Package: GPUmatrix 1.0.2
GPUmatrix: Basic Linear Algebra with GPU
GPUs are great resources for data analysis, especially in statistics and linear algebra. Unfortunately, very few packages connect R to the GPU, and none of them are transparent enough to run the computations on the GPU without substantial changes to the code. The maintenance of these packages is cumbersome: several of the earlier attempts have been removed from their respective repositories. It would be desirable to have a properly maintained R package that takes advantage of the GPU with minimal changes to the existing code. We have developed the GPUmatrix package (available on CRAN). GPUmatrix mimics the behavior of the Matrix package and extends R to use the GPU for computations. It includes single(FP32) and double(FP64) precision data types, and provides support for sparse matrices. It is easy to learn, and requires very few code changes to perform the operations on the GPU. GPUmatrix relies on either the Torch or Tensorflow R packages to perform the GPU operations. We have demonstrated its usefulness for several statistical applications and machine learning applications: non-negative matrix factorization, logistic regression and general linear models. We have also included a comparison of GPU and CPU performance on different matrix operations.
Authors:
GPUmatrix_1.0.2.tar.gz
GPUmatrix_1.0.2.zip(r-4.5)GPUmatrix_1.0.2.zip(r-4.4)GPUmatrix_1.0.2.zip(r-4.3)
GPUmatrix_1.0.2.tgz(r-4.4-any)GPUmatrix_1.0.2.tgz(r-4.3-any)
GPUmatrix_1.0.2.tar.gz(r-4.5-noble)GPUmatrix_1.0.2.tar.gz(r-4.4-noble)
GPUmatrix_1.0.2.tgz(r-4.4-emscripten)GPUmatrix_1.0.2.tgz(r-4.3-emscripten)
GPUmatrix.pdf |GPUmatrix.html✨
GPUmatrix/json (API)
| # Install 'GPUmatrix' in R: |
| install.packages('GPUmatrix', repos = c('https://clobatofern95.r-universe.dev', 'https://cloud.r-project.org')) |
This package does not link to any Github/Gitlab/R-forge repository. No issue tracker or development information is available.
Last updated 7 months agofrom:99aa24df83. Checks:OK: 1 NOTE: 6. Indexed: yes.
| Target | Result | Date |
|---|---|---|
| Doc / Vignettes | OK | Sep 15 2024 |
| R-4.5-win | NOTE | Sep 15 2024 |
| R-4.5-linux | NOTE | Sep 15 2024 |
| R-4.4-win | NOTE | Sep 15 2024 |
| R-4.4-mac | NOTE | Sep 15 2024 |
| R-4.3-win | NOTE | Sep 15 2024 |
| R-4.3-mac | NOTE | Sep 15 2024 |
Exports:%^%%x%applyas.arrayas.gpu.matrixas.listas.matrixcbind2cholchol_solvecolMaxscolMeanscolMinscolnamescolRankscolSumscolVarscorcovcov2cordensitydetdeterminantdiagdiag<-distdtypedtype<-eigenexpmGPUfftginvglm.fit.GPUgpu.matrixGPUglmheadhistinstallTorchLR_GradientConjugate_gpumatrixmeanmvfftncolNMFgpumatrixnrowprintqrqr.coefqr.Qqr.qtyqr.qyqr.Rqr.residqr.solveqr.Xrbind2rowMaxsrowMeansrowMinsrownamesrowRanksrowSumsrowVarssolvesortsvdttailto_denseto_sparsewhich.maxwhich.min
Dependencies:
Readme and manuals
Help Manual
| Help page | Topics |
|---|---|
| Array Transposition | t t,gpu.matrix.tensorflow-method t,gpu.matrix.torch-method t-methods |
| Apply Functions over 'gpu.matrix-class' margins | apply apply,gpu.matrix.tensorflow-method apply,gpu.matrix.torch-method apply-methods |
| as_methods | as.array as.array,gpu.matrix.tensorflow-method as.array,gpu.matrix.torch-method as.array-methods as.list as.list,gpu.matrix.tensorflow-method as.list,gpu.matrix.torch-method as.list-methods as.matrix as.matrix,gpu.matrix.tensorflow-method as.matrix,gpu.matrix.torch-method as.matrix-methods as.numeric as.numeric,gpu.matrix.tensorflow-method as.numeric,gpu.matrix.torch-method as.numeric-methods as.vector as.vector,gpu.matrix.tensorflow-method as.vector,gpu.matrix.torch-method as.vector-methods is.numeric is.numeric,gpu.matrix.tensorflow-method is.numeric,gpu.matrix.torch-method is.numeric-methods |
