I am fitting an auto encoder but the lug::fit method is not working, and returns an unclear error message. Adding PYTORCH_ENABLE_MPS_FALLBACK=1 to the R Environment did not work
Epoch 1/100
Error in (function (x) : object 'delf' not found
In addition: Warning message:
Some torch operators might not yet be implemented for the MPS device. A temporary fix is to set the PYTORCH_ENABLE_MPS_FALLBACK=1 to
use the CPU as a fall back for those operators:
ℹ Add PYTORCH_ENABLE_MPS_FALLBACK=1 to your .Renviron file, for example use usethis::edit_r_environ().
✖ Using Sys.setenv() doesn't work because the env var must be set before R starts.
I am fitting an auto encoder but the lug::fit method is not working, and returns an unclear error message. Adding PYTORCH_ENABLE_MPS_FALLBACK=1 to the R Environment did not work
train_dl <- torch::dataloader( zip_train, batch_size = 10, shuffle = TRUE)
my_optimizer <- function(parameters){ torch::optim_adadelta(parameters, lr=1) }
after.setup <-luz::setup( autoencoder, loss = torch::nnf_mse_loss, optimizer = my_optimizer)
fit auto encoder
fitted <- luz::fit(after.setup, train_dl, epochs=100)
Error Message:
Epoch 1/100 Error in (function (x) : object 'delf' not found In addition: Warning message: Some torch operators might not yet be implemented for the MPS device. A temporary fix is to set the
PYTORCH_ENABLE_MPS_FALLBACK=1
to use the CPU as a fall back for those operators: ℹ AddPYTORCH_ENABLE_MPS_FALLBACK=1
to your.Renviron
file, for example useusethis::edit_r_environ()
. ✖ UsingSys.setenv()
doesn't work because the env var must be set before R starts.