Package: gpuMagic 1.21.0
Jiefei Wang
gpuMagic: An openCL compiler with the capacity to compile R functions and run the code on GPU
The package aims to help users write openCL code with little or no effort. It is able to compile an user-defined R function and run it on a device such as a CPU or a GPU. The user can also write and run their openCL code directly by calling .kernel function.
Authors:
gpuMagic_1.21.0.tar.gz
gpuMagic_1.21.0.zip(r-4.5)gpuMagic_1.21.0.zip(r-4.4)gpuMagic_1.21.0.zip(r-4.3)
gpuMagic_1.21.0.tgz(r-4.4-arm64)gpuMagic_1.21.0.tgz(r-4.4-x86_64)gpuMagic_1.21.0.tgz(r-4.3-arm64)gpuMagic_1.21.0.tgz(r-4.3-x86_64)
gpuMagic_1.21.0.tar.gz(r-4.5-noble)gpuMagic_1.21.0.tar.gz(r-4.4-noble)
gpuMagic.pdf |gpuMagic.html✨
gpuMagic/json (API)
NEWS
| # Install 'gpuMagic' in R: |
| install.packages('gpuMagic', repos = c('https://bioc.r-universe.dev', 'https://cloud.r-project.org')) |
Bug tracker:https://github.com/jiefei-wang/gpumagic/issues
On BioConductor:gpuMagic-1.21.0(bioc 3.20)gpuMagic-1.20.0(bioc 3.19)
Last updated 2 months agofrom:ed349fdd53
Exports:.kernelcompileGPUCodedownloadextractVarsgetCurDevicegetDeviceIndexgetDeviceInfogetDeviceListgetJobStatusgetSizegpu_cast_doublegpu_cast_floatgpu_cast_intgpu_cast_longgpu_cast_uintgpu_cast_ulonggpuEmptMatrixgpuMagic.getAvailableTypegpuMagic.getMemUsagegpuMagic.getOptionsgpuMagic.setOptionsgpuMatrixgpuSapplygpuSapply.getOptionisgreaterkernel.getOptionMatrixncolnrowreturn_nocpyScalarsetDevicesubReft_nocpyupload
Dependencies:askpassBiocGenericsbootcellrangerclassclicodetoolscpp11crayoncurldata.tableDerivDescToolsdigeste1071ExactexpmfansigldgluehmshttrjsonlitelatticelifecyclelmomlobstrmagrittrMASSMatrixmimemvtnormopensslpillarpkgconfigprettyunitsprogressproxypryrR6RcppreadxlrematchrlangrootSolverstudioapistringistringrsystibbleutf8vctrswithr
Readme and manuals
Help Manual
| Help page | Topics |
|---|---|
| Excute the openCL function | .kernel |
| extract/set parts of the data in gpuMatrix object | [ [,gpuMatrix,ANY,ANY,missing-method [,gpuMatrix-method [<-,gpuMatrix,ANY,ANY,numeric-method |
| Convert the gpuMatrix object into a matrix | as.matrix.gpuMatrix |
| Convert the gpuMatrix object into a vector | as.vector.gpuMatrix |
| Compile the R function without excute it in the device. | compileGPUCode |
| Query and select the devices | getCurDevice getDeviceIndex getDeviceInfo getDeviceList getJobStatus setDevice |
| Internal usage only, the package export this function only for the other package to access. | extractVars extractVars.default extractVars.expression extractVars.extCode extractVars.varInfo gpu_cast_double gpu_cast_float gpu_cast_int gpu_cast_long gpu_cast_uint gpu_cast_ulong isgreater |
| Get all the available openCL variable type | gpuMagic.getAvailableType |
| Get the device memory usage | gpuMagic.getMemUsage |
| Get the openCL options | gpuMagic.getOptions |
| Set the openCL options | gpuMagic.setOptions |
| gpuMatrix class | dim dim,gpuMatrix-method download download,ANY-method download,gpuMatrix-method getSize getSize,gpuMatrix-method gpuEmptMatrix gpuMatrix length length,gpuMatrix-method nrow nrow,gpuMatrix-method upload upload,gpuMatrix-method |
| A GPU version of the sapply function | gpuSapply |
| Get the package compilation options | gpuSapply.getOption |
| Get the openCL compilation options | kernel.getOption |
| Create a matrix | Matrix |
| gpuMatrix class | ncol ncol,gpuMatrix-method |
| Print the available options in a pretty format | print.deviceList print.options print.plainText print.varInfo |
| No copy method | return_nocpy t_nocpy |
| Create a scalar variable | Scalar |
| Get a reference of the subset of a matrix | subRef |
