given a pre-trained HMM and an observation sequence, this computes and returns the log-likelihood of that sequence being observed from that HMM. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
input *mat.Dense File containing observations, required
inputModel hmmModel File containing HMM. required
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
logLikelihood float64 Log-likelihood of the sequence.

Detailed documentation

This utility takes an already-trained HMM, specified with the InputModel parameter, and evaluates the log-likelihood of a sequence of observations, given with the Input parameter. The computed log-likelihood is given as output.

Example

For example, to compute the log-likelihood of the sequence seq with the pre-trained HMM hmm, the following command may be used:

// Initialize optional parameters for HmmLoglik().
param := mlpack.HmmLoglikOptions()

_ := mlpack.HmmLoglik(seq, &hmm, param)

See also

HmmViterbi()

Hidden Markov Model (HMM) Viterbi State Prediction

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for HmmViterbi().
param := mlpack.HmmViterbiOptions()

output := mlpack.HmmViterbi(input, inputModel, )

A utility for computing the most probable hidden state sequence for Hidden Markov Models (HMMs). Given a pre-trained HMM and an observed sequence, this uses the Viterbi algorithm to compute and return the most probable hidden state sequence. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
input *mat.Dense Matrix containing observations, required
inputModel hmmModel Trained HMM to use. required
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense (with ints) File to save predicted state sequence to.

Detailed documentation

This utility takes an already-trained HMM, specified as InputModel, and evaluates the most probable hidden state sequence of a given sequence of observations (specified as ‘Input, using the Viterbi algorithm. The computed state sequence may be saved using the Output output parameter.

Example

For example, to predict the state sequence of the observations obs using the HMM hmm, storing the predicted state sequence to states, the following command could be used:

// Initialize optional parameters for HmmViterbi().
param := mlpack.HmmViterbiOptions()

states := mlpack.HmmViterbi(obs, &hmm, param)

See also

HmmGenerate()

Hidden Markov Model (HMM) Sequence Generator

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for HmmGenerate().
param := mlpack.HmmGenerateOptions()
param.Seed = 0
param.StartState = 0

output, state := mlpack.HmmGenerate(length, model, param)

A utility to generate random sequences from a pre-trained Hidden Markov Model (HMM). The length of the desired sequence can be specified, and a random sequence of observations is returned. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
length int Length of sequence to generate. required
model hmmModel Trained HMM to generate sequences with. required
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
StartState int Starting state of sequence. 0
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Matrix to save observation sequence to.
state *mat.Dense (with ints) Matrix to save hidden state sequence to.

Detailed documentation

This utility takes an already-trained HMM, specified as the Model parameter, and generates a random observation sequence and hidden state sequence based on its parameters. The observation sequence may be saved with the Output output parameter, and the internal state sequence may be saved with the State output parameter.

The state to start the sequence in may be specified with the StartState parameter.

Example

For example, to generate a sequence of length 150 from the HMM hmm and save the observation sequence to observations and the hidden state sequence to states, the following command may be used:

// Initialize optional parameters for HmmGenerate().
param := mlpack.HmmGenerateOptions()

observations, states := mlpack.HmmGenerate(&hmm, 150, param)

See also

HoeffdingTree()

Hoeffding trees

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for HoeffdingTree().
param := mlpack.HoeffdingTreeOptions()
param.BatchMode = false
param.Bins = 10
param.Confidence = 0.95
param.InfoGain = false
param.InputModel = nil
param.Labels = mat.NewDense(1, 1, nil)
param.MaxSamples = 5000
param.MinSamples = 100
param.NumericSplitStrategy = "binary"
param.ObservationsBeforeBinning = 100
param.Passes = 1
param.Test = mat.NewDense(1, 1, nil)
param.TestLabels = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)

output_model, predictions, probabilities := mlpack.HoeffdingTree(param)

An implementation of Hoeffding trees, a form of streaming decision tree for classification. Given labeled data, a Hoeffding tree can be trained and saved for later use, or a pre-trained Hoeffding tree can be used for predicting the classifications of new points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
BatchMode bool If true, samples will be considered in batch instead of as a stream. This generally results in better trees but at the cost of memory usage and runtime. false
Bins int If the ‘domingos’ split strategy is used, this specifies the number of bins for each numeric split. 10
Confidence float64 Confidence before splitting (between 0 and 1). 0.95
InfoGain bool If set, information gain is used instead of Gini impurity for calculating Hoeffding bounds. false
InputModel hoeffdingTreeModel Input trained Hoeffding tree model. nil
Labels *mat.Dense (1d with ints) Labels for training dataset. mat.NewDense(1, 1, nil)
MaxSamples int Maximum number of samples before splitting. 5000
MinSamples int Minimum number of samples before splitting. 100
NumericSplitStrategy string The splitting strategy to use for numeric features: ‘domingos’ or ‘binary’. "binary"
ObservationsBeforeBinning int If the ‘domingos’ split strategy is used, this specifies the number of samples observed before binning is performed. 100
Passes int Number of passes to take over the dataset. 1
Test matrixWithInfo Testing dataset (may be categorical). mat.NewDense(1, 1, nil)
TestLabels *mat.Dense (1d with ints) Labels of test data. mat.NewDense(1, 1, nil)
Training matrixWithInfo Training dataset (may be categorical). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel hoeffdingTreeModel Output for trained Hoeffding tree model.
predictions *mat.Dense (1d with ints) Matrix to output label predictions for test data into.
probabilities *mat.Dense In addition to predicting labels, provide rediction probabilities in this matrix.

Detailed documentation

This program implements Hoeffding trees, a form of streaming decision tree suited best for large (or streaming) datasets. This program supports both categorical and numeric data. Given an input dataset, this program is able to train the tree with numerous training options, and save the model to a file. The program is also able to use a trained model or a model from file in order to predict classes for a given test set.

The training file and associated labels are specified with the Training and Labels parameters, respectively. Optionally, if Labels is not specified, the labels are assumed to be the last dimension of the training dataset.

The training may be performed in batch mode (like a typical decision tree algorithm) by specifying the BatchMode option, but this may not be the best option for large datasets.

When a model is trained, it may be saved via the OutputModel output parameter. A model may be loaded from file for further training or testing with the InputModel parameter.

Test data may be specified with the Test parameter, and if performance statistics are desired for that test set, labels may be specified with the TestLabels parameter. Predictions for each test point may be saved with the Predictions output parameter, and class probabilities for each prediction may be saved with the Probabilities output parameter.

Example

For example, to train a Hoeffding tree with confidence 0.99 with data dataset, saving the trained tree to tree, the following command may be used:

// Initialize optional parameters for HoeffdingTree().
param := mlpack.HoeffdingTreeOptions()
param.Training = dataset
param.Confidence = 0.99

tree, _, _ := mlpack.HoeffdingTree(param)

Then, this tree may be used to make predictions on the test set test_set, saving the predictions into predictions and the class probabilities into class_probs with the following command:

// Initialize optional parameters for HoeffdingTree().
param := mlpack.HoeffdingTreeOptions()
param.InputModel = &tree
param.Test = test_set

_, predictions, class_probs := mlpack.HoeffdingTree(param)

See also

Kde()

Kernel Density Estimation

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Kde().
param := mlpack.KdeOptions()
param.AbsError = 0
param.Algorithm = "dual-tree"
param.Bandwidth = 1
param.InitialSampleSize = 100
param.InputModel = nil
param.Kernel = "gaussian"
param.McBreakCoef = 0.4
param.McEntryCoef = 3
param.McProbability = 0.95
param.MonteCarlo = false
param.Query = mat.NewDense(1, 1, nil)
param.Reference = mat.NewDense(1, 1, nil)
param.RelError = 0.05
param.Tree = "kd-tree"

output_model, predictions := mlpack.Kde(param)

An implementation of kernel density estimation with dual-tree algorithms. Given a set of reference points and query points and a kernel function, this can estimate the density function at the location of each query point using trees; trees that are built can be saved for later use. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
AbsError float64 Relative error tolerance for the prediction. 0
Algorithm string Algorithm to use for the prediction.(‘dual-tree’, ‘single-tree’). "dual-tree"
Bandwidth float64 Bandwidth of the kernel. 1
InitialSampleSize int Initial sample size for Monte Carlo estimations. 100
InputModel kdeModel Contains pre-trained KDE model. nil
Kernel string Kernel to use for the prediction.(‘gaussian’, ‘epanechnikov’, ‘laplacian’, ‘spherical’, ‘triangular’). "gaussian"
McBreakCoef float64 Controls what fraction of the amount of node’s descendants is the limit for the sample size before it recurses. 0.4
McEntryCoef float64 Controls how much larger does the amount of node descendants has to be compared to the initial sample size in order to be a candidate for Monte Carlo estimations. 3
McProbability float64 Probability of the estimation being bounded by relative error when using Monte Carlo estimations. 0.95
MonteCarlo bool Whether to use Monte Carlo estimations when possible. false
Query *mat.Dense Query dataset to KDE on. mat.NewDense(1, 1, nil)
Reference *mat.Dense Input reference dataset use for KDE. mat.NewDense(1, 1, nil)
RelError float64 Relative error tolerance for the prediction. 0.05
Tree string Tree to use for the prediction.(‘kd-tree’, ‘ball-tree’, ‘cover-tree’, ‘octree’, ‘r-tree’). "kd-tree"
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel kdeModel If specified, the KDE model will be saved here.
predictions *mat.Dense (1d) Vector to store density predictions.

Detailed documentation

This program performs a Kernel Density Estimation. KDE is a non-parametric way of estimating probability density function. For each query point the program will estimate its probability density by applying a kernel function to each reference point. The computational complexity of this is O(N^2) where there are N query points and N reference points, but this implementation will typically see better performance as it uses an approximate dual or single tree algorithm for acceleration.

Dual or single tree optimization avoids many barely relevant calculations (as kernel function values decrease with distance), so it is an approximate computation. You can specify the maximum relative error tolerance for each query value with RelError as well as the maximum absolute error tolerance with the parameter AbsError. This program runs using an Euclidean metric. Kernel function can be selected using the Kernel option. You can also choose what which type of tree to use for the dual-tree algorithm with Tree. It is also possible to select whether to use dual-tree algorithm or single-tree algorithm using the Algorithm option.

Monte Carlo estimations can be used to accelerate the KDE estimate when the Gaussian Kernel is used. This provides a probabilistic guarantee on the the error of the resulting KDE instead of an absolute guarantee.To enable Monte Carlo estimations, the MonteCarlo flag can be used, and success probability can be set with the McProbability option. It is possible to set the initial sample size for the Monte Carlo estimation using InitialSampleSize. This implementation will only consider a node, as a candidate for the Monte Carlo estimation, if its number of descendant nodes is bigger than the initial sample size. This can be controlled using a coefficient that will multiply the initial sample size and can be set using McEntryCoef. To avoid using the same amount of computations an exact approach would take, this program recurses the tree whenever a fraction of the amount of the node’s descendant points have already been computed. This fraction is set using McBreakCoef.

Example

For example, the following will run KDE using the data in ref_data for training and the data in qu_data as query data. It will apply an Epanechnikov kernel with a 0.2 bandwidth to each reference point and use a KD-Tree for the dual-tree optimization. The returned predictions will be within 5% of the real KDE value for each query point.

// Initialize optional parameters for Kde().
param := mlpack.KdeOptions()
param.Reference = ref_data
param.Query = qu_data
param.Bandwidth = 0.2
param.Kernel = "epanechnikov"
param.Tree = "kd-tree"
param.RelError = 0.05

_, out_data := mlpack.Kde(param)

the predicted density estimations will be stored in out_data. If no Query is provided, then KDE will be computed on the Reference dataset. It is possible to select either a reference dataset or an input model but not both at the same time. If an input model is selected and parameter values are not set (e.g. Bandwidth) then default parameter values will be used.

In addition to the last program call, it is also possible to activate Monte Carlo estimations if a Gaussian kernel is used. This can provide faster results, but the KDE will only have a probabilistic guarantee of meeting the desired error bound (instead of an absolute guarantee). The following example will run KDE using a Monte Carlo estimation when possible. The results will be within a 5% of the real KDE value with a 95% probability. Initial sample size for the Monte Carlo estimation will be 200 points and a node will be a candidate for the estimation only when it contains 700 (i.e. 3.5200) points. If a node contains 700 points and 420 (i.e. 0.6700) have already been sampled, then the algorithm will recurse instead of keep sampling.

// Initialize optional parameters for Kde().
param := mlpack.KdeOptions()
param.Reference = ref_data
param.Query = qu_data
param.Bandwidth = 0.2
param.Kernel = "gaussian"
param.Tree = "kd-tree"
param.RelError = 0.05
param.MonteCarlo = 
param.McProbability = 0.95
param.InitialSampleSize = 200
param.McEntryCoef = 3.5
param.McBreakCoef = 0.6

_, out_data := mlpack.Kde(param)

See also

KernelPca()

Kernel Principal Components Analysis

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for KernelPca().
param := mlpack.KernelPcaOptions()
param.Bandwidth = 1
param.Center = false
param.Degree = 1
param.KernelScale = 1
param.NewDimensionality = 0
param.NystroemMethod = false
param.Offset = 0
param.Sampling = "kmeans"

output := mlpack.KernelPca(input, kernel, param)

An implementation of Kernel Principal Components Analysis (KPCA). This can be used to perform nonlinear dimensionality reduction or preprocessing on a given dataset. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Bandwidth float64 Bandwidth, for ‘gaussian’ and ‘laplacian’ kernels. 1
Center bool If set, the transformed data will be centered about the origin. false
Degree float64 Degree of polynomial, for ‘polynomial’ kernel. 1
input *mat.Dense Input dataset to perform KPCA on. required
kernel string The kernel to use; see the above documentation for the list of usable kernels. required
KernelScale float64 Scale, for ‘hyptan’ kernel. 1
NewDimensionality int If not 0, reduce the dimensionality of the output dataset by ignoring the dimensions with the smallest eigenvalues. 0
NystroemMethod bool If set, the Nystroem method will be used. false
Offset float64 Offset, for ‘hyptan’ and ‘polynomial’ kernels. 0
Sampling string Sampling scheme to use for the Nystroem method: ‘kmeans’, ‘random’, ‘ordered’ "kmeans"
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Matrix to save modified dataset to.

Detailed documentation

This program performs Kernel Principal Components Analysis (KPCA) on the specified dataset with the specified kernel. This will transform the data onto the kernel principal components, and optionally reduce the dimensionality by ignoring the kernel principal components with the smallest eigenvalues.

For the case where a linear kernel is used, this reduces to regular PCA.

The kernels that are supported are listed below:

  • ‘linear’: the standard linear dot product (same as normal PCA): K(x, y) = x^T y

  • ‘gaussian’: a Gaussian kernel; requires bandwidth: K(x, y) = exp(-(|| x - y || ^ 2) / (2 * (bandwidth ^ 2)))

  • ‘polynomial’: polynomial kernel; requires offset and degree: K(x, y) = (x^T y + offset) ^ degree

  • ‘hyptan’: hyperbolic tangent kernel; requires scale and offset: K(x, y) = tanh(scale * (x^T y) + offset)

  • ‘laplacian’: Laplacian kernel; requires bandwidth: K(x, y) = exp(-(|| x - y ||) / bandwidth)

  • ‘epanechnikov’: Epanechnikov kernel; requires bandwidth: K(x, y) = max(0, 1 - || x - y ||^2 / bandwidth^2)

  • ‘cosine’: cosine distance: K(x, y) = 1 - (x^T y) / (|| x || * || y ||)

The parameters for each of the kernels should be specified with the options Bandwidth, KernelScale, Offset, or Degree (or a combination of those parameters).

Optionally, the Nystroem method (“Using the Nystroem method to speed up kernel machines”, 2001) can be used to calculate the kernel matrix by specifying the NystroemMethod parameter. This approach works by using a subset of the data as basis to reconstruct the kernel matrix; to specify the sampling scheme, the Sampling parameter is used. The sampling scheme for the Nystroem method can be chosen from the following list: ‘kmeans’, ‘random’, ‘ordered’.