| cbind_rbind_methods | cbind2 cbind2,ANY,gpu.matrix.tensorflow-method cbind2,ANY,gpu.matrix.torch-method cbind2,gpu.matrix.tensorflow,ANY-method cbind2,gpu.matrix.torch,ANY-method cbind2-methods rbind2 rbind2,ANY,gpu.matrix.tensorflow-method rbind2,ANY,gpu.matrix.torch-method rbind2,gpu.matrix.tensorflow,ANY-method rbind2,gpu.matrix.torch,ANY-method rbind2-methods |
| concatenate_gpu.matrix | c,gpu.matrix.tensorflow-method c,gpu.matrix.torch-method c,numMatrixLike-method c-methods |
| Correlation, Variance and Covariance for 'GPUmatrix' objects | cor cor,ANY,gpu.matrix.tensorflow,ANY,ANY-method cor,ANY,gpu.matrix.torch,ANY,ANY-method cor,gpu.matrix.tensorflow,ANY,ANY,ANY-method cor,gpu.matrix.tensorflow,ANY,missing,character-method cor,gpu.matrix.tensorflow,missing,ANY,ANY-method cor,gpu.matrix.tensorflow,missing,missing,character-method cor,gpu.matrix.torch,ANY,ANY,ANY-method cor,gpu.matrix.torch,ANY,missing,character-method cor,gpu.matrix.torch,missing,ANY,ANY-method cor,gpu.matrix.torch,missing,missing,character-method cor,gpu.matrix.torch,missing,missing,missing-method cor-methods cov cov,ANY,gpu.matrix.tensorflow-method cov,ANY,gpu.matrix.torch-method cov,gpu.matrix.tensorflow,ANY-method cov,gpu.matrix.tensorflow,missing-method cov,gpu.matrix.tensorflow-method cov,gpu.matrix.torch,ANY-method cov,gpu.matrix.torch,missing-method cov,gpu.matrix.torch-method cov-methods cov2cor cov2cor,gpu.matrix.tensorflow-method cov2cor,gpu.matrix.torch-method cov2cor-methods |
| Kernel Density Estimation and Histograms | density density,gpu.matrix.tensorflow-method density,gpu.matrix.torch-method density-methods hist hist,gpu.matrix.tensorflow-method hist,gpu.matrix.torch-method hist-methods |
| Calculate the Determinant of a 'GPUMatrix' | det det,gpu.matrix.tensorflow-method det,gpu.matrix.torch-method det-methods determinant determinant,gpu.matrix.tensorflow,logical-method determinant,gpu.matrix.tensorflow,missing-method determinant,gpu.matrix.torch,logical-method determinant,gpu.matrix.torch,missing-method determinant-methods |
| diag | diag diag,gpu.matrix.tensorflow-method diag,gpu.matrix.torch-method diag-methods diag<- diag<-,gpu.matrix.tensorflow,numeric-method diag<-,gpu.matrix.torch,numeric-method diag<--methods |
| Number of rows and columns and its corresponding names | colnames colnames,gpu.matrix.tensorflow-method colnames,gpu.matrix.torch-method colnames-methods dim dim,gpu.matrix.tensorflow-method dim,gpu.matrix.torch-method dim-methods dim<- dim<-,gpu.matrix.tensorflow,vector-method dim<-,gpu.matrix.torch,vector-method dim<--methods dimnames dimnames,gpu.matrix.tensorflow-method dimnames,gpu.matrix.torch-method dimnames-methods dimnames<- dimnames<-,gpu.matrix.tensorflow,vector-method dimnames<-,gpu.matrix.torch,vector-method dimnames<--methods length length,gpu.matrix.tensorflow-method length,gpu.matrix.torch-method length-methods ncol ncol,gpu.matrix.tensorflow-method ncol,gpu.matrix.torch-method ncol-methods nrow nrow,gpu.matrix.tensorflow-method nrow,gpu.matrix.torch-method nrow-methods rownames rownames,gpu.matrix.tensorflow-method rownames,gpu.matrix.torch-method rownames-methods |
| Distance Matrix Computation with GPU | dist dist,gpu.matrix.torch-method dist-methods |
| 'GPUmatrix' Exponential | expmGPU expmGPU,gpu.matrix.tensorflow-method expmGPU,gpu.matrix.torch-method expmGPU-methods |
