torch
Tensors and Neural Networks with 'GPU' Acceleration
Provides functionality to define and train neural networks similar to
'PyTorch' by Paszke et al (2019)
torch 
Installation
torch can be installed from CRAN with:
install.packages("torch")
You can also install the development version with:
remotes::install_github("mlverse/torch")
At the first package load additional software will be installed. See also the full installation guide here.
Examples
You can create torch tensors from R objects with the torch_tensor
function and convert them back to R objects with as_array.
library(torch)
x <- array(runif(8), dim = c(2, 2, 2))
y <- torch_tensor(x, dtype = torch_float64())
y
#> torch_tensor
#> (1,.,.) =
#> 0.6192 0.5800
#> 0.2488 0.3681
#>
#> (2,.,.) =
#> 0.0042 0.9206
#> 0.4388 0.5664
#> [ CPUDoubleType{2,2,2} ]
identical(x, as_array(y))
#> [1] TRUE
Simple Autograd Example
In the following snippet we let torch, using the autograd feature, calculate the derivatives:
x <- torch_tensor(1, requires_grad = TRUE)
w <- torch_tensor(2, requires_grad = TRUE)
b <- torch_tensor(3, requires_grad = TRUE)
y <- w * x + b
y$backward()
x$grad
#> torch_tensor
#> 2
#> [ CPUFloatType{1} ]
w$grad
#> torch_tensor
#> 1
#> [ CPUFloatType{1} ]
b$grad
#> torch_tensor
#> 1
#> [ CPUFloatType{1} ]
Contributing
No matter your current skills it’s possible to contribute to torch
development. See the contributing
guide for more
information.