density {GPUmatrix}

Kernel Density Estimation and Histograms

Description

The function density mimics the function density of the library stats to operate on gpu.matrix-class objects: "It computes kernel density estimates. Its default method does so with the given kernel and bandwidth for univariate observations."

The function 'hist' mimics the function 'hist' of the library 'graphics' to operate on gpu.matrix-class objects: "It computes a histogram of the given data values."

Usage

## S4 method for signature 'gpu.matrix.tensorflow'
density(x)
## S4 method for signature 'gpu.matrix.torch'
density(x)
## S4 method for signature 'gpu.matrix.tensorflow'
hist(x,...)
## S4 method for signature 'gpu.matrix.torch'
hist(x,...)

Arguments

x

the gpu.matrix object from which the estimate density is to be computed or the histogram is desired.

...

further arguments and graphical parameters.

Details

The two functions (density and hist) have been programmed to call their corresponding counterpart functions with their default parameters. Therefore, the internal operations to obtain each graph are computed by the CPU, regardless of whether the input value is stored in the GPU.

For more information on these functions see density, and hist.

Value

The function density returns the same output as its counterpart function density from the library stats: It returns "an object with class 'density' whose underlying structure is a list containing the following components.

x

the n coordinates of the points where the density is estimated.

y

the estimated density values. These will be non-negative, but can be zero.

bw

the bandwidth used.

n

the sample size after elimination of missing values.

call

the call which produced the result.

data.name

the deparsed name of the x argument.

has.na

logical, for compatibility (always FALSE).

The print method reports summary values on the x and y components." (taken from density).

On the other hand, the function hist returns the same output as its counterpart function hist from the library graphics: It returns "an object of class 'histogram' which is a list with components:

breaks

the n+1n+1 cell boundaries (= breaks if that was a vector). These are the nominal breaks, not with the boundary fuzz.

counts

n integers; for each cell, the number of x[] inside.

density

values \hat{f}(x_i), as estimated density values. If all(diff(breaks) == 1), the are the relative frequencies counts/n and in general satisfy \sum_i{\hat{f}(x_i)(b_{i+1}-b_i)=1}, where b_i = breaks[i]

.

mids

the n cell midpoints.

xname

a character string with the actual x argument name.

equidist

logical, indicating if the distances between breaks are all the same."

(Taken from hist)

See Also

For more information see: density, and hist

Examples

if(installTorch()){

  a <- gpu.matrix(rnorm(20*100),20,100)

  density(a[1,]) #density information
  plot(density(a[1,])) #plot the estimated density function

  hist(a[1,]) #plot the histogram

}