| extract_gpu.matrix | - -,gpu.matrix.tensorflow,missing-method -,gpu.matrix.torch,missing-method --methods [ [,gpu.matrix.tensorflow,index,index-method [,gpu.matrix.tensorflow,index,missing-method [,gpu.matrix.tensorflow,matrix,missing-method [,gpu.matrix.tensorflow,missing,index-method [,gpu.matrix.torch,index,index-method [,gpu.matrix.torch,index,missing-method [,gpu.matrix.torch,matrix,missing-method [,gpu.matrix.torch,missing,index-method [-methods [<- [<-,gpu.matrix.tensorflow,index,index-method [<-,gpu.matrix.tensorflow,index,missing-method [<-,gpu.matrix.tensorflow,matrix,missing-method [<-,gpu.matrix.tensorflow,missing,index-method [<-,gpu.matrix.torch,index,index-method [<-,gpu.matrix.torch,index,missing-method [<-,gpu.matrix.torch,matrix,missing-method [<-,gpu.matrix.torch,missing,index-method [<--methods [[ [[,gpu.matrix.tensorflow,index-method [[,gpu.matrix.torch,index-method [[-methods [[<- [[<-,gpu.matrix.tensorflow,index-method [[<-,gpu.matrix.torch,index-method [[<--methods |
| Fast Discrete Fourier Transform (FFT) | fft fft,gpu.matrix.tensorflow,missing-method fft,gpu.matrix.tensorflow-method fft,gpu.matrix.torch,logical-method fft,gpu.matrix.torch,missing-method fft,gpu.matrix.torch-method fft-methods mvfft mvfft,gpu.matrix.tensorflow,missing-method mvfft,gpu.matrix.tensorflow-method mvfft,gpu.matrix.torch,logical-method mvfft,gpu.matrix.torch,missing-method mvfft,gpu.matrix.torch-method mvfft-methods |
| create and store a matrix in the GPU | as.gpu.matrix as.gpu.matrix,ANY-method as.gpu.matrix-methods gpu.matrix |
| Class 'gpu.matrix' for matrix stored in GPU | gpu.matrix-class |
| Fitting Generalized Linear Models using GPUmatrix objects | glm.fit.GPU GPUglm |
| installTorch | installTorch |
| kroneker Products | %x% %x%,ANY,gpu.matrix.tensorflow-method %x%,ANY,gpu.matrix.torch-method %x%,gpu.matrix.tensorflow,ANY-method %x%,gpu.matrix.torch,ANY-method %x%-methods |
| Logistic Regression with Conjugate Gradient method | LR_GradientConjugate_gpumatrix |
| Decomposition of a matrix with GPU | chol chol,gpu.matrix.tensorflow-method chol,gpu.matrix.torch-method chol-methods eigen eigen,gpu.matrix.tensorflow-method eigen,gpu.matrix.torch-method eigen-methods svd svd,gpu.matrix.tensorflow-method svd,gpu.matrix.torch-method svd-methods |
| Return the first or last part of a GPUmatrix object | head head,gpu.matrix.tensorflow-method head,gpu.matrix.torch-method head-methods print print,gpu.matrix.torch-method print-methods show show,gpu.matrix.tensorflow-method show,gpu.matrix.torch-method show-methods tail tail,gpu.matrix.tensorflow-method tail,gpu.matrix.torch-method tail-methods |
| Get different statistics for a gpu.matrix-class. | colMaxs colMaxs,gpu.matrix.tensorflow-method colMaxs,gpu.matrix.torch-method colMaxs-methods colMeans colMeans,gpu.matrix.tensorflow-method colMeans,gpu.matrix.torch-method colMeans-methods colMins colMins,gpu.matrix.tensorflow-method colMins,gpu.matrix.torch-method colMins-methods colRanks colRanks,gpu.matrix.tensorflow-method colRanks,gpu.matrix.torch-method colRanks-methods colSums colSums,gpu.matrix.tensorflow-method colSums,gpu.matrix.torch-method colSums-methods colVars colVars,gpu.matrix.tensorflow-method colVars,gpu.matrix.torch-method colVars-methods max,gpu.matrix.tensorflow-method max,gpu.matrix.torch-method max-methods mean mean,gpu.matrix.tensorflow-method mean,gpu.matrix.torch-method mean-methods min min,gpu.matrix.tensorflow-method min,gpu.matrix.torch-method min-methods rowMaxs rowMaxs,gpu.matrix.tensorflow-method rowMaxs,gpu.matrix.torch-method rowMaxs-methods rowMeans rowMeans,gpu.matrix.tensorflow-method rowMeans,gpu.matrix.torch-method rowMeans-methods rowMins rowMins,gpu.matrix.tensorflow-method