Example

For example, the following command will perform KPCA on the dataset input using the Gaussian kernel, and saving the transformed data to transformed:

// Initialize optional parameters for KernelPca().
param := mlpack.KernelPcaOptions()

transformed := mlpack.KernelPca(input, "gaussian", param)

See also

Kmeans()

K-Means Clustering

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Kmeans().
param := mlpack.KmeansOptions()
param.Algorithm = "naive"
param.AllowEmptyClusters = false
param.InPlace = false
param.InitialCentroids = mat.NewDense(1, 1, nil)
param.KillEmptyClusters = false
param.LabelsOnly = false
param.MaxIterations = 1000
param.Percentage = 0.02
param.RefinedStart = false
param.Samplings = 100
param.Seed = 0

centroid, output := mlpack.Kmeans(clusters, input, param)

An implementation of several strategies for efficient k-means clustering. Given a dataset and a value of k, this computes and returns a k-means clustering on that data. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Algorithm string Algorithm to use for the Lloyd iteration (‘naive’, ‘pelleg-moore’, ‘elkan’, ‘hamerly’, ‘dualtree’, or ‘dualtree-covertree’). "naive"
AllowEmptyClusters bool Allow empty clusters to be persist. false
clusters int Number of clusters to find (0 autodetects from initial centroids). required
InPlace bool If specified, a column containing the learned cluster assignments will be added to the input dataset file. In this case, –output_file is overridden. (Do not use in Python.) false
InitialCentroids *mat.Dense Start with the specified initial centroids. mat.NewDense(1, 1, nil)
input *mat.Dense Input dataset to perform clustering on. required
KillEmptyClusters bool Remove empty clusters when they occur. false
LabelsOnly bool Only output labels into output file. false
MaxIterations int Maximum number of iterations before k-means terminates. 1000
Percentage float64 Percentage of dataset to use for each refined start sampling (use when –refined_start is specified). 0.02
RefinedStart bool Use the refined initial point strategy by Bradley and Fayyad to choose initial points. false
Samplings int Number of samplings to perform for refined start (use when –refined_start is specified). 100
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
centroid *mat.Dense If specified, the centroids of each cluster will be written to the given file.
output *mat.Dense Matrix to store output labels or labeled data to.

Detailed documentation

This program performs K-Means clustering on the given dataset. It can return the learned cluster assignments, and the centroids of the clusters. Empty clusters are not allowed by default; when a cluster becomes empty, the point furthest from the centroid of the cluster with maximum variance is taken to fill that cluster.

Optionally, the Bradley and Fayyad approach (“Refining initial points for k-means clustering”, 1998) can be used to select initial points by specifying the RefinedStart parameter. This approach works by taking random samplings of the dataset; to specify the number of samplings, the Samplings parameter is used, and to specify the percentage of the dataset to be used in each sample, the Percentage parameter is used (it should be a value between 0.0 and 1.0).

There are several options available for the algorithm used for each Lloyd iteration, specified with the Algorithm option. The standard O(kN) approach can be used (‘naive’). Other options include the Pelleg-Moore tree-based algorithm (‘pelleg-moore’), Elkan’s triangle-inequality based algorithm (‘elkan’), Hamerly’s modification to Elkan’s algorithm (‘hamerly’), the dual-tree k-means algorithm (‘dualtree’), and the dual-tree k-means algorithm using the cover tree (‘dualtree-covertree’).

The behavior for when an empty cluster is encountered can be modified with the AllowEmptyClusters option. When this option is specified and there is a cluster owning no points at the end of an iteration, that cluster’s centroid will simply remain in its position from the previous iteration. If the KillEmptyClusters option is specified, then when a cluster owns no points at the end of an iteration, the cluster centroid is simply filled with DBL_MAX, killing it and effectively reducing k for the rest of the computation. Note that the default option when neither empty cluster option is specified can be time-consuming to calculate; therefore, specifying either of these parameters will often accelerate runtime.

Initial clustering assignments may be specified using the InitialCentroids parameter, and the maximum number of iterations may be specified with the MaxIterations parameter.

Example

As an example, to use Hamerly’s algorithm to perform k-means clustering with k=10 on the dataset data, saving the centroids to centroids and the assignments for each point to assignments, the following command could be used:

// Initialize optional parameters for Kmeans().
param := mlpack.KmeansOptions()

centroids, assignments := mlpack.Kmeans(data, 10, param)

To run k-means on that same dataset with initial centroids specified in initial with a maximum of 500 iterations, storing the output centroids in final the following command may be used:

// Initialize optional parameters for Kmeans().
param := mlpack.KmeansOptions()
param.InitialCentroids = initial
param.MaxIterations = 500

final, _ := mlpack.Kmeans(data, 10, param)

See also

Lars()

LARS

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Lars().
param := mlpack.LarsOptions()
param.Input = mat.NewDense(1, 1, nil)
param.InputModel = nil
param.Lambda1 = 0
param.Lambda2 = 0
param.Responses = mat.NewDense(1, 1, nil)
param.Test = mat.NewDense(1, 1, nil)
param.UseCholesky = false

output_model, output_predictions := mlpack.Lars(param)

An implementation of Least Angle Regression (Stagewise/laSso), also known as LARS. This can train a LARS/LASSO/Elastic Net model and use that model or a pre-trained model to output regression predictions for a test set. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Input *mat.Dense Matrix of covariates (X). mat.NewDense(1, 1, nil)
InputModel lars Trained LARS model to use. nil
Lambda1 float64 Regularization parameter for l1-norm penalty. 0
Lambda2 float64 Regularization parameter for l2-norm penalty. 0
Responses *mat.Dense Matrix of responses/observations (y). mat.NewDense(1, 1, nil)
Test *mat.Dense Matrix containing points to regress on (test points). mat.NewDense(1, 1, nil)
UseCholesky bool Use Cholesky decomposition during computation rather than explicitly computing the full Gram matrix. false
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel lars Output LARS model.
outputPredictions *mat.Dense If –test_file is specified, this file is where the predicted responses will be saved.

Detailed documentation

An implementation of LARS: Least Angle Regression (Stagewise/laSso). This is a stage-wise homotopy-based algorithm for L1-regularized linear regression (LASSO) and L1+L2-regularized linear regression (Elastic Net).

This program is able to train a LARS/LASSO/Elastic Net model or load a model from file, output regression predictions for a test set, and save the trained model to a file. The LARS algorithm is described in more detail below:

Let X be a matrix where each row is a point and each column is a dimension, and let y be a vector of targets.

The Elastic Net problem is to solve

min_beta 0.5 || X * beta - y ||_2^2 + lambda_1 ||beta||_1 + 0.5 lambda_2 ||beta||_2^2

If lambda1 > 0 and lambda2 = 0, the problem is the LASSO. If lambda1 > 0 and lambda2 > 0, the problem is the Elastic Net. If lambda1 = 0 and lambda2 > 0, the problem is ridge regression. If lambda1 = 0 and lambda2 = 0, the problem is unregularized linear regression.

For efficiency reasons, it is not recommended to use this algorithm with Lambda1 = 0. In that case, use the ‘linear_regression’ program, which implements both unregularized linear regression and ridge regression.

To train a LARS/LASSO/Elastic Net model, the Input and Responses parameters must be given. The Lambda1, Lambda2, and UseCholesky parameters control the training options. A trained model can be saved with the OutputModel. If no training is desired at all, a model can be passed via the InputModel parameter.

The program can also provide predictions for test data using either the trained model or the given input model. Test points can be specified with the Test parameter. Predicted responses to the test points can be saved with the OutputPredictions output parameter.

Example

For example, the following command trains a model on the data data and responses responses with lambda1 set to 0.4 and lambda2 set to 0 (so, LASSO is being solved), and then the model is saved to lasso_model:

// Initialize optional parameters for Lars().
param := mlpack.LarsOptions()
param.Input = data
param.Responses = responses
param.Lambda1 = 0.4
param.Lambda2 = 0

lasso_model, _ := mlpack.Lars(param)

The following command uses the lasso_model to provide predicted responses for the data test and save those responses to test_predictions:

// Initialize optional parameters for Lars().
param := mlpack.LarsOptions()
param.InputModel = &lasso_model
param.Test = test

_, test_predictions := mlpack.Lars(param)

See also

LinearRegression()

Simple Linear Regression and Prediction

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for LinearRegression().
param := mlpack.LinearRegressionOptions()
param.InputModel = nil
param.Lambda = 0
param.Test = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)
param.TrainingResponses = mat.NewDense(1, 1, nil)

output_model, output_predictions := mlpack.LinearRegression(param)

An implementation of simple linear regression and ridge regression using ordinary least squares. Given a dataset and responses, a model can be trained and saved for later use, or a pre-trained model can be used to output regression predictions for a test set. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
InputModel linearRegression Existing LinearRegression model to use. nil
Lambda float64 Tikhonov regularization for ridge regression. If 0, the method reduces to linear regression. 0
Test *mat.Dense Matrix containing X’ (test regressors). mat.NewDense(1, 1, nil)
Training *mat.Dense Matrix containing training set X (regressors). mat.NewDense(1, 1, nil)
TrainingResponses *mat.Dense (1d) Optional vector containing y (responses). If not given, the responses are assumed to be the last row of the input file. mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel linearRegression Output LinearRegression model.
outputPredictions *mat.Dense (1d) If –test_file is specified, this matrix is where the predicted responses will be saved.

Detailed documentation

An implementation of simple linear regression and simple ridge regression using ordinary least squares. This solves the problem

y = X * b + e

where X (specified by Training) and y (specified either as the last column of the input matrix Training or via the TrainingResponses parameter) are known and b is the desired variable. If the covariance matrix (X’X) is not invertible, or if the solution is overdetermined, then specify a Tikhonov regularization constant (with Lambda) greater than 0, which will regularize the covariance matrix to make it invertible. The calculated b may be saved with the OutputPredictions output parameter.

Optionally, the calculated value of b is used to predict the responses for another matrix X’ (specified by the Test parameter):

y’ = X’ * b

and the predicted responses y’ may be saved with the OutputPredictions output parameter. This type of regression is related to least-angle regression, which mlpack implements as the ‘lars’ program.

Example

For example, to run a linear regression on the dataset X with responses y, saving the trained model to lr_model, the following command could be used:

// Initialize optional parameters for LinearRegression().
param := mlpack.LinearRegressionOptions()
param.Training = X
param.TrainingResponses = y

lr_model, _ := mlpack.LinearRegression(param)

Then, to use lr_model to predict responses for a test set X_test, saving the predictions to X_test_responses, the following command could be used:

// Initialize optional parameters for LinearRegression().
param := mlpack.LinearRegressionOptions()
param.InputModel = &lr_model
param.Test = X_test

_, X_test_responses := mlpack.LinearRegression(param)

See also

LinearSvm()

Linear SVM is an L2-regularized support vector machine.

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for LinearSvm().
param := mlpack.LinearSvmOptions()
param.Delta = 1
param.Epochs = 50
param.InputModel = nil
param.Labels = mat.NewDense(1, 1, nil)
param.Lambda = 0.0001
param.MaxIterations = 10000
param.NoIntercept = false
param.NumClasses = 0
param.Optimizer = "lbfgs"
param.Seed = 0
param.Shuffle = false
param.StepSize = 0.01
param.Test = mat.NewDense(1, 1, nil)
param.TestLabels = mat.NewDense(1, 1, nil)
param.Tolerance = 1e-10
param.Training = mat.NewDense(1, 1, nil)

output_model, predictions, probabilities := mlpack.LinearSvm(param)

An implementation of linear SVM for multiclass classification. Given labeled data, a model can be trained and saved for future use; or, a pre-trained model can be used to classify new points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Delta float64 Margin of difference between correct class and other classes. 1
Epochs int Maximum number of full epochs over dataset for psgd 50
InputModel linearsvmModel Existing model (parameters). nil
Labels *mat.Dense (1d with ints) A matrix containing labels (0 or 1) for the points in the training set (y). mat.NewDense(1, 1, nil)
Lambda float64 L2-regularization parameter for training. 0.0001
MaxIterations int Maximum iterations for optimizer (0 indicates no limit). 10000
NoIntercept bool Do not add the intercept term to the model. false
NumClasses int Number of classes for classification; if unspecified (or 0), the number of classes found in the labels will be used. 0
Optimizer string Optimizer to use for training (‘lbfgs’ or ‘psgd’). "lbfgs"
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
Shuffle bool Don’t shuffle the order in which data points are visited for parallel SGD. false
StepSize float64 Step size for parallel SGD optimizer. 0.01
Test *mat.Dense Matrix containing test dataset. mat.NewDense(1, 1, nil)
TestLabels *mat.Dense (1d with ints) Matrix containing test labels. mat.NewDense(1, 1, nil)
Tolerance float64 Convergence tolerance for optimizer. 1e-10
Training *mat.Dense A matrix containing the training set (the matrix of predictors, X). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel linearsvmModel Output for trained linear svm model.
predictions *mat.Dense (1d with ints) If test data is specified, this matrix is where the predictions for the test set will be saved.
probabilities *mat.Dense If test data is specified, this matrix is where the class probabilities for the test set will be saved.

Detailed documentation

An implementation of linear SVMs that uses either L-BFGS or parallel SGD (stochastic gradient descent) to train the model.

This program allows loading a linear SVM model (via the InputModel parameter) or training a linear SVM model given training data (specified with the Training parameter), or both those things at once. In addition, this program allows classification on a test dataset (specified with the Test parameter) and the classification results may be saved with the Predictions output parameter. The trained linear SVM model may be saved using the OutputModel output parameter.

The training data, if specified, may have class labels as its last dimension. Alternately, the Labels parameter may be used to specify a separate vector of labels.

When a model is being trained, there are many options. L2 regularization (to prevent overfitting) can be specified with the Lambda option, and the number of classes can be manually specified with the NumClassesand if an intercept term is not desired in the model, the NoIntercept parameter can be specified.Margin of difference between correct class and other classes can be specified with the Delta option.The optimizer used to train the model can be specified with the Optimizer parameter. Available options are ‘psgd’ (parallel stochastic gradient descent) and ‘lbfgs’ (the L-BFGS optimizer). There are also various parameters for the optimizer; the MaxIterations parameter specifies the maximum number of allowed iterations, and the Tolerance parameter specifies the tolerance for convergence. For the parallel SGD optimizer, the StepSize parameter controls the step size taken at each iteration by the optimizer and the maximum number of epochs (specified with Epochs). If the objective function for your data is oscillating between Inf and 0, the step size is probably too large. There are more parameters for the optimizers, but the C++ interface must be used to access these.

Optionally, the model can be used to predict the labels for another matrix of data points, if Test is specified. The Test parameter can be specified without the Training parameter, so long as an existing linear SVM model is given with the InputModel parameter. The output predictions from the linear SVM model may be saved with the Predictions parameter.

