Package 'mlapi'

Fits model to data

Description

Generic function to fit models (inherited from mlapiEstimation)

Usage

fit(x, model, y = NULL, ...)

## S3 method for class 'Matrix'
fit(x, model, y = NULL, ...)

## S3 method for class 'matrix'
fit(x, model, y = NULL, ...)

Arguments

x	A matrix like object, should inherit from Matrix or matrix
model	instance of class estimator which should implement method with signature $fit(x, y, ...)
y	NULL by default. Optional response variable for supervised learning models. Should inherit from vector or Matrix or matrix. See documentation for corresponding models.
...	additional data/model dependent arguments to downstream functions.

Value

invisible(object$self())

---

Fit model to the data, then transforms data

Description

Generic function to fit transformers (inherits from mlapiTransformation)

Usage

fit_transform(x, model, y = NULL, ...)

## S3 method for class 'Matrix'
fit_transform(x, model, y = NULL, ...)

## S3 method for class 'matrix'
fit_transform(x, model, y = NULL, ...)

Arguments

x	A matrix like object, should inherit from Matrix or matrix
model	instance of class estimator which should implement method with signature $fit(x, ...)
y	NULL by default. Optional response variable for supervised models. Should inherit from vector Matrix or matrix. See documentation for corresponding models.
...	additional data/model dependent arguments to downstream functions.

Value

Transformed version of the x

---

Base abstract class for all decompositions

Description

Base class for all decompositions which are methods which can decompose matrix into 2 low-dimensional matrices x = f(A, B). (Think of this Latent Dirichlet Allocation, Non-negative Matrix Factorization, etc). It iherits from mlapiTransformation and additionally requires to implement components member.

Base class for all decompositions which are methods which can decompose matrix into 2 low-dimensional matrices x = f(A, B) incrementally. It iherits from mlapiDecomposition and additionally requires to implement partial_fit method which can learn components incrementally.

Usage

mlapiDecomposition

mlapiDecompositionOnline

Format

R6Class object.

Fields

components

features embeddings. So if matrix is decomposed in a form x = f(A, B) where X = n\*m, A = n\*k, B = k\*m them B = components

components

features embeddings. So if matrix is decomposed in a form x = f(A, B) where X = n\*m, A = n\*k, B = k\*m them B = components

Methods

$fit_transform(x, y = NULL, ...)

$transform(x, ...)

Performs transformation of the new data (after model was trained)

$fit_transform(x, y = NULL, ...)

$partial_fit(x, y = NULL, ...)

$transform(x, ...)

Performs transformation of the new data (after model was trained)

Arguments

x

A matrix like object, should inherit from Matrix or matrix. Allowed classes should be defined in child classes.

y

NULL. Optional taget variable. Usually this should be NULL. There few cases when it could be used.

...

additional parameters with default values

x

A matrix like object, should inherit from Matrix or matrix. Allowed classes should be defined in child classes.

y

NULL. Optional taget variable. Usually this should be NULL. There few cases when it could be used.

...

additional parameters with default values

Examples

TruncatedSVD = R6::R6Class(
  classname = "TruncatedSVD",
  inherit = mlapi::mlapiDecomposition,
  public = list(
    initialize = function(rank = 10) {
      private$rank = rank
      super$set_internal_matrix_formats(dense = "matrix", sparse = NULL)
    },
    fit_transform = function(x, ...) {
      x = super$check_convert_input(x)
      private$n_features = ncol(x)
      svd_fit = svd(x, nu = private$rank, nv = private$rank, ...)
      sing_values = svd_fit$d[seq_len(private$rank)]
      result = svd_fit$u %*% diag(x = sqrt(sing_values))
      private$components_ = t(svd_fit$v %*% diag(x = sqrt(sing_values)))
      rm(svd_fit)
      rownames(result) = rownames(x)
      colnames(private$components_) = colnames(x)
      private$fitted = TRUE
      invisible(result)
    },
    transform = function(x, ...) {
      if (private$fitted) {
        stopifnot(ncol(x) == ncol(private$components_))
        lhs = tcrossprod(private$components_)
        rhs = as.matrix(tcrossprod(private$components_, x))
        t(solve(lhs, rhs))
      }
      else
        stop("Fit the model first woth model$fit_transform()!")
    }
  ),
  private = list(
    rank = NULL,
    n_features = NULL,
    fitted = NULL
  )
)
set.seed(1)
model = TruncatedSVD$new(2)
x = matrix(sample(100 * 10, replace = TRUE), ncol = 10)
x_trunc = model$fit_transform(x)
dim(x_trunc)

x_trunc_2 = model$transform(x)
sum(x_trunc_2 - x_trunc)

#' check pipe-compatible S3 interface
x_trunc_2_s3 = transform(x, model)
identical(x_trunc_2, x_trunc_2_s3)

---

Base abstract class for all classification/regression models

Description

Base class for all estimators. Defines minimal set of members and methods(with signatires) which have to be implemented in child classes.

