DataSet.Array: R6 class representing an array dataset

In modello: Homemade Deep Learning Library

Description

R6 class representing an array dataset

R6 class representing an array dataset

Details

It loads the data from an R array

Methods

Public methods

• DataSet.Array$new()

• DataSet.Array$length()

• DataSet.Array$has.next()

• DataSet.Array$reset()

• DataSet.Array$..pop.index..()

• DataSet.Array$..get.data..()

• DataSet.Array$feed()

• DataSet.Array$clone()

---

Method new()

Initialisation method.

Usage

DataSet.Array$new(X, indexes, yi, si)

Arguments

X

array

indexes

subset of indexes along the sliceing dimension

yi

subset of indexes to return as targets

si

index of slicing dimension

---

Method length()

Returns the length of the dataset

Usage

DataSet.Array$length()

Examples

X = matrix(rnorm(100), 20)
ds = DataSet.Array$new(X, 1:20, 1, 1)
ds$length() # 20

---

Method has.next()

Returns TRUE if the dataset has still data

Usage

DataSet.Array$has.next()

Examples

X = matrix(rnorm(100), 20)
ds = DataSet.Array$new(X, 1:20, 1, 1)
ds$has.next() # TRUE

---

Method reset()

Reset the dataset to its initial state

Usage

DataSet.Array$reset()

Examples

X = matrix(rnorm(100), 20)
ds = DataSet.Array$new(X, 1:20, 1, 1)
ds$..get.data..()
ds$..get.data..()
ds$reset()
ds$..get.data..()

---

Method ..pop.index..()

Auxiliary method not to be called directly. It pops an index from the dataset according to the given method

Usage

DataSet.Array$..pop.index..(mode = "head")

Arguments

mode

popping mode

---

Method ..get.data..()

Auxiliary method not to be called directly.

Usage

DataSet.Array$..get.data..(mode = "head")

Arguments

mode

popping mode

---

Method feed()

Method to feed the bathces of data to 'numbers'.

Usage

DataSet.Array$feed(x, mode = "head")

Arguments

x

number

mode

popping mode

Examples

modello.init(100, 100, 100, 100)
X = matrix(rnorm(100), 20)
ds = DataSet.Array$new(X, 1:20, 1, 1)
x = number(rnorm(5))
print(x$v)
ds$feed(x)
print(x$v)
modello.close()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

DataSet.Array$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Author(s)

Filippo Monari

Examples

## ------------------------------------------------
## Method `DataSet.Array$length`
## ------------------------------------------------

X = matrix(rnorm(100), 20)
ds = DataSet.Array$new(X, 1:20, 1, 1)
ds$length() # 20

## ------------------------------------------------
## Method `DataSet.Array$has.next`
## ------------------------------------------------

X = matrix(rnorm(100), 20)
ds = DataSet.Array$new(X, 1:20, 1, 1)
ds$has.next() # TRUE

## ------------------------------------------------
## Method `DataSet.Array$reset`
## ------------------------------------------------

X = matrix(rnorm(100), 20)
ds = DataSet.Array$new(X, 1:20, 1, 1)
ds$..get.data..()
ds$..get.data..()
ds$reset()
ds$..get.data..()

## ------------------------------------------------
## Method `DataSet.Array$feed`
## ------------------------------------------------

modello.init(100, 100, 100, 100)
X = matrix(rnorm(100), 20)
ds = DataSet.Array$new(X, 1:20, 1, 1)
x = number(rnorm(5))
print(x$v)
ds$feed(x)
print(x$v)
modello.close()

modello documentation built on Feb. 2, 2021, 9:06 a.m.