Example

As an example, to train a LinaerSVM on the data ‘data’ with labels ‘labels’ with L2 regularization of 0.1, saving the model to ‘lsvm_model’, the following command may be used:

// Initialize optional parameters for LinearSvm().
param := mlpack.LinearSvmOptions()
param.Training = data
param.Labels = labels
param.Lambda = 0.1
param.Delta = 1
param.NumClasses = 0

lsvm_model, _, _ := mlpack.LinearSvm(param)

Then, to use that model to predict classes for the dataset ‘test’, storing the output predictions in ‘predictions’, the following command may be used:

// Initialize optional parameters for LinearSvm().
param := mlpack.LinearSvmOptions()
param.InputModel = &lsvm_model
param.Test = test

_, predictions, _ := mlpack.LinearSvm(param)

See also

Lmnn()

Large Margin Nearest Neighbors (LMNN)

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Lmnn().
param := mlpack.LmnnOptions()
param.BatchSize = 50
param.Center = false
param.Distance = mat.NewDense(1, 1, nil)
param.K = 1
param.Labels = mat.NewDense(1, 1, nil)
param.LinearScan = false
param.MaxIterations = 100000
param.Normalize = false
param.Optimizer = "amsgrad"
param.Passes = 50
param.PrintAccuracy = false
param.Range = 1
param.Rank = 0
param.Regularization = 0.5
param.Seed = 0
param.StepSize = 0.01
param.Tolerance = 1e-07

centered_data, output, transformed_data := mlpack.Lmnn(input, param)

An implementation of Large Margin Nearest Neighbors (LMNN), a distance learning technique. Given a labeled dataset, this learns a transformation of the data that improves k-nearest-neighbor performance; this can be useful as a preprocessing step. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
BatchSize int Batch size for mini-batch SGD. 50
Center bool Perform mean-centering on the dataset. It is useful when the centroid of the data is far from the origin. false
Distance *mat.Dense Initial distance matrix to be used as starting point mat.NewDense(1, 1, nil)
input *mat.Dense Input dataset to run LMNN on. required
K int Number of target neighbors to use for each datapoint. 1
Labels *mat.Dense (1d with ints) Labels for input dataset. mat.NewDense(1, 1, nil)
LinearScan bool Don’t shuffle the order in which data points are visited for SGD or mini-batch SGD. false
MaxIterations int Maximum number of iterations for L-BFGS (0 indicates no limit). 100000
Normalize bool Use a normalized starting point for optimization. Itis useful for when points are far apart, or when SGD is returning NaN. false
Optimizer string Optimizer to use; ‘amsgrad’, ‘bbsgd’, ‘sgd’, or ‘lbfgs’. "amsgrad"
Passes int Maximum number of full passes over dataset for AMSGrad, BB_SGD and SGD. 50
PrintAccuracy bool Print accuracies on initial and transformed dataset false
Range int Number of iterations after which impostors needs to be recalculated 1
Rank int Rank of distance matrix to be optimized. 0
Regularization float64 Regularization for LMNN objective function 0.5
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
StepSize float64 Step size for AMSGrad, BB_SGD and SGD (alpha). 0.01
Tolerance float64 Maximum tolerance for termination of AMSGrad, BB_SGD, SGD or L-BFGS. 1e-07
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
centeredData *mat.Dense Output matrix for mean-centered dataset.
output *mat.Dense Output matrix for learned distance matrix.
transformedData *mat.Dense Output matrix for transformed dataset.

Detailed documentation

This program implements Large Margin Nearest Neighbors, a distance learning technique. The method seeks to improve k-nearest-neighbor classification on a dataset. The method employes the strategy of reducing distance between similar labeled data points (a.k.a target neighbors) and increasing distance between differently labeled points (a.k.a impostors) using standard optimization techniques over the gradient of the distance between data points.

To work, this algorithm needs labeled data. It can be given as the last row of the input dataset (specified with Input), or alternatively as a separate matrix (specified with Labels). Additionally, a starting point for optimization (specified with Distancecan be given, having (r x d) dimensionality. Here r should satisfy 1 <= r <= d, Consequently a Low-Rank matrix will be optimized. Alternatively, Low-Rank distance can be learned by specifying the Rankparameter (A Low-Rank matrix with uniformly distributed values will be used as initial learning point).

The program also requires number of targets neighbors to work with ( specified with K), A regularization parameter can also be passed, It acts as a trade of between the pulling and pushing terms (specified with Regularization), In addition, this implementation of LMNN includes a parameter to decide the interval after which impostors must be re-calculated (specified with Range).

Output can either be the learned distance matrix (specified with Output), or the transformed dataset (specified with TransformedData), or both. Additionally mean-centered dataset (specified with CenteredData) can be accessed given mean-centering (specified with Center) is performed on the dataset. Accuracy on initial dataset and final transformed dataset can be printed by specifying the PrintAccuracyparameter.

This implementation of LMNN uses AdaGrad, BigBatch_SGD, stochastic gradient descent, mini-batch stochastic gradient descent, or the L_BFGS optimizer.

AdaGrad, specified by the value ‘adagrad’ for the parameter Optimizer, uses maximum of past squared gradients. It primarily on six parameters: the step size (specified with StepSize), the batch size (specified with BatchSize), the maximum number of passes (specified with Passes). Inaddition, a normalized starting point can be used by specifying the Normalize parameter.

BigBatch_SGD, specified by the value ‘bbsgd’ for the parameter Optimizer, depends primarily on four parameters: the step size (specified with StepSize), the batch size (specified with BatchSize), the maximum number of passes (specified with Passes). In addition, a normalized starting point can be used by specifying the Normalize parameter.

Stochastic gradient descent, specified by the value ‘sgd’ for the parameter Optimizer, depends primarily on three parameters: the step size (specified with StepSize), the batch size (specified with BatchSize), and the maximum number of passes (specified with Passes). In addition, a normalized starting point can be used by specifying the Normalize parameter. Furthermore, mean-centering can be performed on the dataset by specifying the Centerparameter.

The L-BFGS optimizer, specified by the value ‘lbfgs’ for the parameter Optimizer, uses a back-tracking line search algorithm to minimize a function. The following parameters are used by L-BFGS: MaxIterations, Tolerance(the optimization is terminated when the gradient norm is below this value). For more details on the L-BFGS optimizer, consult either the mlpack L-BFGS documentation (in lbfgs.hpp) or the vast set of published literature on L-BFGS. In addition, a normalized starting point can be used by specifying the Normalize parameter.

By default, the AMSGrad optimizer is used.

Example

Example - Let’s say we want to learn distance on iris dataset with number of targets as 3 using BigBatch_SGD optimizer. A simple call for the same will look like:

// Initialize optional parameters for MlpackLmnn().
param := mlpack.MlpackLmnnOptions()
param.Labels = iris_labels
param.K = 3
param.Optimizer = "bbsgd"

_, output, _ := mlpack.MlpackLmnn(iris, param)

An another program call making use of range & regularization parameter with dataset having labels as last column can be made as:

// Initialize optional parameters for MlpackLmnn().
param := mlpack.MlpackLmnnOptions()
param.K = 5
param.Range = 10
param.Regularization = 0.4

_, output, _ := mlpack.MlpackLmnn(letter_recognition, param)

See also

LocalCoordinateCoding()

Local Coordinate Coding

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for LocalCoordinateCoding().
param := mlpack.LocalCoordinateCodingOptions()
param.Atoms = 0
param.InitialDictionary = mat.NewDense(1, 1, nil)
param.InputModel = nil
param.Lambda = 0
param.MaxIterations = 0
param.Normalize = false
param.Seed = 0
param.Test = mat.NewDense(1, 1, nil)
param.Tolerance = 0.01
param.Training = mat.NewDense(1, 1, nil)

codes, dictionary, output_model := mlpack.LocalCoordinateCoding(param)

An implementation of Local Coordinate Coding (LCC), a data transformation technique. Given input data, this transforms each point to be expressed as a linear combination of a few points in the dataset; once an LCC model is trained, it can be used to transform points later also. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Atoms int Number of atoms in the dictionary. 0
InitialDictionary *mat.Dense Optional initial dictionary. mat.NewDense(1, 1, nil)
InputModel localCoordinateCoding Input LCC model. nil
Lambda float64 Weighted l1-norm regularization parameter. 0
MaxIterations int Maximum number of iterations for LCC (0 indicates no limit). 0
Normalize bool If set, the input data matrix will be normalized before coding. false
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
Test *mat.Dense Test points to encode. mat.NewDense(1, 1, nil)
Tolerance float64 Tolerance for objective function. 0.01
Training *mat.Dense Matrix of training data (X). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
codes *mat.Dense Output codes matrix.
dictionary *mat.Dense Output dictionary matrix.
outputModel localCoordinateCoding Output for trained LCC model.

Detailed documentation

An implementation of Local Coordinate Coding (LCC), which codes data that approximately lives on a manifold using a variation of l1-norm regularized sparse coding. Given a dense data matrix X with n points and d dimensions, LCC seeks to find a dense dictionary matrix D with k atoms in d dimensions, and a coding matrix Z with n points in k dimensions. Because of the regularization method used, the atoms in D should lie close to the manifold on which the data points lie.

The original data matrix X can then be reconstructed as D * Z. Therefore, this program finds a representation of each point in X as a sparse linear combination of atoms in the dictionary D.

The coding is found with an algorithm which alternates between a dictionary step, which updates the dictionary D, and a coding step, which updates the coding matrix Z.

To run this program, the input matrix X must be specified (with -i), along with the number of atoms in the dictionary (-k). An initial dictionary may also be specified with the InitialDictionary parameter. The l1-norm regularization parameter is specified with the Lambda parameter.

Example

For example, to run LCC on the dataset data using 200 atoms and an l1-regularization parameter of 0.1, saving the dictionary Dictionary and the codes into Codes, use

// Initialize optional parameters for LocalCoordinateCoding().
param := mlpack.LocalCoordinateCodingOptions()
param.Training = data
param.Atoms = 200
param.Lambda = 0.1

codes, dict, _ := mlpack.LocalCoordinateCoding(param)

The maximum number of iterations may be specified with the MaxIterations parameter. Optionally, the input data matrix X can be normalized before coding with the Normalize parameter.

An LCC model may be saved using the OutputModel output parameter. Then, to encode new points from the dataset points with the previously saved model lcc_model, saving the new codes to new_codes, the following command can be used:

// Initialize optional parameters for LocalCoordinateCoding().
param := mlpack.LocalCoordinateCodingOptions()
param.InputModel = &lcc_model
param.Test = points

new_codes, _, _ := mlpack.LocalCoordinateCoding(param)

See also

LogisticRegression()

L2-regularized Logistic Regression and Prediction

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for LogisticRegression().
param := mlpack.LogisticRegressionOptions()
param.BatchSize = 64
param.DecisionBoundary = 0.5
param.InputModel = nil
param.Labels = mat.NewDense(1, 1, nil)
param.Lambda = 0
param.MaxIterations = 10000
param.Optimizer = "lbfgs"
param.StepSize = 0.01
param.Test = mat.NewDense(1, 1, nil)
param.Tolerance = 1e-10
param.Training = mat.NewDense(1, 1, nil)

output, output_model, output_probabilities, predictions, probabilities :=
    mlpack.LogisticRegression(param)

An implementation of L2-regularized logistic regression for two-class classification. Given labeled data, a model can be trained and saved for future use; or, a pre-trained model can be used to classify new points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
BatchSize int Batch size for SGD. 64
DecisionBoundary float64 Decision boundary for prediction; if the logistic function for a point is less than the boundary, the class is taken to be 0; otherwise, the class is 1. 0.5
InputModel logisticRegression Existing model (parameters). nil
Labels *mat.Dense (1d with ints) A matrix containing labels (0 or 1) for the points in the training set (y). mat.NewDense(1, 1, nil)
Lambda float64 L2-regularization parameter for training. 0
MaxIterations int Maximum iterations for optimizer (0 indicates no limit). 10000
Optimizer string Optimizer to use for training (‘lbfgs’ or ‘sgd’). "lbfgs"
StepSize float64 Step size for SGD optimizer. 0.01
Test *mat.Dense Matrix containing test dataset. mat.NewDense(1, 1, nil)
Tolerance float64 Convergence tolerance for optimizer. 1e-10
Training *mat.Dense A matrix containing the training set (the matrix of predictors, X). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense (1d with ints) If test data is specified, this matrix is where the predictions for the test set will be saved.
outputModel logisticRegression Output for trained logistic regression model.
outputProbabilities *mat.Dense If test data is specified, this matrix is where the class probabilities for the test set will be saved.
predictions *mat.Dense (1d with ints) If test data is specified, this matrix is where the predictions for the test set will be saved.
probabilities *mat.Dense If test data is specified, this matrix is where the class probabilities for the test set will be saved.

Detailed documentation

An implementation of L2-regularized logistic regression using either the L-BFGS optimizer or SGD (stochastic gradient descent). This solves the regression problem

y = (1 / 1 + e^-(X * b))

where y takes values 0 or 1.

This program allows loading a logistic regression model (via the InputModel parameter) or training a logistic regression model given training data (specified with the Training parameter), or both those things at once. In addition, this program allows classification on a test dataset (specified with the Test parameter) and the classification results may be saved with the Predictions output parameter. The trained logistic regression model may be saved using the OutputModel output parameter.

The training data, if specified, may have class labels as its last dimension. Alternately, the Labels parameter may be used to specify a separate matrix of labels.

When a model is being trained, there are many options. L2 regularization (to prevent overfitting) can be specified with the Lambda option, and the optimizer used to train the model can be specified with the Optimizer parameter. Available options are ‘sgd’ (stochastic gradient descent) and ‘lbfgs’ (the L-BFGS optimizer). There are also various parameters for the optimizer; the MaxIterations parameter specifies the maximum number of allowed iterations, and the Tolerance parameter specifies the tolerance for convergence. For the SGD optimizer, the StepSize parameter controls the step size taken at each iteration by the optimizer. The batch size for SGD is controlled with the BatchSize parameter. If the objective function for your data is oscillating between Inf and 0, the step size is probably too large. There are more parameters for the optimizers, but the C++ interface must be used to access these.

For SGD, an iteration refers to a single point. So to take a single pass over the dataset with SGD, MaxIterations should be set to the number of points in the dataset.

Optionally, the model can be used to predict the responses for another matrix of data points, if Test is specified. The Test parameter can be specified without the Training parameter, so long as an existing logistic regression model is given with the InputModel parameter. The output predictions from the logistic regression model may be saved with the Predictions parameter.

Note : The following parameters are deprecated and will be removed in mlpack 4: Output, OutputProbabilities Use Predictions instead of Output Use Probabilities instead of OutputProbabilities

This implementation of logistic regression does not support the general multi-class case but instead only the two-class case. Any labels must be either 0 or 1. For more classes, see the softmax_regression program.

Example

As an example, to train a logistic regression model on the data ‘data’ with labels ‘labels’ with L2 regularization of 0.1, saving the model to ‘lr_model’, the following command may be used:

// Initialize optional parameters for LogisticRegression().
param := mlpack.LogisticRegressionOptions()
param.Training = data
param.Labels = labels
param.Lambda = 0.1

_, lr_model, _, _, _ := mlpack.LogisticRegression(param)

Then, to use that model to predict classes for the dataset ‘test’, storing the output predictions in ‘predictions’, the following command may be used:

// Initialize optional parameters for LogisticRegression().
param := mlpack.LogisticRegressionOptions()
param.InputModel = &lr_model
param.Test = test

predictions, _, _, _, _ := mlpack.LogisticRegression(param)

See also

Lsh()

K-Approximate-Nearest-Neighbor Search with LSH

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Lsh().
param := mlpack.LshOptions()
param.BucketSize = 500
param.HashWidth = 0
param.InputModel = nil
param.K = 0
param.NumProbes = 0
param.Projections = 10
param.Query = mat.NewDense(1, 1, nil)
param.Reference = mat.NewDense(1, 1, nil)
param.SecondHashSize = 99901
param.Seed = 0
param.Tables = 30
param.TrueNeighbors = mat.NewDense(1, 1, nil)

distances, neighbors, output_model := mlpack.Lsh(param)

An implementation of approximate k-nearest-neighbor search with locality-sensitive hashing (LSH). Given a set of reference points and a set of query points, this will compute the k approximate nearest neighbors of each query point in the reference set; models can be saved for future use. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
BucketSize int The size of a bucket in the second level hash. 500
HashWidth float64 The hash width for the first-level hashing in the LSH preprocessing. By default, the LSH class automatically estimates a hash width for its use. 0
InputModel lshSearch Input LSH model. nil
K int Number of nearest neighbors to find. 0
NumProbes int Number of additional probes for multiprobe LSH; if 0, traditional LSH is used. 0
Projections int The number of hash functions for each table 10
Query *mat.Dense Matrix containing query points (optional). mat.NewDense(1, 1, nil)
Reference *mat.Dense Matrix containing the reference dataset. mat.NewDense(1, 1, nil)
SecondHashSize int The size of the second level hash table. 99901
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
Tables int The number of hash tables to be used. 30
TrueNeighbors *mat.Dense (with ints) Matrix of true neighbors to compute recall with (the recall is printed when -v is specified). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
distances *mat.Dense Matrix to output distances into.
neighbors *mat.Dense (with ints) Matrix to output neighbors into.
outputModel lshSearch Output for trained LSH model.

Detailed documentation

This program will calculate the k approximate-nearest-neighbors of a set of points using locality-sensitive hashing. You may specify a separate set of reference points and query points, or just a reference set which will be used as both the reference and query set.

Example

For example, the following will return 5 neighbors from the data for each point in input and store the distances in distances and the neighbors in neighbors:

// Initialize optional parameters for Lsh().
param := mlpack.LshOptions()
param.K = 5
param.Reference = input

distances, neighbors, _ := mlpack.Lsh(param)

The output is organized such that row i and column j in the neighbors output corresponds to the index of the point in the reference set which is the j’th nearest neighbor from the point in the query set with index i. Row j and column i in the distances output file corresponds to the distance between those two points.

