as_gpu_matrix

Tags a matrix as GPU-backed. In an expression like
<code>as_gpu_matrix(A) %*% B</code>, the multiply dispatches to
<code>ggml_matmul</code> (Vulkan GPU with CPU fallback) instead of R's
built-in. Only one operand needs wrapping. <code><a href="/link/crossprod()?package=ggmlR&version=0.8.1" data-mini-rdoc="ggmlR::crossprod()">crossprod()</a></code> and
<code><a href="/link/tcrossprod()?package=ggmlR&version=0.8.1" data-mini-rdoc="ggmlR::tcrossprod()">tcrossprod()</a></code> on a wrapped matrix likewise use the GPU.

Provides 'R' bindings to the 'GGML' tensor library for machine
learning, optimized for 'Vulkan' GPU acceleration with a transparent CPU
fallback. The package features a 'Keras'-like sequential API and a
'PyTorch'-style 'autograd' engine for building, training, and deploying
neural networks. Key capabilities include high-performance 5D tensor
operations, 'f16' precision, and efficient quantization. It supports
native 'ONNX' model import (50+ operators) and 'GGUF' weight loading
from the 'llama.cpp' and 'Hugging Face' ecosystems. Designed for
zero-overhead inference via dedicated weight buffering, it integrates
seamlessly as a 'parsnip' engine for 'tidymodels' and provides
first-class learners for the 'mlr3' framework.
See <https://github.com/ggml-org/ggml> for more information about the
underlying library.

Yuri Baramykov

ggmlR

'GGML' Tensor Operations for Machine Learning

as_gpu_matrix function

<dl><dt>x</dt>
<dd>A numeric matrix (or something coercible with <code>as.matrix</code>).</dd>
<dt>device</dt>
<dd><code>"auto"</code> (default), <code>"gpu"</code> or <code>"cpu"</code>; see
<code>ggml_matmul</code>.</dd>
<dt>prec</dt>
<dd><code>"f32"</code> (default) or <code>"f16"</code>; see <code>ggml_matmul</code>.</dd></dl>

Arguments

Tags a matrix as GPU-backed. In an expression like
<code>as_gpu_matrix(A) %*% B</code>, the multiply dispatches to
<code>ggml_matmul</code> (Vulkan GPU with CPU fallback) instead of R's
built-in. Only one operand needs wrapping. <code><a href='https://rdrr.io/r/base/crossprod.html'>crossprod()</a></code> and
<code><a href='https://rdrr.io/r/base/crossprod.html'>tcrossprod()</a></code> on a wrapped matrix likewise use the GPU.

Wrap a matrix so multiplies run on the GPU — as_gpu_matrix

<dl>

<dt>x</dt>
<dd>A numeric matrix (or something coercible with <code>as.matrix</code>).</dd>


<dt>device</dt>
<dd><code>"auto"</code> (default), <code>"gpu"</code> or <code>"cpu"</code>; see
<code>ggml_matmul</code>.</dd>


<dt>prec</dt>
<dd><code>"f32"</code> (default) or <code>"f16"</code>; see <code>ggml_matmul</code>.</dd>

</dl>

as_gpu_matrix: Wrap a matrix so multiplies run on the GPU

Description

Usage

Value

Arguments

Examples