gpu.matrix-class: Class 'gpu.matrix' for matrix stored in GPU

In GPUmatrix: Basic Linear Algebra with GPU

Class 'gpu.matrix' for matrix stored in GPU

Description

GPU computational power is a great resource for computational biology specifically in statistics and linear algebra. the gpu.matrix-class is a class of the GPUmatrix package, that store a matrix in the GPU.

The GPUmatrix package is based on S4 objects in R and we have created a constructor function that acts similarly to the default matrix constructor in R for CPU matrices. The constructor function is gpu.matrix and accepts the same parameters as matrix.

Slots

Use the @ operator to access the different slots:

rownames

the row names of the gpu.matrix

colnames

the colunm names of the gpu.matrix

gm

the corresponding tensor

sparse

Logical: indicates if the gpu.matrix is sparse or not

type

If it is tensorflow or torch

See Also

See Also gpu.matrix, Matrix, and matrix..

GPUmatrix documentation built on May 29, 2024, 11:02 a.m.