Because this is approximate-nearest-neighbors search, results may be different from run to run. Thus, the Seed parameter can be specified to set the random seed.

This program also has many other parameters to control its functionality; see the parameter-specific documentation for more information.

See also

MeanShift()

Mean Shift Clustering

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for MeanShift().
param := mlpack.MeanShiftOptions()
param.ForceConvergence = false
param.InPlace = false
param.LabelsOnly = false
param.MaxIterations = 1000
param.Radius = 0

centroid, output := mlpack.MeanShift(input, param)

A fast implementation of mean-shift clustering using dual-tree range search. Given a dataset, this uses the mean shift algorithm to produce and return a clustering of the data. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
ForceConvergence bool If specified, the mean shift algorithm will continue running regardless of max_iterations until the clusters converge. false
InPlace bool If specified, a column containing the learned cluster assignments will be added to the input dataset file. In this case, –output_file is overridden. (Do not use with Python.) false
input *mat.Dense Input dataset to perform clustering on. required
LabelsOnly bool If specified, only the output labels will be written to the file specified by –output_file. false
MaxIterations int Maximum number of iterations before mean shift terminates. 1000
Radius float64 If the distance between two centroids is less than the given radius, one will be removed. A radius of 0 or less means an estimate will be calculated and used for the radius. 0
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
centroid *mat.Dense If specified, the centroids of each cluster will be written to the given matrix.
output *mat.Dense Matrix to write output labels or labeled data to.

Detailed documentation

This program performs mean shift clustering on the given dataset, storing the learned cluster assignments either as a column of labels in the input dataset or separately.

The input dataset should be specified with the Input parameter, and the radius used for search can be specified with the Radius parameter. The maximum number of iterations before algorithm termination is controlled with the MaxIterations parameter.

The output labels may be saved with the Output output parameter and the centroids of each cluster may be saved with the Centroid output parameter.

Example

For example, to run mean shift clustering on the dataset data and store the centroids to centroids, the following command may be used:

// Initialize optional parameters for MeanShift().
param := mlpack.MeanShiftOptions()

centroids, _ := mlpack.MeanShift(data, param)

See also

Nbc()

Parametric Naive Bayes Classifier

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Nbc().
param := mlpack.NbcOptions()
param.IncrementalVariance = false
param.InputModel = nil
param.Labels = mat.NewDense(1, 1, nil)
param.Test = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)

output, output_model, output_probs, predictions, probabilities :=
    mlpack.Nbc(param)

An implementation of the Naive Bayes Classifier, used for classification. Given labeled data, an NBC model can be trained and saved, or, a pre-trained model can be used for classification. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
IncrementalVariance bool The variance of each class will be calculated incrementally. false
InputModel nbcModel Input Naive Bayes model. nil
Labels *mat.Dense (1d with ints) A file containing labels for the training set. mat.NewDense(1, 1, nil)
Test *mat.Dense A matrix containing the test set. mat.NewDense(1, 1, nil)
Training *mat.Dense A matrix containing the training set. mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense (1d with ints) The matrix in which the predicted labels for the test set will be written (deprecated).
outputModel nbcModel File to save trained Naive Bayes model to.
outputProbs *mat.Dense The matrix in which the predicted probability of labels for the test set will be written (deprecated).
predictions *mat.Dense (1d with ints) The matrix in which the predicted labels for the test set will be written.
probabilities *mat.Dense The matrix in which the predicted probability of labels for the test set will be written.

Detailed documentation

This program trains the Naive Bayes classifier on the given labeled training set, or loads a model from the given model file, and then may use that trained model to classify the points in a given test set.

The training set is specified with the Training parameter. Labels may be either the last row of the training set, or alternately the Labels parameter may be specified to pass a separate matrix of labels.

If training is not desired, a pre-existing model may be loaded with the InputModel parameter.

The IncrementalVariance parameter can be used to force the training to use an incremental algorithm for calculating variance. This is slower, but can help avoid loss of precision in some cases.

If classifying a test set is desired, the test set may be specified with the Test parameter, and the classifications may be saved with the Predictionspredictions parameter. If saving the trained model is desired, this may be done with the OutputModel output parameter.

Note: the Output and OutputProbs parameters are deprecated and will be removed in mlpack 4.0.0. Use Predictions and Probabilities instead.

Example

For example, to train a Naive Bayes classifier on the dataset data with labels labels and save the model to nbc_model, the following command may be used:

// Initialize optional parameters for Nbc().
param := mlpack.NbcOptions()
param.Training = data
param.Labels = labels

_, nbc_model, _, _, _ := mlpack.Nbc(param)

Then, to use nbc_model to predict the classes of the dataset test_set and save the predicted classes to predictions, the following command may be used:

// Initialize optional parameters for Nbc().
param := mlpack.NbcOptions()
param.InputModel = &nbc_model
param.Test = test_set

predictions, _, _, _, _ := mlpack.Nbc(param)

See also

Nca()

Neighborhood Components Analysis (NCA)

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Nca().
param := mlpack.NcaOptions()
param.ArmijoConstant = 0.0001
param.BatchSize = 50
param.Labels = mat.NewDense(1, 1, nil)
param.LinearScan = false
param.MaxIterations = 500000
param.MaxLineSearchTrials = 50
param.MaxStep = 1e+20
param.MinStep = 1e-20
param.Normalize = false
param.NumBasis = 5
param.Optimizer = "sgd"
param.Seed = 0
param.StepSize = 0.01
param.Tolerance = 1e-07
param.Wolfe = 0.9

output := mlpack.Nca(input, param)

An implementation of neighborhood components analysis, a distance learning technique that can be used for preprocessing. Given a labeled dataset, this uses NCA, which seeks to improve the k-nearest-neighbor classification, and returns the learned distance metric. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
ArmijoConstant float64 Armijo constant for L-BFGS. 0.0001
BatchSize int Batch size for mini-batch SGD. 50
input *mat.Dense Input dataset to run NCA on. required
Labels *mat.Dense (1d with ints) Labels for input dataset. mat.NewDense(1, 1, nil)
LinearScan bool Don’t shuffle the order in which data points are visited for SGD or mini-batch SGD. false
MaxIterations int Maximum number of iterations for SGD or L-BFGS (0 indicates no limit). 500000
MaxLineSearchTrials int Maximum number of line search trials for L-BFGS. 50
MaxStep float64 Maximum step of line search for L-BFGS. 1e+20
MinStep float64 Minimum step of line search for L-BFGS. 1e-20
Normalize bool Use a normalized starting point for optimization. This is useful for when points are far apart, or when SGD is returning NaN. false
NumBasis int Number of memory points to be stored for L-BFGS. 5
Optimizer string Optimizer to use; ‘sgd’ or ‘lbfgs’. "sgd"
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
StepSize float64 Step size for stochastic gradient descent (alpha). 0.01
Tolerance float64 Maximum tolerance for termination of SGD or L-BFGS. 1e-07
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false
Wolfe float64 Wolfe condition parameter for L-BFGS. 0.9

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Output matrix for learned distance matrix.

Detailed documentation

This program implements Neighborhood Components Analysis, both a linear dimensionality reduction technique and a distance learning technique. The method seeks to improve k-nearest-neighbor classification on a dataset by scaling the dimensions. The method is nonparametric, and does not require a value of k. It works by using stochastic (“soft”) neighbor assignments and using optimization techniques over the gradient of the accuracy of the neighbor assignments.

To work, this algorithm needs labeled data. It can be given as the last row of the input dataset (specified with Input), or alternatively as a separate matrix (specified with Labels).

This implementation of NCA uses stochastic gradient descent, mini-batch stochastic gradient descent, or the L_BFGS optimizer. These optimizers do not guarantee global convergence for a nonconvex objective function (NCA’s objective function is nonconvex), so the final results could depend on the random seed or other optimizer parameters.

Stochastic gradient descent, specified by the value ‘sgd’ for the parameter Optimizer, depends primarily on three parameters: the step size (specified with StepSize), the batch size (specified with BatchSize), and the maximum number of iterations (specified with MaxIterations). In addition, a normalized starting point can be used by specifying the Normalize parameter, which is necessary if many warnings of the form ‘Denominator of p_i is 0!’ are given. Tuning the step size can be a tedious affair. In general, the step size is too large if the objective is not mostly uniformly decreasing, or if zero-valued denominator warnings are being issued. The step size is too small if the objective is changing very slowly. Setting the termination condition can be done easily once a good step size parameter is found; either increase the maximum iterations to a large number and allow SGD to find a minimum, or set the maximum iterations to 0 (allowing infinite iterations) and set the tolerance (specified by Tolerance) to define the maximum allowed difference between objectives for SGD to terminate. Be careful—setting the tolerance instead of the maximum iterations can take a very long time and may actually never converge due to the properties of the SGD optimizer. Note that a single iteration of SGD refers to a single point, so to take a single pass over the dataset, set the value of the MaxIterations parameter equal to the number of points in the dataset.

The L-BFGS optimizer, specified by the value ‘lbfgs’ for the parameter Optimizer, uses a back-tracking line search algorithm to minimize a function. The following parameters are used by L-BFGS: NumBasis (specifies the number of memory points used by L-BFGS), MaxIterations, ArmijoConstant, Wolfe, Tolerance (the optimization is terminated when the gradient norm is below this value), MaxLineSearchTrials, MinStep, and MaxStep (which both refer to the line search routine). For more details on the L-BFGS optimizer, consult either the mlpack L-BFGS documentation (in lbfgs.hpp) or the vast set of published literature on L-BFGS.

By default, the SGD optimizer is used.

See also

Knn()

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Knn().
param := mlpack.KnnOptions()
param.Algorithm = "dual_tree"
param.Epsilon = 0
param.InputModel = nil
param.K = 0
param.LeafSize = 20
param.Query = mat.NewDense(1, 1, nil)
param.RandomBasis = false
param.Reference = mat.NewDense(1, 1, nil)
param.Rho = 0.7
param.Seed = 0
param.Tau = 0
param.TreeType = "kd"
param.TrueDistances = mat.NewDense(1, 1, nil)
param.TrueNeighbors = mat.NewDense(1, 1, nil)

distances, neighbors, output_model := mlpack.Knn(param)

An implementation of k-nearest-neighbor search using single-tree and dual-tree algorithms. Given a set of reference points and query points, this can find the k nearest neighbors in the reference set of each query point using trees; trees that are built can be saved for future use. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Algorithm string Type of neighbor search: ‘naive’, ‘single_tree’, ‘dual_tree’, ‘greedy’. "dual_tree"
Epsilon float64 If specified, will do approximate nearest neighbor search with given relative error. 0
InputModel knnModel Pre-trained kNN model. nil
K int Number of nearest neighbors to find. 0
LeafSize int Leaf size for tree building (used for kd-trees, vp trees, random projection trees, UB trees, R trees, R* trees, X trees, Hilbert R trees, R+ trees, R++ trees, spill trees, and octrees). 20
Query *mat.Dense Matrix containing query points (optional). mat.NewDense(1, 1, nil)
RandomBasis bool Before tree-building, project the data onto a random orthogonal basis. false
Reference *mat.Dense Matrix containing the reference dataset. mat.NewDense(1, 1, nil)
Rho float64 Balance threshold (only valid for spill trees). 0.7
Seed int Random seed (if 0, std::time(NULL) is used). 0
Tau float64 Overlapping size (only valid for spill trees). 0
TreeType string Type of tree to use: ‘kd’, ‘vp’, ‘rp’, ‘max-rp’, ‘ub’, ‘cover’, ‘r’, ‘r-star’, ‘x’, ‘ball’, ‘hilbert-r’, ‘r-plus’, ‘r-plus-plus’, ‘spill’, ‘oct’. "kd"
TrueDistances *mat.Dense Matrix of true distances to compute the effective error (average relative error) (it is printed when -v is specified). mat.NewDense(1, 1, nil)
TrueNeighbors *mat.Dense (with ints) Matrix of true neighbors to compute the recall (it is printed when -v is specified). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
distances *mat.Dense Matrix to output distances into.
neighbors *mat.Dense (with ints) Matrix to output neighbors into.
outputModel knnModel If specified, the kNN model will be output here.

Detailed documentation

This program will calculate the k-nearest-neighbors of a set of points using kd-trees or cover trees (cover tree support is experimental and may be slow). You may specify a separate set of reference points and query points, or just a reference set which will be used as both the reference and query set.

Example

For example, the following command will calculate the 5 nearest neighbors of each point in input and store the distances in distances and the neighbors in neighbors:

// Initialize optional parameters for Knn().
param := mlpack.KnnOptions()
param.K = 5
param.Reference = input

distances, neighbors, _ := mlpack.Knn(param)

The output is organized such that row i and column j in the neighbors output matrix corresponds to the index of the point in the reference set which is the j’th nearest neighbor from the point in the query set with index i. Row j and column i in the distances output matrix corresponds to the distance between those two points.

See also

Kfn()

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Kfn().
param := mlpack.KfnOptions()
param.Algorithm = "dual_tree"
param.Epsilon = 0
param.InputModel = nil
param.K = 0
param.LeafSize = 20
param.Percentage = 1
param.Query = mat.NewDense(1, 1, nil)
param.RandomBasis = false
param.Reference = mat.NewDense(1, 1, nil)
param.Seed = 0
param.TreeType = "kd"
param.TrueDistances = mat.NewDense(1, 1, nil)
param.TrueNeighbors = mat.NewDense(1, 1, nil)

distances, neighbors, output_model := mlpack.Kfn(param)

An implementation of k-furthest-neighbor search using single-tree and dual-tree algorithms. Given a set of reference points and query points, this can find the k furthest neighbors in the reference set of each query point using trees; trees that are built can be saved for future use. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Algorithm string Type of neighbor search: ‘naive’, ‘single_tree’, ‘dual_tree’, ‘greedy’. "dual_tree"
Epsilon float64 If specified, will do approximate furthest neighbor search with given relative error. Must be in the range [0,1). 0
InputModel kfnModel Pre-trained kFN model. nil
K int Number of furthest neighbors to find. 0
LeafSize int Leaf size for tree building (used for kd-trees, vp trees, random projection trees, UB trees, R trees, R* trees, X trees, Hilbert R trees, R+ trees, R++ trees, and octrees). 20
Percentage float64 If specified, will do approximate furthest neighbor search. Must be in the range (0,1] (decimal form). Resultant neighbors will be at least (p*100) % of the distance as the true furthest neighbor. 1
Query *mat.Dense Matrix containing query points (optional). mat.NewDense(1, 1, nil)
RandomBasis bool Before tree-building, project the data onto a random orthogonal basis. false
Reference *mat.Dense Matrix containing the reference dataset. mat.NewDense(1, 1, nil)
Seed int Random seed (if 0, std::time(NULL) is used). 0
TreeType string Type of tree to use: ‘kd’, ‘vp’, ‘rp’, ‘max-rp’, ‘ub’, ‘cover’, ‘r’, ‘r-star’, ‘x’, ‘ball’, ‘hilbert-r’, ‘r-plus’, ‘r-plus-plus’, ‘oct’. "kd"
TrueDistances *mat.Dense Matrix of true distances to compute the effective error (average relative error) (it is printed when -v is specified). mat.NewDense(1, 1, nil)
TrueNeighbors *mat.Dense (with ints) Matrix of true neighbors to compute the recall (it is printed when -v is specified). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
distances *mat.Dense Matrix to output distances into.
neighbors *mat.Dense (with ints) Matrix to output neighbors into.
outputModel kfnModel If specified, the kFN model will be output here.

Detailed documentation

This program will calculate the k-furthest-neighbors of a set of points. You may specify a separate set of reference points and query points, or just a reference set which will be used as both the reference and query set.

Example

For example, the following will calculate the 5 furthest neighbors of eachpoint in input and store the distances in distances and the neighbors in neighbors:

// Initialize optional parameters for Kfn().
param := mlpack.KfnOptions()
param.K = 5
param.Reference = input

distances, neighbors, _ := mlpack.Kfn(param)

The output files are organized such that row i and column j in the neighbors output matrix corresponds to the index of the point in the reference set which is the j’th furthest neighbor from the point in the query set with index i. Row i and column j in the distances output file corresponds to the distance between those two points.