Usage

mlapiEstimation

Format

R6Class object.

Methods

$fit(x, y, ...)

$predict(x, ...)

Makes predictions on new data (after model was trained)

Arguments

x

A matrix like object, should inherit from Matrix or matrix. Allowed classes should be defined in child classes.

y

target - usually vector, but also can be a matrix like object. Allowed classes should be defined in child classes.

...

additional parameters with default values

Examples

SimpleLinearModel = R6::R6Class(
classname = "mlapiSimpleLinearModel",
inherit = mlapi::mlapiEstimation,
public = list(
  initialize = function(tol = 1e-7) {
    private$tol = tol
    super$set_internal_matrix_formats(dense = "matrix", sparse = NULL)
  },
  fit = function(x, y, ...) {
    x = super$check_convert_input(x)
   stopifnot(is.vector(y))
   stopifnot(is.numeric(y))
   stopifnot(nrow(x) == length(y))

   private$n_features = ncol(x)
   private$coefficients = .lm.fit(x, y, tol = private$tol)[["coefficients"]]
 },
 predict = function(x) {
   stopifnot(ncol(x) == private$n_features)
   x %*% matrix(private$coefficients, ncol = 1)
 }
),
private = list(
  tol = NULL,
  coefficients = NULL,
  n_features = NULL
))
set.seed(1)
model = SimpleLinearModel$new()
x = matrix(sample(100 * 10, replace = TRUE), ncol = 10)
y = sample(c(0, 1), 100, replace = TRUE)
model$fit(as.data.frame(x), y)
res1 = model$predict(x)
# check pipe-compatible S3 interface
res2 = predict(x, model)
identical(res1, res2)

---

Base abstract class for all classification/regression models which can be trained incremendally (online)

Description

Base class for all online estimators. This class inherits from mlapiEstimation and additionally requires to implement $partial_fit(x, y, ...) method. Idea is that user can pass x, y in chunks and model will be updated/refined incrementally.

Usage

mlapiEstimationOnline

Format

R6Class object.

Methods

$fit(x, y, ...)

$partial_fit(x, y, ...)

$predict(x, ...)

Makes predictions on new data (after model was trained)

Arguments

x

A matrix like object, should inherit from Matrix or matrix. Allowed classes should be defined in child classes.

y

target - usually vector, but also can be a matrix like object. Allowed classes should be defined in child classes.

...

additional parameters with default values

---

Base abstract class for all transformations

Description

Base class for all online transformations.

Usage

mlapiTransformation

Format

R6Class object.

Methods

$fit_transform(x, y = NULL, ...)

$transform(x, ...)

Performs transformation of the new data (after model was trained)

Arguments

x

A matrix like object, should inherit from Matrix or matrix. Allowed classes should be defined in child classes.

y

NULL. Optional taget variable. Usually this should be NULL. There few cases when it could be used.

...

additional parameters with default values

---

Base abstract class for all transformations which can be trained incremendally (online)

Description

Base class for all online transformations. This class inherits from mlapiTransformation and additionally requires to implement $partial_fit(x, y, ...) method. Idea is that user can pass x, y in chunks and model will be updated/refined incrementally.

Usage

mlapiTransformationOnline

Format

R6Class object.

Methods

$fit_transform(x, y = NULL, ...)

$transform(x, ...)

Performs transformation of the new data (after model was trained)

Arguments

x

A matrix like object, should inherit from Matrix or matrix. Allowed classes should be defined in child classes.

y

NULL. Optional taget variable. Usually this should be NULL. There few cases when it could be used.

...

additional parameters with default values

---

Makes predictions on new data using pre-trained model

Description

Makes predictions on new data using pre-trained model (inherits from mlapiEstimation)

Usage

## S3 method for class 'matrix'
predict(object, model, ...)

## S3 method for class 'Matrix'
predict(object, model, ...)

Arguments

object	= x in other methods. A matrix like object, should inherit from Matrix or matrix
model	object which inherits class mlapiEstimation which implements method model$predict(x, ...)
...	additional data/model dependent arguments to downstream functions

---

Transforms new data using pre-trained model

Description

Generic function to transform data with pre-trained model (inherits from mlapiTransformation)

Usage

## S3 method for class 'Matrix'
transform(`_data`, model, ...)

## S3 method for class 'matrix'
transform(`_data`, model, ...)

Arguments

_data	= x in other methods. A matrix like object, should inherit from Matrix or matrix
model	object of class mlapiTransformation which implements method $transform(x, ...)
...	additional data/model dependent arguments to downstream functions.

---