rowMins,gpu.matrix.torch-method rowMins-methods rowRanks rowRanks,gpu.matrix.tensorflow-method rowRanks,gpu.matrix.torch-method rowRanks-methods rowSums rowSums,gpu.matrix.tensorflow-method rowSums,gpu.matrix.torch-method rowSums-methods rowVars rowVars,gpu.matrix.tensorflow-method rowVars,gpu.matrix.torch-method rowVars-methods sum sum,gpu.matrix.tensorflow-method sum,gpu.matrix.torch-method sum-methods which.max,gpu.matrix.tensorflow-method which.max,gpu.matrix.torch-method which.max-methods which.min which.min,gpu.matrix.tensorflow-method which.min,gpu.matrix.torch-method which.min-methods |
| Matrix Products | %*% %*%,ANY,gpu.matrix.tensorflow-method %*%,ANY,gpu.matrix.torch-method %*%,gpu.matrix.tensorflow,ANY-method %*%,gpu.matrix.torch,ANY-method %*%-methods crossprod crossprod,ANY,gpu.matrix.tensorflow-method crossprod,ANY,gpu.matrix.torch-method crossprod,gpu.matrix.tensorflow,ANY-method crossprod,gpu.matrix.tensorflow,missing-method crossprod,gpu.matrix.torch,ANY-method crossprod,gpu.matrix.torch,missing-method crossprod-methods tcrossprod tcrossprod,ANY,gpu.matrix.tensorflow-method tcrossprod,ANY,gpu.matrix.torch-method tcrossprod,gpu.matrix.tensorflow,ANY-method tcrossprod,gpu.matrix.tensorflow,missing-method tcrossprod,gpu.matrix.torch,ANY-method tcrossprod,gpu.matrix.torch,missing-method tcrossprod-methods |
| Non negative factorization of a matrix | NMFgpumatrix |
| Compute the kth power of a matrix. | %^% %^%,gpu.matrix.tensorflow,numeric-method %^%,gpu.matrix.torch,numeric-method |
| The QR Decomposition of a GPUmatrix object | qr qr,gpu.matrix.tensorflow-method qr,gpu.matrix.torch-method qr-methods qr.coef qr.coef,list-method qr.coef-methods qr.Q qr.Q,list-method qr.Q-methods qr.qty qr.qty,list-method qr.qty-methods qr.qy qr.qy,list-method qr.qy-methods qr.R qr.R,list-method qr.R-methods qr.resid qr.resid,list-method qr.resid-methods qr.solve qr.solve,ANY,gpu.matrix.tensorflow-method qr.solve,ANY,gpu.matrix.torch-method qr.solve,gpu.matrix.tensorflow,ANY-method qr.solve,gpu.matrix.tensorflow,gpu.matrix.tensorflow-method qr.solve,gpu.matrix.torch,ANY-method qr.solve,gpu.matrix.torch,gpu.matrix.torch-method qr.solve,list,ANY-method qr.solve-methods qr.X qr.X,list-method qr.X-methods |
| rounding of numers | round round,gpu.matrix.tensorflow,ANY-method round,gpu.matrix.tensorflow,missing-method round,gpu.matrix.tensorflow,numeric-method round,gpu.matrix.torch,missing-method round,gpu.matrix.torch,numeric-method round-methods |
| Solve a System of Equations | chol_solve chol_solve,ANY,gpu.matrix.tensorflow-method chol_solve,ANY,gpu.matrix.torch-method chol_solve,gpu.matrix.tensorflow,ANY-method chol_solve,gpu.matrix.torch,ANY-method chol_solve-methods ginv ginv,gpu.matrix.tensorflow-method ginv,gpu.matrix.torch-method ginv-methods solve solve,ANY,gpu.matrix.tensorflow-method solve,ANY,gpu.matrix.torch-method solve,gpu.matrix.tensorflow,ANY-method solve,gpu.matrix.tensorflow,missing-method solve,gpu.matrix.torch,ANY-method solve,gpu.matrix.torch,missing-method solve-methods |
| sort | sort sort,gpu.matrix.tensorflow,logical-method sort,gpu.matrix.tensorflow-method sort,gpu.matrix.torch,logical-method sort,gpu.matrix.torch-method sort-methods |
| Spicify type of 'GPUmatrix' | dtype dtype,gpu.matrix.tensorflow-method dtype,gpu.matrix.torch-method dtype-methods dtype<- dtype<-,gpu.matrix.tensorflow-method dtype<-,gpu.matrix.torch-method dtype<--methods to_dense to_dense,gpu.matrix.tensorflow-method to_dense,gpu.matrix.torch-method to_dense-methods to_sparse to_sparse,gpu.matrix.tensorflow-method to_sparse,gpu.matrix.torch-method to_sparse-methods |