See also

Nmf()

Non-negative Matrix Factorization

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Nmf().
param := mlpack.NmfOptions()
param.InitialH = mat.NewDense(1, 1, nil)
param.InitialW = mat.NewDense(1, 1, nil)
param.MaxIterations = 10000
param.MinResidue = 1e-05
param.Seed = 0
param.UpdateRules = "multdist"

h, w := mlpack.Nmf(input, rank, param)

An implementation of non-negative matrix factorization. This can be used to decompose an input dataset into two low-rank non-negative components. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
InitialH *mat.Dense Initial H matrix. mat.NewDense(1, 1, nil)
InitialW *mat.Dense Initial W matrix. mat.NewDense(1, 1, nil)
input *mat.Dense Input dataset to perform NMF on. required
MaxIterations int Number of iterations before NMF terminates (0 runs until convergence. 10000
MinResidue float64 The minimum root mean square residue allowed for each iteration, below which the program terminates. 1e-05
rank int Rank of the factorization. required
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
UpdateRules string Update rules for each iteration; ( multdist | multdiv | als ). "multdist"
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
h *mat.Dense Matrix to save the calculated H to.
w *mat.Dense Matrix to save the calculated W to.

Detailed documentation

This program performs non-negative matrix factorization on the given dataset, storing the resulting decomposed matrices in the specified files. For an input dataset V, NMF decomposes V into two matrices W and H such that

V = W * H

where all elements in W and H are non-negative. If V is of size (n x m), then W will be of size (n x r) and H will be of size (r x m), where r is the rank of the factorization (specified by the Rank parameter).

Optionally, the desired update rules for each NMF iteration can be chosen from the following list:

  • multdist: multiplicative distance-based update rules (Lee and Seung 1999)
  • multdiv: multiplicative divergence-based update rules (Lee and Seung 1999)
  • als: alternating least squares update rules (Paatero and Tapper 1994)

The maximum number of iterations is specified with MaxIterations, and the minimum residue required for algorithm termination is specified with the MinResidue parameter.

Example

For example, to run NMF on the input matrix V using the ‘multdist’ update rules with a rank-10 decomposition and storing the decomposed matrices into W and H, the following command could be used:

// Initialize optional parameters for Nmf().
param := mlpack.NmfOptions()
param.UpdateRules = "multdist"

H, W := mlpack.Nmf(V, 10, param)

See also

Pca()

Principal Components Analysis

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Pca().
param := mlpack.PcaOptions()
param.DecompositionMethod = "exact"
param.NewDimensionality = 0
param.Scale = false
param.VarToRetain = 0

output := mlpack.Pca(input, param)

An implementation of several strategies for principal components analysis (PCA), a common preprocessing step. Given a dataset and a desired new dimensionality, this can reduce the dimensionality of the data using the linear transformation determined by PCA. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
DecompositionMethod string Method used for the principal components analysis: ‘exact’, ‘randomized’, ‘randomized-block-krylov’, ‘quic’. "exact"
input *mat.Dense Input dataset to perform PCA on. required
NewDimensionality int Desired dimensionality of output dataset. If 0, no dimensionality reduction is performed. 0
Scale bool If set, the data will be scaled before running PCA, such that the variance of each feature is 1. false
VarToRetain float64 Amount of variance to retain; should be between 0 and 1. If 1, all variance is retained. Overrides -d. 0
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Matrix to save modified dataset to.

Detailed documentation

This program performs principal components analysis on the given dataset using the exact, randomized, randomized block Krylov, or QUIC SVD method. It will transform the data onto its principal components, optionally performing dimensionality reduction by ignoring the principal components with the smallest eigenvalues.

Use the Input parameter to specify the dataset to perform PCA on. A desired new dimensionality can be specified with the NewDimensionality parameter, or the desired variance to retain can be specified with the VarToRetain parameter. If desired, the dataset can be scaled before running PCA with the Scale parameter.

Multiple different decomposition techniques can be used. The method to use can be specified with the DecompositionMethod parameter, and it may take the values ‘exact’, ‘randomized’, or ‘quic’.

Example

For example, to reduce the dimensionality of the matrix data to 5 dimensions using randomized SVD for the decomposition, storing the output matrix to data_mod, the following command can be used:

// Initialize optional parameters for Pca().
param := mlpack.PcaOptions()
param.NewDimensionality = 5
param.DecompositionMethod = "randomized"

data_mod := mlpack.Pca(data, param)

See also

Perceptron()

Perceptron

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Perceptron().
param := mlpack.PerceptronOptions()
param.InputModel = nil
param.Labels = mat.NewDense(1, 1, nil)
param.MaxIterations = 1000
param.Test = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)

output, output_model, predictions := mlpack.Perceptron(param)

An implementation of a perceptron—a single level neural network–=for classification. Given labeled data, a perceptron can be trained and saved for future use; or, a pre-trained perceptron can be used for classification on new points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
InputModel perceptronModel Input perceptron model. nil
Labels *mat.Dense (1d with ints) A matrix containing labels for the training set. mat.NewDense(1, 1, nil)
MaxIterations int The maximum number of iterations the perceptron is to be run 1000
Test *mat.Dense A matrix containing the test set. mat.NewDense(1, 1, nil)
Training *mat.Dense A matrix containing the training set. mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense (1d with ints) The matrix in which the predicted labels for the test set will be written.
outputModel perceptronModel Output for trained perceptron model.
predictions *mat.Dense (1d with ints) The matrix in which the predicted labels for the test set will be written.

Detailed documentation

This program implements a perceptron, which is a single level neural network. The perceptron makes its predictions based on a linear predictor function combining a set of weights with the feature vector. The perceptron learning rule is able to converge, given enough iterations (specified using the MaxIterations parameter), if the data supplied is linearly separable. The perceptron is parameterized by a matrix of weight vectors that denote the numerical weights of the neural network.

This program allows loading a perceptron from a model (via the InputModel parameter) or training a perceptron given training data (via the Training parameter), or both those things at once. In addition, this program allows classification on a test dataset (via the Test parameter) and the classification results on the test set may be saved with the Predictions output parameter. The perceptron model may be saved with the OutputModel output parameter.

Note: the following parameter is deprecated and will be removed in mlpack 4.0.0: Output. Use Predictions instead of Output.

Example

The training data given with the Training option may have class labels as its last dimension (so, if the training data is in CSV format, labels should be the last column). Alternately, the Labels parameter may be used to specify a separate matrix of labels.

All these options make it easy to train a perceptron, and then re-use that perceptron for later classification. The invocation below trains a perceptron on training_data with labels training_labels, and saves the model to perceptron_model.

// Initialize optional parameters for Perceptron().
param := mlpack.PerceptronOptions()
param.Training = training_data
param.Labels = training_labels

_, perceptron_model, _ := mlpack.Perceptron(param)

Then, this model can be re-used for classification on the test data test_data. The example below does precisely that, saving the predicted classes to predictions.

// Initialize optional parameters for Perceptron().
param := mlpack.PerceptronOptions()
param.InputModel = &perceptron_model
param.Test = test_data

_, _, predictions := mlpack.Perceptron(param)

Note that all of the options may be specified at once: predictions may be calculated right after training a model, and model training can occur even if an existing perceptron model is passed with the InputModel parameter. However, note that the number of classes and the dimensionality of all data must match. So you cannot pass a perceptron model trained on 2 classes and then re-train with a 4-class dataset. Similarly, attempting classification on a 3-dimensional dataset with a perceptron that has been trained on 8 dimensions will cause an error.

See also

PreprocessSplit()

Split Data

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for PreprocessSplit().
param := mlpack.PreprocessSplitOptions()
param.InputLabels = mat.NewDense(1, 1, nil)
param.NoShuffle = false
param.Seed = 0
param.TestRatio = 0.2

test, test_labels, training, training_labels :=
    mlpack.PreprocessSplit(input, param)

A utility to split data into a training and testing dataset. This can also split labels according to the same split. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
input *mat.Dense Matrix containing data. required
InputLabels *mat.Dense (with ints) Matrix containing labels. mat.NewDense(1, 1, nil)
NoShuffle bool Avoid shuffling and splitting the data. false
Seed int Random seed (0 for std::time(NULL)). 0
TestRatio float64 Ratio of test set; if not set,the ratio defaults to 0.2 0.2
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
test *mat.Dense Matrix to save test data to.
testLabels *mat.Dense (with ints) Matrix to save test labels to.
training *mat.Dense Matrix to save training data to.
trainingLabels *mat.Dense (with ints) Matrix to save train labels to.

Detailed documentation

This utility takes a dataset and optionally labels and splits them into a training set and a test set. Before the split, the points in the dataset are randomly reordered. The percentage of the dataset to be used as the test set can be specified with the TestRatio parameter; the default is 0.2 (20%).

The output training and test matrices may be saved with the Training and Test output parameters.

Optionally, labels can be also be split along with the data by specifying the InputLabels parameter. Splitting labels works the same way as splitting the data. The output training and test labels may be saved with the TrainingLabels and TestLabels output parameters, respectively.

Example

So, a simple example where we want to split the dataset X into X_train and X_test with 60% of the data in the training set and 40% of the dataset in the test set, we could run

// Initialize optional parameters for PreprocessSplit().
param := mlpack.PreprocessSplitOptions()
param.TestRatio = 0.4

X_test, _, X_train, _ := mlpack.PreprocessSplit(X, param)

Also by default the dataset is shuffled and split; you can provide the NoShuffle option to avoid shuffling the data; an example to avoid shuffling of data is:

// Initialize optional parameters for PreprocessSplit().
param := mlpack.PreprocessSplitOptions()
param.TestRatio = 0.4
param.NoShuffle = true

X_test, _, X_train, _ := mlpack.PreprocessSplit(X, param)

If we had a dataset X and associated labels y, and we wanted to split these into X_train, y_train, X_test, and y_test, with 30% of the data in the test set, we could run

// Initialize optional parameters for PreprocessSplit().
param := mlpack.PreprocessSplitOptions()
param.InputLabels = y
param.TestRatio = 0.3

X_test, y_test, X_train, y_train := mlpack.PreprocessSplit(X, param)

See also

PreprocessBinarize()

Binarize Data

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for PreprocessBinarize().
param := mlpack.PreprocessBinarizeOptions()
param.Dimension = 0
param.Threshold = 0

output := mlpack.PreprocessBinarize(input, param)

A utility to binarize a dataset. Given a dataset, this utility converts each value in the desired dimension(s) to 0 or 1; this can be a useful preprocessing step. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Dimension int Dimension to apply the binarization. If not set, the program will binarize every dimension by default. 0
input *mat.Dense Input data matrix. required
Threshold float64 Threshold to be applied for binarization. If not set, the threshold defaults to 0.0. 0
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Matrix in which to save the output.

Detailed documentation

This utility takes a dataset and binarizes the variables into either 0 or 1 given threshold. User can apply binarization on a dimension or the whole dataset. The dimension to apply binarization to can be specified using the Dimension parameter; if left unspecified, every dimension will be binarized. The threshold for binarization can also be specified with the Threshold parameter; the default threshold is 0.0.

The binarized matrix may be saved with the Output output parameter.

Example

For example, if we want to set all variables greater than 5 in the dataset X to 1 and variables less than or equal to 5.0 to 0, and save the result to Y, we could run

// Initialize optional parameters for PreprocessBinarize().
param := mlpack.PreprocessBinarizeOptions()
param.Threshold = 5

Y := mlpack.PreprocessBinarize(X, param)

But if we want to apply this to only the first (0th) dimension of X, we could instead run

// Initialize optional parameters for PreprocessBinarize().
param := mlpack.PreprocessBinarizeOptions()
param.Threshold = 5
param.Dimension = 0

Y := mlpack.PreprocessBinarize(X, param)

See also

PreprocessDescribe()

Descriptive Statistics

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for PreprocessDescribe().
param := mlpack.PreprocessDescribeOptions()
param.Dimension = 0
param.Population = false
param.Precision = 4
param.RowMajor = false
param.Width = 8

 := mlpack.PreprocessDescribe(input, param)

A utility for printing descriptive statistics about a dataset. This prints a number of details about a dataset in a tabular format. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Dimension int Dimension of the data. Use this to specify a dimension 0
input *mat.Dense Matrix containing data, required
Population bool If specified, the program will calculate statistics assuming the dataset is the population. By default, the program will assume the dataset as a sample. false
Precision int Precision of the output statistics. 4
RowMajor bool If specified, the program will calculate statistics across rows, not across columns. (Remember that in mlpack, a column represents a point, so this option is generally not necessary.) false
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false
Width int Width of the output table. 8

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

| name | type | description | |————|————|——————-|

Detailed documentation

This utility takes a dataset and prints out the descriptive statistics of the data. Descriptive statistics is the discipline of quantitatively describing the main features of a collection of information, or the quantitative description itself. The program does not modify the original file, but instead prints out the statistics to the console. The printed result will look like a table.

Optionally, width and precision of the output can be adjusted by a user using the Width and Precision parameters. A user can also select a specific dimension to analyze if there are too many dimensions. The Population parameter can be specified when the dataset should be considered as a population. Otherwise, the dataset will be considered as a sample.

Example

So, a simple example where we want to print out statistical facts about the dataset X using the default settings, we could run

// Initialize optional parameters for PreprocessDescribe().
param := mlpack.PreprocessDescribeOptions()
param.Verbose = true

 := mlpack.PreprocessDescribe(X, param)

If we want to customize the width to 10 and precision to 5 and consider the dataset as a population, we could run

// Initialize optional parameters for PreprocessDescribe().
param := mlpack.PreprocessDescribeOptions()
param.Width = 10
param.Precision = 5
param.Verbose = true

 := mlpack.PreprocessDescribe(X, param)

See also

Impute Data

This utility provides several imputation strategies for missing data. Given a dataset with missing values, this can impute according to several strategies, including user-defined values. .

PreprocessScale()

Scale Data

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for PreprocessScale().
param := mlpack.PreprocessScaleOptions()
param.Epsilon = 1e-06
param.InputModel = nil
param.InverseScaling = false
param.MaxValue = 1
param.MinValue = 0
param.ScalerMethod = "standard_scaler"
param.Seed = 0

output, output_model := mlpack.PreprocessScale(input, param)

A utility to perform feature scaling on datasets using one of sixtechniques. Both scaling and inverse scaling are supported, andscalers can be saved and then applied to other datasets. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Epsilon float64 regularization Parameter for pcawhitening, or zcawhitening, should be between -1 to 1. 1e-06
input *mat.Dense Matrix containing data. required
InputModel scalingModel Input Scaling model. nil
InverseScaling bool Inverse Scaling to get original dataset false
MaxValue int Ending value of range for min_max_scaler. 1
MinValue int Starting value of range for min_max_scaler. 0
ScalerMethod string method to use for scaling, the default is standard_scaler. "standard_scaler"
Seed int Random seed (0 for std::time(NULL)). 0
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Matrix to save scaled data to.
outputModel scalingModel Output scaling model.

Detailed documentation

This utility takes a dataset and performs feature scaling using one of the six scaler methods namely: ‘max_abs_scaler’, ‘mean_normalization’, ‘min_max_scaler’ ,’standard_scaler’, ‘pca_whitening’ and ‘zca_whitening’. The function takes a matrix as Input and a scaling method type which you can specify using ScalerMethod parameter; the default is standard scaler, and outputs a matrix with scaled feature.

The output scaled feature matrix may be saved with the Output output parameters.

The model to scale features can be saved using OutputModel and later can be loaded back usingInputModel.

Example

So, a simple example where we want to scale the dataset X into X_scaled with standard_scaler as scaler_method, we could run

// Initialize optional parameters for PreprocessScale().
param := mlpack.PreprocessScaleOptions()
param.ScalerMethod = "standard_scaler"

X_scaled, _ := mlpack.PreprocessScale(X, param)

A simple example where we want to whiten the dataset X into X_whitened with PCA as whitening_method and use 0.01 as regularization parameter, we could run

// Initialize optional parameters for PreprocessScale().
param := mlpack.PreprocessScaleOptions()
param.ScalerMethod = "pca_whitening"
param.Epsilon = 0.01

X_scaled, _ := mlpack.PreprocessScale(X, param)

You can also retransform the scaled dataset back usingInverseScaling. An example to rescale : X_scaled into Xusing the saved model InputModel is:

// Initialize optional parameters for PreprocessScale().
param := mlpack.PreprocessScaleOptions()
param.InverseScaling = true
param.InputModel = &saved

X, _ := mlpack.PreprocessScale(X_scaled, param)

Another simple example where we want to scale the dataset X into X_scaled with min_max_scaler as scaler method, where scaling range is 1 to 3 instead of default 0 to 1. We could run

// Initialize optional parameters for PreprocessScale().
param := mlpack.PreprocessScaleOptions()
param.ScalerMethod = "min_max_scaler"
param.MinValue = 1
param.MaxValue = 3

X_scaled, _ := mlpack.PreprocessScale(X, param)

See also

PreprocessOneHotEncoding()

One Hot Encoding

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for PreprocessOneHotEncoding().
param := mlpack.PreprocessOneHotEncodingOptions()

output := mlpack.PreprocessOneHotEncoding(dimensions, input, )

A utility to do one-hot encoding on features of dataset. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
dimensions array of ints Index of dimensions thatneed to be one-hot encoded. required
input *mat.Dense Matrix containing data. required
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Matrix to save one-hot encoded features data to.

Detailed documentation

This utility takes a dataset and a vector of indices and does one-hot encoding of the respective features at those indices. Indices represent the IDs of the dimensions to be one-hot encoded.

The output matrix with encoded features may be saved with the Output parameters.

Example

So, a simple example where we want to encode 1st and 3rd feature from dataset X into X_output would be

// Initialize optional parameters for PreprocessOneHotEncoding().
param := mlpack.PreprocessOneHotEncodingOptions()

X_ouput := mlpack.PreprocessOneHotEncoding(X, 1, 3, param)

See also

ImageConverter()

Image Converter

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for ImageConverter().
param := mlpack.ImageConverterOptions()
param.Channels = 0
param.Dataset = mat.NewDense(1, 1, nil)
param.Height = 0
param.Quality = 90
param.Save = false
param.Width = 0

output := mlpack.ImageConverter(input, param)

A utility to load an image or set of images into a single dataset that can then be used by other mlpack methods and utilities. This can also unpack an image dataset into individual files, for instance after mlpack methods have been used. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Channels int Number of channels in the image. 0
Dataset *mat.Dense Input matrix to save as images. mat.NewDense(1, 1, nil)
Height int Height of the images. 0
input array of strings Image filenames which have to be loaded/saved. required
Quality int Compression of the image if saved as jpg (0-100). 90
Save bool Save a dataset as images. false
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false
Width int Width of the image. 0

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
output *mat.Dense Matrix to save images data to, Onlyneeded if you are specifying ‘save’ option.

Detailed documentation

This utility takes an image or an array of images and loads them to a matrix. You can optionally specify the height Height width Width and channel Channels of the images that needs to be loaded; otherwise, these parameters will be automatically detected from the image. There are other options too, that can be specified such as Quality.

You can also provide a dataset and save them as images using Dataset and Save as an parameter.

Example

An example to load an image :

// Initialize optional parameters for ImageConverter().
param := mlpack.ImageConverterOptions()
param.Height = 256
param.Width = 256
param.Channels = 3

Y := mlpack.ImageConverter(X, param)

An example to save an image is :

// Initialize optional parameters for ImageConverter().
param := mlpack.ImageConverterOptions()
param.Height = 256
param.Width = 256
param.Channels = 3
param.Dataset = Y
param.Save = true

_ := mlpack.ImageConverter(X, param)

See also

Radical()

RADICAL

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Radical().
param := mlpack.RadicalOptions()
param.Angles = 150
param.NoiseStdDev = 0.175
param.Objective = false
param.Replicates = 30
param.Seed = 0
param.Sweeps = 0

output_ic, output_unmixing := mlpack.Radical(input, param)

An implementation of RADICAL, a method for independent component analysis (ICA). Given a dataset, this can decompose the dataset into an unmixing matrix and an independent component matrix; this can be useful for preprocessing. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Angles int Number of angles to consider in brute-force search during Radical2D. 150
input *mat.Dense Input dataset for ICA. required
NoiseStdDev float64 Standard deviation of Gaussian noise. 0.175
Objective bool If set, an estimate of the final objective function is printed. false
Replicates int Number of Gaussian-perturbed replicates to use (per point) in Radical2D. 30
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
Sweeps int Number of sweeps; each sweep calls Radical2D once for each pair of dimensions. 0
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputIc *mat.Dense Matrix to save independent components to.
outputUnmixing *mat.Dense Matrix to save unmixing matrix to.

Detailed documentation

An implementation of RADICAL, a method for independent component analysis (ICA). Assuming that we have an input matrix X, the goal is to find a square unmixing matrix W such that Y = W * X and the dimensions of Y are independent components. If the algorithm is running particularly slowly, try reducing the number of replicates.

The input matrix to perform ICA on should be specified with the Input parameter. The output matrix Y may be saved with the OutputIc output parameter, and the output unmixing matrix W may be saved with the OutputUnmixing output parameter.

Example

For example, to perform ICA on the matrix X with 40 replicates, saving the independent components to ic, the following command may be used:

// Initialize optional parameters for Radical().
param := mlpack.RadicalOptions()
param.Replicates = 40

ic, _ := mlpack.Radical(X, param)

See also

RandomForest()

Random forests

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for RandomForest().
param := mlpack.RandomForestOptions()
param.InputModel = nil
param.Labels = mat.NewDense(1, 1, nil)
param.MaximumDepth = 0
param.MinimumGainSplit = 0
param.MinimumLeafSize = 1
param.NumTrees = 10
param.PrintTrainingAccuracy = false
param.Seed = 0
param.SubspaceDim = 0
param.Test = mat.NewDense(1, 1, nil)
param.TestLabels = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)

output_model, predictions, probabilities := mlpack.RandomForest(param)

An implementation of the standard random forest algorithm by Leo Breiman for classification. Given labeled data, a random forest can be trained and saved for future use; or, a pre-trained random forest can be used for classification. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
InputModel randomForestModel Pre-trained random forest to use for classification. nil
Labels *mat.Dense (1d with ints) Labels for training dataset. mat.NewDense(1, 1, nil)
MaximumDepth int Maximum depth of the tree (0 means no limit). 0
MinimumGainSplit float64 Minimum gain needed to make a split when building a tree. 0
MinimumLeafSize int Minimum number of points in each leaf node. 1
NumTrees int Number of trees in the random forest. 10
PrintTrainingAccuracy bool If set, then the accuracy of the model on the training set will be predicted (verbose must also be specified). false
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
SubspaceDim int Dimensionality of random subspace to use for each split. ‘0’ will autoselect the square root of data dimensionality. 0
Test *mat.Dense Test dataset to produce predictions for. mat.NewDense(1, 1, nil)
TestLabels *mat.Dense (1d with ints) Test dataset labels, if accuracy calculation is desired. mat.NewDense(1, 1, nil)
Training *mat.Dense Training dataset. mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel randomForestModel Model to save trained random forest to.
predictions *mat.Dense (1d with ints) Predicted classes for each point in the test set.
probabilities *mat.Dense Predicted class probabilities for each point in the test set.

Detailed documentation

This program is an implementation of the standard random forest classification algorithm by Leo Breiman. A random forest can be trained and saved for later use, or a random forest may be loaded and predictions or class probabilities for points may be generated.

The training set and associated labels are specified with the Training and Labels parameters, respectively. The labels should be in the range [0, num_classes - 1]. Optionally, if Labels is not specified, the labels are assumed to be the last dimension of the training dataset.

When a model is trained, the OutputModel output parameter may be used to save the trained model. A model may be loaded for predictions with the InputModelparameter. The InputModel parameter may not be specified when the Training parameter is specified. The MinimumLeafSize parameter specifies the minimum number of training points that must fall into each leaf for it to be split. The NumTrees controls the number of trees in the random forest. The MinimumGainSplit parameter controls the minimum required gain for a decision tree node to split. Larger values will force higher-confidence splits. The MaximumDepth parameter specifies the maximum depth of the tree. The SubspaceDim parameter is used to control the number of random dimensions chosen for an individual node’s split. If PrintTrainingAccuracy is specified, the calculated accuracy on the training set will be printed.

Test data may be specified with the Test parameter, and if performance measures are desired for that test set, labels for the test points may be specified with the TestLabels parameter. Predictions for each test point may be saved via the Predictionsoutput parameter. Class probabilities for each prediction may be saved with the Probabilities output parameter.

Example

For example, to train a random forest with a minimum leaf size of 20 using 10 trees on the dataset contained in datawith labels labels, saving the output random forest to rf_model and printing the training error, one could call

// Initialize optional parameters for RandomForest().
param := mlpack.RandomForestOptions()
param.Training = data
param.Labels = labels
param.MinimumLeafSize = 20
param.NumTrees = 10
param.PrintTrainingAccuracy = true

rf_model, _, _ := mlpack.RandomForest(param)

Then, to use that model to classify points in test_set and print the test error given the labels test_labels using that model, while saving the predictions for each point to predictions, one could call

// Initialize optional parameters for RandomForest().
param := mlpack.RandomForestOptions()
param.InputModel = &rf_model
param.Test = test_set
param.TestLabels = test_labels

_, predictions, _ := mlpack.RandomForest(param)

See also

An implementation of range search with single-tree and dual-tree algorithms. Given a set of reference points and a set of query points and a range, this can find the set of reference points within the desired range for each query point, and any trees built during the computation can be saved for reuse with future range searches. .

Krann()

K-Rank-Approximate-Nearest-Neighbors (kRANN)

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for Krann().
param := mlpack.KrannOptions()
param.Alpha = 0.95
param.FirstLeafExact = false
param.InputModel = nil
param.K = 0
param.LeafSize = 20
param.Naive = false
param.Query = mat.NewDense(1, 1, nil)
param.RandomBasis = false
param.Reference = mat.NewDense(1, 1, nil)
param.SampleAtLeaves = false
param.Seed = 0
param.SingleMode = false
param.SingleSampleLimit = 20
param.Tau = 5
param.TreeType = "kd"

distances, neighbors, output_model := mlpack.Krann(param)

An implementation of rank-approximate k-nearest-neighbor search (kRANN) using single-tree and dual-tree algorithms. Given a set of reference points and query points, this can find the k nearest neighbors in the reference set of each query point using trees; trees that are built can be saved for future use. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Alpha float64 The desired success probability. 0.95
FirstLeafExact bool The flag to trigger sampling only after exactly exploring the first leaf. false
InputModel rannModel Pre-trained kNN model. nil
K int Number of nearest neighbors to find. 0
LeafSize int Leaf size for tree building (used for kd-trees, UB trees, R trees, R* trees, X trees, Hilbert R trees, R+ trees, R++ trees, and octrees). 20
Naive bool If true, sampling will be done without using a tree. false
Query *mat.Dense Matrix containing query points (optional). mat.NewDense(1, 1, nil)
RandomBasis bool Before tree-building, project the data onto a random orthogonal basis. false
Reference *mat.Dense Matrix containing the reference dataset. mat.NewDense(1, 1, nil)
SampleAtLeaves bool The flag to trigger sampling at leaves. false
Seed int Random seed (if 0, std::time(NULL) is used). 0
SingleMode bool If true, single-tree search is used (as opposed to dual-tree search. false
SingleSampleLimit int The limit on the maximum number of samples (and hence the largest node you can approximate). 20
Tau float64 The allowed rank-error in terms of the percentile of the data. 5
TreeType string Type of tree to use: ‘kd’, ‘ub’, ‘cover’, ‘r’, ‘x’, ‘r-star’, ‘hilbert-r’, ‘r-plus’, ‘r-plus-plus’, ‘oct’. "kd"
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
distances *mat.Dense Matrix to output distances into.
neighbors *mat.Dense (with ints) Matrix to output neighbors into.
outputModel rannModel If specified, the kNN model will be output here.

Detailed documentation

This program will calculate the k rank-approximate-nearest-neighbors of a set of points. You may specify a separate set of reference points and query points, or just a reference set which will be used as both the reference and query set. You must specify the rank approximation (in %) (and optionally the success probability).

Example

For example, the following will return 5 neighbors from the top 0.1% of the data (with probability 0.95) for each point in input and store the distances in distances and the neighbors in neighbors.csv:

// Initialize optional parameters for Krann().
param := mlpack.KrannOptions()
param.Reference = input
param.K = 5
param.Tau = 0.1

distances, neighbors, _ := mlpack.Krann(param)

Note that tau must be set such that the number of points in the corresponding percentile of the data is greater than k. Thus, if we choose tau = 0.1 with a dataset of 1000 points and k = 5, then we are attempting to choose 5 nearest neighbors out of the closest 1 point – this is invalid and the program will terminate with an error message.

The output matrices are organized such that row i and column j in the neighbors output file corresponds to the index of the point in the reference set which is the i’th nearest neighbor from the point in the query set with index j. Row i and column j in the distances output file corresponds to the distance between those two points.

See also

SoftmaxRegression()

Softmax Regression

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for SoftmaxRegression().
param := mlpack.SoftmaxRegressionOptions()
param.InputModel = nil
param.Labels = mat.NewDense(1, 1, nil)
param.Lambda = 0.0001
param.MaxIterations = 400
param.NoIntercept = false
param.NumberOfClasses = 0
param.Test = mat.NewDense(1, 1, nil)
param.TestLabels = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)

output_model, predictions := mlpack.SoftmaxRegression(param)

An implementation of softmax regression for classification, which is a multiclass generalization of logistic regression. Given labeled data, a softmax regression model can be trained and saved for future use, or, a pre-trained softmax regression model can be used for classification of new points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
InputModel softmaxRegression File containing existing model (parameters). nil
Labels *mat.Dense (1d with ints) A matrix containing labels (0 or 1) for the points in the training set (y). The labels must order as a row. mat.NewDense(1, 1, nil)
Lambda float64 L2-regularization constant 0.0001
MaxIterations int Maximum number of iterations before termination. 400
NoIntercept bool Do not add the intercept term to the model. false
NumberOfClasses int Number of classes for classification; if unspecified (or 0), the number of classes found in the labels will be used. 0
Test *mat.Dense Matrix containing test dataset. mat.NewDense(1, 1, nil)
TestLabels *mat.Dense (1d with ints) Matrix containing test labels. mat.NewDense(1, 1, nil)
Training *mat.Dense A matrix containing the training set (the matrix of predictors, X). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
outputModel softmaxRegression File to save trained softmax regression model to.
predictions *mat.Dense (1d with ints) Matrix to save predictions for test dataset into.

Detailed documentation

This program performs softmax regression, a generalization of logistic regression to the multiclass case, and has support for L2 regularization. The program is able to train a model, load an existing model, and give predictions (and optionally their accuracy) for test data.

Training a softmax regression model is done by giving a file of training points with the Training parameter and their corresponding labels with the Labels parameter. The number of classes can be manually specified with the NumberOfClasses parameter, and the maximum number of iterations of the L-BFGS optimizer can be specified with the MaxIterations parameter. The L2 regularization constant can be specified with the Lambda parameter and if an intercept term is not desired in the model, the NoIntercept parameter can be specified.

The trained model can be saved with the OutputModel output parameter. If training is not desired, but only testing is, a model can be loaded with the InputModel parameter. At the current time, a loaded model cannot be trained further, so specifying both InputModel and Training is not allowed.

The program is also able to evaluate a model on test data. A test dataset can be specified with the Test parameter. Class predictions can be saved with the Predictions output parameter. If labels are specified for the test data with the TestLabels parameter, then the program will print the accuracy of the predictions on the given test set and its corresponding labels.

Example

For example, to train a softmax regression model on the data dataset with labels labels with a maximum of 1000 iterations for training, saving the trained model to sr_model, the following command can be used:

// Initialize optional parameters for SoftmaxRegression().
param := mlpack.SoftmaxRegressionOptions()
param.Training = dataset
param.Labels = labels

sr_model, _ := mlpack.SoftmaxRegression(param)

Then, to use sr_model to classify the test points in test_points, saving the output predictions to predictions, the following command can be used:

// Initialize optional parameters for SoftmaxRegression().
param := mlpack.SoftmaxRegressionOptions()
param.InputModel = &sr_model
param.Test = test_points

_, predictions := mlpack.SoftmaxRegression(param)

See also

SparseCoding()

Sparse Coding

import (
  "mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
)

// Initialize optional parameters for SparseCoding().
param := mlpack.SparseCodingOptions()
param.Atoms = 15
param.InitialDictionary = mat.NewDense(1, 1, nil)
param.InputModel = nil
param.Lambda1 = 0
param.Lambda2 = 0
param.MaxIterations = 0
param.NewtonTolerance = 1e-06
param.Normalize = false
param.ObjectiveTolerance = 0.01
param.Seed = 0
param.Test = mat.NewDense(1, 1, nil)
param.Training = mat.NewDense(1, 1, nil)

codes, dictionary, output_model := mlpack.SparseCoding(param)

An implementation of Sparse Coding with Dictionary Learning. Given a dataset, this will decompose the dataset into a sparse combination of a few dictionary elements, where the dictionary is learned during computation; a dictionary can be reused for future sparse coding of new points. Detailed documentation.

Input options

There are two types of input options: required options, which are passed directly to the function call, and optional options, which are passed via an initialized struct, which allows keyword access to each of the options.

name type description default
Atoms int Number of atoms in the dictionary. 15
InitialDictionary *mat.Dense Optional initial dictionary matrix. mat.NewDense(1, 1, nil)
InputModel sparseCoding File containing input sparse coding model. nil
Lambda1 float64 Sparse coding l1-norm regularization parameter. 0
Lambda2 float64 Sparse coding l2-norm regularization parameter. 0
MaxIterations int Maximum number of iterations for sparse coding (0 indicates no limit). 0
NewtonTolerance float64 Tolerance for convergence of Newton method. 1e-06
Normalize bool If set, the input data matrix will be normalized before coding. false
ObjectiveTolerance float64 Tolerance for convergence of the objective function. 0.01
Seed int Random seed. If 0, ‘std::time(NULL)’ is used. 0
Test *mat.Dense Optional matrix to be encoded by trained model. mat.NewDense(1, 1, nil)
Training *mat.Dense Matrix of training data (X). mat.NewDense(1, 1, nil)
Verbose bool Display informational messages and the full list of parameters and timers at the end of execution. false

Output options

Output options are returned via Go’s support for multiple return values, in the order listed below.

name type description
codes *mat.Dense Matrix to save the output sparse codes of the test matrix (–test_file) to.
dictionary *mat.Dense Matrix to save the output dictionary to.
outputModel sparseCoding File to save trained sparse coding model to.

Detailed documentation

An implementation of Sparse Coding with Dictionary Learning, which achieves sparsity via an l1-norm regularizer on the codes (LASSO) or an (l1+l2)-norm regularizer on the codes (the Elastic Net). Given a dense data matrix X with d dimensions and n points, sparse coding seeks to find a dense dictionary matrix D with k atoms in d dimensions, and a sparse coding matrix Z with n points in k dimensions.

The original data matrix X can then be reconstructed as Z * D. Therefore, this program finds a representation of each point in X as a sparse linear combination of atoms in the dictionary D.

The sparse coding is found with an algorithm which alternates between a dictionary step, which updates the dictionary D, and a sparse coding step, which updates the sparse coding matrix.

Once a dictionary D is found, the sparse coding model may be used to encode other matrices, and saved for future usage.

To run this program, either an input matrix or an already-saved sparse coding model must be specified. An input matrix may be specified with the Training option, along with the number of atoms in the dictionary (specified with the Atoms parameter). It is also possible to specify an initial dictionary for the optimization, with the InitialDictionary parameter. An input model may be specified with the InputModel parameter.

Example

As an example, to build a sparse coding model on the dataset data using 200 atoms and an l1-regularization parameter of 0.1, saving the model into model, use

// Initialize optional parameters for SparseCoding().
param := mlpack.SparseCodingOptions()
param.Training = data
param.Atoms = 200
param.Lambda1 = 0.1

_, _, model := mlpack.SparseCoding(param)

Then, this model could be used to encode a new matrix, otherdata, and save the output codes to codes:

// Initialize optional parameters for SparseCoding().
param := mlpack.SparseCodingOptions()
param.InputModel = &model
param.Test = otherdata

codes, _, _ := mlpack.SparseCoding(param)

See also

changelog/history

mlpack ?.?.?

????-??-??
  • Added Mean Absolute Percentage Error.

mlpack 3.4.1

2020-09-07
  • Fix incorrect parsing of required matrix/model parameters for command-line bindings #2600.

  • Add manual type specification support to data::Load() and data::Save() (#2084, #2135, #2602).

  • Remove use of internal Armadillo functionality (#2596, #2601, #2602).

mlpack 3.4.0

2020-09-01
  • Issue warnings when metrics produce NaNs in KFoldCV #2595.

  • Added bindings for R during Google Summer of Code #2556.

  • Added common striptype function for all bindings #2556.

  • Refactored common utility function of bindings to bindings/util #2556.

  • Renamed InformationGain to HoeffdingInformationGain in methods/hoeffding_trees/information_gain.hpp #2556.

  • Added macro for changing stream of printing and warnings/errors #2556.

  • Added Spatial Dropout layer #2564.

  • Force CMake to show error when it didn’t find Python/modules #2568.

  • Refactor ProgramInfo() to separate out all the different information #2558.

  • Add bindings for one-hot encoding #2325.

  • Added Soft Actor-Critic to RL methods #2487.

  • Added Categorical DQN to q_networks #2454.

  • Added N-step DQN to q_networks #2461.

  • Add Silhoutte Score metric and Pairwise Distances #2406.

  • Add Go bindings for some missed models #2460.

  • Replace boost program_options dependency with CLI11 #2459.

  • Additional functionality for the ARFF loader #2486; use case sensitive categories #2516.

  • Add bayesian_linear_regression binding for the command-line, Python, Julia, and Go. Also called “Bayesian Ridge”, this is equivalent to a version of linear regression where the regularization parameter is automatically tuned #2030.

  • Fix defeatist search for spill tree traversals (#2566, #1269).

  • Fix incremental training of logistic regression models #2560.

  • Change default configuration of BUILD_PYTHON_BINDINGS to OFF #2575.

mlpack 3.3.2

2020-06-18
  • Added Noisy DQN to q_networks #2446.

  • Add Go bindings #1884.

  • Added Dueling DQN to q_networks, Noisy linear layer to ann/layer and Empty loss to ann/loss_functions #2414.

  • Storing and adding accessor method for action in q_learning #2413.

  • Added accessor methods for ANN layers #2321.

  • Addition of Elliot activation function #2268.

  • Add adaptive max pooling and adaptive mean pooling layers #2195.

  • Add parameter to avoid shuffling of data in preprocess_split #2293.

  • Add MatType parameter to LSHSearch, allowing sparse matrices to be used for search #2395.

  • Documentation fixes to resolve Doxygen warnings and issues #2400.

  • Add Load and Save of Sparse Matrix #2344.

  • Add Intersection over Union (IoU) metric for bounding boxes #2402.

  • Add Non Maximal Supression (NMS) metric for bounding boxes #2410.

  • Fix no_intercept and probability computation for linear SVM bindings #2419.

  • Fix incorrect neighbors for k > 1 searches in approx_kfn binding, for the QDAFN algorithm #2448.

  • Fix serialization of kernels with state for FastMKS #2452.

  • Add RBF layer in ann module to make RBFN architecture #2261.

mlpack 3.3.1

2020-04-29
  • Minor Julia and Python documentation fixes #2373.

  • Updated terminal state and fixed bugs for Pendulum environment (#2354, #2369).

  • Added EliSH activation function #2323.

  • Add L1 Loss function #2203.

  • Pass CMAKE_CXX_FLAGS (compilation options) correctly to Python build #2367.

  • Expose ensmallen Callbacks for sparseautoencoder #2198.

  • Bugfix for LARS class causing invalid read #2374.

  • Add serialization support from Julia; use mlpack.serialize() and mlpack.deserialize() to save and load from IOBuffers.

mlpack 3.3.0

2020-04-07
  • Added Normal Distribution to ann/dists #2382.

  • Templated return type of Forward function of loss functions #2339.

  • Added R2 Score regression metric #2323.

  • Added poisson negative log likelihood loss function #2196.

  • Added huber loss function #2199.

  • Added mean squared logarithmic error loss function for neural networks #2210.

  • Added mean bias loss function for neural networks #2210.

  • The DecisionStump class has been marked deprecated; use the DecisionTree class with NoRecursion=true or use ID3DecisionStump instead #2099.

  • Added probabilities_file parameter to get the probabilities matrix of AdaBoost classifier #2050.

  • Fix STB header search paths #2104.

  • Add DISABLE_DOWNLOADS CMake configuration option #2104.

  • Add padding layer in TransposedConvolutionLayer #2082.

  • Fix pkgconfig generation on non-Linux systems #2101.

  • Use log-space to represent HMM initial state and transition probabilities #2081.

  • Add functions to access parameters of Convolution and AtrousConvolution layers #1985.

  • Add Compute Error function in lars regression and changing Train function to return computed error #2139.

  • Add Julia bindings #1949. Build settings can be controlled with the BUILD_JULIA_BINDINGS=(ON/OFF) and JULIA_EXECUTABLE=/path/to/julia CMake parameters.

  • CMake fix for finding STB include directory #2145.

  • Add bindings for loading and saving images #2019; mlpack_image_converter from the command-line, mlpack.image_converter() from Python.

  • Add normalization support for CF binding #2136.

  • Add Mish activation function #2158.

  • Update init_rules in AMF to allow users to merge two initialization rules #2151.

  • Add GELU activation function #2183.

  • Better error handling of eigendecompositions and Cholesky decompositions (#2088, #1840).

  • Add LiSHT activation function #2182.

  • Add Valid and Same Padding for Transposed Convolution layer #2163.

  • Add CELU activation function #2191

  • Add Log-Hyperbolic-Cosine Loss function #2207.

  • Change neural network types to avoid unnecessary use of rvalue references #2259.

  • Bump minimum Boost version to 1.58 #2305.

  • Refactor STB support so HAS_STB macro is not needed when compiling against mlpack #2312.

  • Add Hard Shrink Activation Function #2186.

  • Add Soft Shrink Activation Function #2174.

  • Add Hinge Embedding Loss Function #2229.

  • Add Cosine Embedding Loss Function #2209.

  • Add Margin Ranking Loss Function #2264.

  • Bugfix for incorrect parameter vector sizes in logistic regression and softmax regression #2359.

mlpack 3.2.2

2019-11-26
  • Add valid and same padding option in Convolution and Atrous Convolution layer #1988.

  • Add Model() to the FFN class to access individual layers #2043.

  • Update documentation for pip and conda installation packages #2044.

  • Add bindings for linear SVM #1935; mlpack_linear_svm from the command-line, linear_svm() from Python.

  • Add support to return the layer name as std::string #1987.

  • Speed and memory improvements for the Transposed Convolution layer #1493.

  • Fix Windows Python build configuration #1885.

  • Validate md5 of STB library after download #2087.

  • Add __version__ to __init__.py #2092.

  • Correctly handle RNN sequences that are shorter than the value of rho #2102.

mlpack 3.2.1

2019-10-01
  • Enforce CMake version check for ensmallen #2032.

  • Fix CMake check for Armadillo version #2029.

  • Better handling of when STB is not installed #2033.

  • Fix Naive Bayes classifier computations in high dimensions #2022.

mlpack 3.2.0

2019-09-25
  • Fix some potential infinity errors in Naive Bayes Classifier #2022.

  • Fix occasionally-failing RADICAL test #1924.

  • Fix gcc 9 OpenMP compilation issue #1970.

  • Added support for loading and saving of images #1903.

  • Add Multiple Pole Balancing Environment (#1901, #1951).

  • Added functionality for scaling of data #1876; see the command-line binding mlpack_preprocess_scale or Python binding preprocess_scale().

  • Add new parameter maximum_depth to decision tree and random forest bindings #1916.

  • Fix prediction output of softmax regression when test set accuracy is calculated #1922.

  • Pendulum environment now checks for termination. All RL environments now have an option to terminate after a set number of time steps (no limit by default) #1941.

  • Add support for probabilistic KDE (kernel density estimation) error bounds when using the Gaussian kernel #1934.

  • Fix negative distances for cover tree computation #1979.

  • Fix cover tree building when all pairwise distances are 0 #1986.

  • Improve KDE pruning by reclaiming not used error tolerance (#1954, #1984).

  • Optimizations for sparse matrix accesses in z-score normalization for CF #1989.

  • Add kmeans_max_iterations option to GMM training binding gmm_train_main.

  • Bump minimum Armadillo version to 8.400.0 due to ensmallen dependency requirement #2015.

mlpack 3.1.1

2019-05-26
  • Fix random forest bug for numerical-only data #1887.

  • Significant speedups for random forest #1887.

  • Random forest now has minimum_gain_split and subspace_dim parameters #1887.

  • Decision tree parameter print_training_error deprecated in favor of print_training_accuracy.

  • output option changed to predictions for adaboost and perceptron binding. Old options are now deprecated and will be preserved until mlpack 4.0.0 #1882.

  • Concatenated ReLU layer #1843.

  • Accelerate NormalizeLabels function using hashing instead of linear search (see src/mlpack/core/data/normalize_labels_impl.hpp) #1780.

  • Add ConfusionMatrix() function for checking performance of classifiers #1798.

  • Install ensmallen headers when it is downloaded during build #1900.

mlpack 3.1.0

2019-04-25
  • Add DiagonalGaussianDistribution and DiagonalGMM classes to speed up the diagonal covariance computation and deprecate DiagonalConstraint #1666.

  • Add kernel density estimation (KDE) implementation with bindings to other languages #1301.

  • Where relevant, all models with a Train() method now return a double value representing the goodness of fit (i.e. final objective value, error, etc.) #1678.

  • Add implementation for linear support vector machine (see src/mlpack/methods/linear_svm).

  • Change DBSCAN to use PointSelectionPolicy and add OrderedPointSelection #1625.

  • Residual block support #1594.

  • Bidirectional RNN #1626.

  • Dice loss layer (#1674, #1714) and hard sigmoid layer #1776.

  • output option changed to predictions and output_probabilities to probabilities for Naive Bayes binding (mlpack_nbc/nbc()). Old options are now deprecated and will be preserved until mlpack 4.0.0 #1616.

  • Add support for Diagonal GMMs to HMM code (#1658, #1666). This can provide large speedup when a diagonal GMM is acceptable as an emission probability distribution.

  • Python binding improvements: check parameter type #1717, avoid copying Pandas dataframes #1711, handle Pandas Series objects #1700.

mlpack 3.0.4

2018-11-13
  • Bump minimum CMake version to 3.3.2.

  • CMake fixes for Ninja generator by Marc Espie.

mlpack 3.0.3

2018-07-27
  • Fix Visual Studio compilation issue #1443.

  • Allow running local_coordinate_coding binding with no initial_dictionary parameter when input_model is not specified #1457.

  • Make use of OpenMP optional via the CMake ‘USE_OPENMP’ configuration variable #1474.

  • Accelerate FNN training by 20-30% by avoiding redundant calculations #1467.

  • Fix math::RandomSeed() usage in tests (#1462, #1440).

  • Generate better Python setup.py with documentation #1460.

mlpack 3.0.2

2018-06-08
  • Documentation generation fixes for Python bindings #1421.

  • Fix build error for man pages if command-line bindings are not being built #1424.

  • Add ‘shuffle’ parameter and Shuffle() method to KFoldCV #1412. This will shuffle the data when the object is constructed, or when Shuffle() is called.

  • Added neural network layers: AtrousConvolution #1390, Embedding #1401, and LayerNorm (layer normalization) #1389.

  • Add Pendulum environment for reinforcement learning #1388 and update Mountain Car environment #1394.

mlpack 3.0.1

2018-05-10
  • Fix intermittently failing tests #1387.

  • Add big-batch SGD (BBSGD) optimizer in src/mlpack/core/optimizers/bigbatch_sgd/ #1131.

  • Fix simple compiler warnings (#1380, #1373).

  • Simplify NeighborSearch constructor and Train() overloads #1378.

  • Add warning for OpenMP setting differences (#1358/#1382). When mlpack is compiled with OpenMP but another application is not (or vice versa), a compilation warning will now be issued.

  • Restructured loss functions in src/mlpack/methods/ann/ #1365.

  • Add environments for reinforcement learning tests (#1368, #1370, #1329).

  • Allow single outputs for multiple timestep inputs for recurrent neural networks #1348.

  • Add He and LeCun normal initializations for neural networks #1342. Neural networks: add He and LeCun normal initializations #1342, add FReLU and SELU activation functions (#1346, #1341), add alpha-dropout #1349.

mlpack 3.0.0

2018-03-30
  • Speed and memory improvements for DBSCAN. –single_mode can now be used for situations where previously RAM usage was too high.

  • Bump minimum required version of Armadillo to 6.500.0.

  • Add automatically generated Python bindings. These have the same interface as the command-line programs.

  • Add deep learning infrastructure in src/mlpack/methods/ann/.

  • Add reinforcement learning infrastructure in src/mlpack/methods/reinforcement_learning/.

  • Add optimizers: AdaGrad, CMAES, CNE, FrankeWolfe, GradientDescent, GridSearch, IQN, Katyusha, LineSearch, ParallelSGD, SARAH, SCD, SGDR, SMORMS3, SPALeRA, SVRG.

  • Add hyperparameter tuning infrastructure and cross-validation infrastructure in src/mlpack/core/cv/ and src/mlpack/core/hpt/.

  • Fix bug in mean shift.

  • Add random forests (see src/mlpack/methods/random_forest).

  • Numerous other bugfixes and testing improvements.

  • Add randomized Krylov SVD and Block Krylov SVD.

mlpack 2.2.5

2017-08-25
  • Compilation fix for some systems #1082.

  • Fix PARAM_INT_OUT() #1100.

mlpack 2.2.4

2017-07-18
  • Speed and memory improvements for DBSCAN. –single_mode can now be used for situations where previously RAM usage was too high.

  • Fix bug in CF causing incorrect recommendations.

mlpack 2.2.3

2017-05-24
  • Bug fix for –predictions_file in mlpack_decision_tree program.

mlpack 2.2.2

2017-05-04
  • Install backwards-compatibility mlpack_allknn and mlpack_allkfn programs; note they are deprecated and will be removed in mlpack 3.0.0 #992.

  • Fix RStarTree bug that surfaced on OS X only #964.

  • Small fixes for MiniBatchSGD and SGD and tests.

mlpack 2.2.1

2017-04-13
  • Compilation fix for mlpack_nca and mlpack_test on older Armadillo versions #984.

mlpack 2.2.0

2017-03-21
  • Bugfix for mlpack_knn program #816.

  • Add decision tree implementation in methods/decision_tree/. This is very similar to a C4.5 tree learner.

  • Add DBSCAN implementation in methods/dbscan/.

  • Add support for multidimensional discrete distributions (#810, #830).

  • Better output for Log::Debug/Log::Info/Log::Warn/Log::Fatal for Armadillo objects (#895, #928).

  • Refactor categorical CSV loading with boost::spirit for faster loading #681.

mlpack 2.1.1

2016-12-22
  • HMMs now use random initialization; this should fix some convergence issues #828.

  • HMMs now initialize emissions according to the distribution of observations #833.

  • Minor fix for formatted output #814.

  • Fix DecisionStump to properly work with any input type.

mlpack 2.1.0

2016-10-31
  • Fixed CoverTree to properly handle single-point datasets.

  • Fixed a bug in CosineTree (and thus QUIC-SVD) that caused split failures for some datasets #717.

  • Added mlpack_preprocess_describe program, which can be used to print statistics on a given dataset #742.

  • Fix prioritized recursion for k-furthest-neighbor search (mlpack_kfn and the KFN class), leading to orders-of-magnitude speedups in some cases.

  • Bump minimum required version of Armadillo to 4.200.0.

  • Added simple Gradient Descent optimizer, found in src/mlpack/core/optimizers/gradient_descent/ #792.

  • Added approximate furthest neighbor search algorithms QDAFN and DrusillaSelect in src/mlpack/methods/approx_kfn/, with command-line program mlpack_approx_kfn.

mlpack 2.0.3

2016-07-21
  • Added multiprobe LSH #691. The parameter ‘T’ to LSHSearch::Search() can now be used to control the number of extra bins that are probed, as can the -T (–num_probes) option to mlpack_lsh.

  • Added the Hilbert R tree to src/mlpack/core/tree/rectangle_tree/ #664. It can be used as the typedef HilbertRTree, and it is now an option in the mlpack_knn, mlpack_kfn, mlpack_range_search, and mlpack_krann command-line programs.

  • Added the mlpack_preprocess_split and mlpack_preprocess_binarize programs, which can be used for preprocessing code (#650, #666).

  • Added OpenMP support to LSHSearch and mlpack_lsh #700.

mlpack 2.0.2

2016-06-20
  • Added the function LSHSearch::Projections(), which returns an arma::cube with each projection table in a slice #663. Instead of Projection(i), you should now use Projections().slice(i).

  • A new constructor has been added to LSHSearch that creates objects using projection tables provided in an arma::cube #663.

  • Handle zero-variance dimensions in DET #515.

  • Add MiniBatchSGD optimizer (src/mlpack/core/optimizers/minibatch_sgd/) and allow its use in mlpack_logistic_regression and mlpack_nca programs.

  • Add better backtrace support from Grzegorz Krajewski for Log::Fatal messages when compiled with debugging and profiling symbols. This requires libbfd and libdl to be present during compilation.

  • CosineTree test fix from Mikhail Lozhnikov #358.

  • Fixed HMM initial state estimation #600.

  • Changed versioning macros __MLPACK_VERSION_MAJOR, __MLPACK_VERSION_MINOR, and __MLPACK_VERSION_PATCH to MLPACK_VERSION_MAJOR, MLPACK_VERSION_MINOR, and MLPACK_VERSION_PATCH. The old names will remain in place until mlpack 3.0.0.

  • Renamed mlpack_allknn, mlpack_allkfn, and mlpack_allkrann to mlpack_knn, mlpack_kfn, and mlpack_krann. The mlpack_allknn, mlpack_allkfn, and mlpack_allkrann programs will remain as copies until mlpack 3.0.0.

  • Add –random_initialization option to mlpack_hmm_train, for use when no labels are provided.

  • Add –kill_empty_clusters option to mlpack_kmeans and KillEmptyClusters policy for the KMeans class (#595, #596).

mlpack 2.0.1

2016-02-04
  • Fix CMake to properly detect when MKL is being used with Armadillo.

  • Minor parameter handling fixes to mlpack_logistic_regression (#504, #505).

  • Properly install arma_config.hpp.

  • Memory handling fixes for Hoeffding tree code.

  • Add functions that allow changing training-time parameters to HoeffdingTree class.

  • Fix infinite loop in sparse coding test.

  • Documentation spelling fixes #501.

  • Properly handle covariances for Gaussians with large condition number #496, preventing GMMs from filling with NaNs during training (and also HMMs that use GMMs).

  • CMake fixes for finding LAPACK and BLAS as Armadillo dependencies when ATLAS is used.

  • CMake fix for projects using mlpack’s CMake configuration from elsewhere #512.

mlpack 2.0.0

2015-12-24
  • Removed overclustering support from k-means because it is not well-tested, may be buggy, and is (I think) unused. If this was support you were using, open a bug or get in touch with us; it would not be hard for us to reimplement it.

  • Refactored KMeans to allow different types of Lloyd iterations.

  • Added implementations of k-means: Elkan’s algorithm, Hamerly’s algorithm, Pelleg-Moore’s algorithm, and the DTNN (dual-tree nearest neighbor) algorithm.

  • Significant acceleration of LRSDP via the use of accu(a % b) instead of trace(a * b).

  • Added MatrixCompletion class (matrix_completion), which performs nuclear norm minimization to fill unknown values of an input matrix.

  • No more dependence on Boost.Random; now we use C++11 STL random support.

  • Add softmax regression, contributed by Siddharth Agrawal and QiaoAn Chen.

  • Changed NeighborSearch, RangeSearch, FastMKS, LSH, and RASearch API; these classes now take the query sets in the Search() method, instead of in the constructor.

  • Use OpenMP, if available. For now OpenMP support is only available in the DET training code.

  • Add support for predicting new test point values to LARS and the command-line ‘lars’ program.

  • Add serialization support for Perceptron and LogisticRegression.

  • Refactor SoftmaxRegression to predict into an arma::Row object, and add a softmax_regression program.

  • Refactor LSH to allow loading and saving of models.

  • ToString() is removed entirely #487.

  • Add –input_model_file and –output_model_file options to appropriate machine learning algorithms.

  • Rename all executables to start with an “mlpack” prefix #229.

  • Add HoeffdingTree and mlpack_hoeffding_tree, an implementation of the streaming decision tree methodology from Domingos and Hulten in 2000.

mlpack 1.0.12

2015-01-07
  • Switch to 3-clause BSD license (from LGPL).

mlpack 1.0.11

2014-12-11
  • Proper handling of dimension calculation in PCA.

  • Load parameter vectors properly for LinearRegression models.

  • Linker fixes for AugLagrangian specializations under Visual Studio.

  • Add support for observation weights to LinearRegression.

  • MahalanobisDistance<> now takes the root of the distance by default and therefore satisfies the triangle inequality (TakeRoot now defaults to true).

  • Better handling of optional Armadillo HDF5 dependency.

  • Fixes for numerous intermittent test failures.

  • math::RandomSeed() now sets the random seed for recent (>=3.930) Armadillo versions.

  • Handle Newton method convergence better for SparseCoding::OptimizeDictionary() and make maximum iterations a parameter.

  • Known bug: CosineTree construction may fail in some cases on i386 systems #358.

mlpack 1.0.10

2014-08-29
  • Bugfix for NeighborSearch regression which caused very slow allknn/allkfn. Speeds are now restored to approximately 1.0.8 speeds, with significant improvement for the cover tree #347.

  • Detect dependencies correctly when ARMA_USE_WRAPPER is not being defined (i.e., libarmadillo.so does not exist).

  • Bugfix for compilation under Visual Studio #348.

mlpack 1.0.9

2014-07-28
  • GMM initialization is now safer and provides a working GMM when constructed with only the dimensionality and number of Gaussians #301.

  • Check for division by 0 in Forward-Backward Algorithm in HMMs #301.

  • Fix MaxVarianceNewCluster (used when re-initializing clusters for k-means) #301.

  • Fixed implementation of Viterbi algorithm in HMM::Predict() #303.

  • Significant speedups for dual-tree algorithms using the cover tree (#235, #314) including a faster implementation of FastMKS.

  • Fix for LRSDP optimizer so that it compiles and can be used #312.

  • CF (collaborative filtering) now expects users and items to be zero-indexed, not one-indexed #311.

  • CF::GetRecommendations() API change: now requires the number of recommendations as the first parameter. The number of users in the local neighborhood should be specified with CF::NumUsersForSimilarity().

  • Removed incorrect PeriodicHRectBound #58.

  • Refactor LRSDP into LRSDP class and standalone function to be optimized #305.

  • Fix for centering in kernel PCA #337.

  • Added simulated annealing (SA) optimizer, contributed by Zhihao Lou.

  • HMMs now support initial state probabilities; these can be set in the constructor, trained, or set manually with HMM::Initial() #302.

  • Added Nyström method for kernel matrix approximation by Marcus Edel.

  • Kernel PCA now supports using Nyström method for approximation.

  • Ball trees now work with dual-tree algorithms, via the BallBound<> bound structure #307; fixed by Yash Vadalia.

  • The NMF class is now AMF<>, and supports far more types of factorizations, by Sumedh Ghaisas.

  • A QUIC-SVD implementation has returned, written by Siddharth Agrawal and based on older code from Mudit Gupta.

  • Added perceptron and decision stump by Udit Saxena (these are weak learners for an eventual AdaBoost class).

  • Sparse autoencoder added by Siddharth Agrawal.

mlpack 1.0.8

2014-01-06
  • Memory leak in NeighborSearch index-mapping code fixed #298.

  • GMMs can be trained using the existing model as a starting point by specifying an additional boolean parameter to GMM::Estimate() #296.

  • Logistic regression implementation added in methods/logistic_regression (see also #293).

  • L-BFGS optimizer now returns its function via Function().

  • Version information is now obtainable via mlpack::util::GetVersion() or the __MLPACK_VERSION_MAJOR, __MLPACK_VERSION_MINOR, and __MLPACK_VERSION_PATCH macros #297.

  • Fix typos in allkfn and allkrann output.

mlpack 1.0.7

2013-10-04
  • Cover tree support for range search (range_search), rank-approximate nearest neighbors (allkrann), minimum spanning tree calculation (emst), and FastMKS (fastmks).

  • Dual-tree FastMKS implementation added and tested.

  • Added collaborative filtering package (cf) that can provide recommendations when given users and items.

  • Fix for correctness of Kernel PCA (kernel_pca) #270.

  • Speedups for PCA and Kernel PCA #198.

  • Fix for correctness of Neighborhood Components Analysis (NCA) #279.

  • Minor speedups for dual-tree algorithms.

  • Fix for Naive Bayes Classifier (nbc) #269.

  • Added a ridge regression option to LinearRegression (linear_regression) #286.

  • Gaussian Mixture Models (gmm::GMM<>) now support arbitrary covariance matrix constraints #283.

  • MVU (mvu) removed because it is known to not work #183.

  • Minor updates and fixes for kernels (in mlpack::kernel).

mlpack 1.0.6

2013-06-13
  • Minor bugfix so that FastMKS gets built.

mlpack 1.0.5

2013-05-01
  • Speedups of cover tree traversers #235.

  • Addition of rank-approximate nearest neighbors (RANN), found in src/mlpack/methods/rann/.

  • Addition of fast exact max-kernel search (FastMKS), found in src/mlpack/methods/fastmks/.

  • Fix for EM covariance estimation; this should improve GMM training time.

  • More parameters for GMM estimation.

  • Force GMM and GaussianDistribution covariance matrices to be positive definite, so that training converges much more often.

  • Add parameter for the tolerance of the Baum-Welch algorithm for HMM training.

  • Fix for compilation with clang compiler.

  • Fix for k-furthest-neighbor-search.

mlpack 1.0.4

2013-02-08
  • Force minimum Armadillo version to 2.4.2.

  • Better output of class types to streams; a class with a ToString() method implemented can be sent to a stream with operator«.

  • Change return type of GMM::Estimate() to double #257.

  • Style fixes for k-means and RADICAL.

  • Handle size_t support correctly with Armadillo 3.6.2 #258.

  • Add locality-sensitive hashing (LSH), found in src/mlpack/methods/lsh/.

  • Better tests for SGD (stochastic gradient descent) and NCA (neighborhood components analysis).

mlpack 1.0.3

2012-09-16
  • Remove internal sparse matrix support because Armadillo 3.4.0 now includes it. When using Armadillo versions older than 3.4.0, sparse matrix support is not available.

  • NCA (neighborhood components analysis) now support an arbitrary optimizer #245, including stochastic gradient descent #249.

mlpack 1.0.2

2012-08-15
  • Added density estimation trees, found in src/mlpack/methods/det/.

  • Added non-negative matrix factorization, found in src/mlpack/methods/nmf/.

  • Added experimental cover tree implementation, found in src/mlpack/core/tree/cover_tree/ #157.

  • Better reporting of boost::program_options errors #225.

  • Fix for timers on Windows (#212, #211).

  • Fix for allknn and allkfn output #204.

  • Sparse coding dictionary initialization is now a template parameter #220.

mlpack 1.0.1

2012-03-03
  • Added kernel principal components analysis (kernel PCA), found in src/mlpack/methods/kernel_pca/ #74.

  • Fix for Lovasz-Theta AugLagrangian tests #182.

  • Fixes for allknn output (#185, #186).

  • Added range search executable #192.

  • Adapted citations in documentation to BibTeX; no citations in -h output #195.

  • Stop use of ‘const char*’ and prefer ‘std::string’ #176.

  • Support seeds for random numbers #177.

mlpack 1.0.0

2011-12-17
  • Initial release. See any resolved tickets numbered less than #196 or execute this query: http://www.mlpack.org/trac/query?status=closed&milestone=mlpack+1.0.0