tensor size errors for bce_with_logits loss / net which outputs 1d tensor #373
Comments
|
Thanks for this question! There are in principle two solutions:
library(mlr3torch)
#> Loading required package: mlr3
#> Loading required package: mlr3pipelines
#> Loading required package: torch
# currently requires adding class "nn_loss" because as_torch_loss only has a method for this class
nn_bce_loss2 = nn_module(c("nn_bce_with_logits_loss2", "nn_loss"),
initialize = function(weight = NULL, reduction = "mean", pos_weight = NULL) {
self$loss = nn_bce_with_logits_loss(weight, reduction, pos_weight)
},
forward = function(input, target) {
# EDIT: added the -1 after comment by tdhock
self$loss(input$reshape(-1), target$to(dtype = torch_float()) - 1)
}
)
loss = nn_bce_loss2()
loss(torch_randn(10, 1), torch_randint(0, 1, 10))
#> torch_tensor
#> 0.500878
#> [ CPUFloatType{} ]
task = tsk("sonar")
graph = po("torch_ingress_num") %>>%
nn("linear", out_features = 1) %>>%
po("torch_loss", loss = nn_bce_loss2) %>>%
po("torch_optimizer") %>>%
po("torch_model_classif",
epochs = 1, batch_size = 32
)
glrn = as_learner(graph)
glrn$train(task)
# this does not work, because the targets are still loaded as ints but bce_loss expects floats
graph2 = po("torch_ingress_num") %>>%
nn("linear", out_features = 1) %>>%
nn("flatten", start_dim = 1) %>>%
po("torch_loss", loss = nn_bce_loss) %>>%
po("torch_optimizer") %>>%
po("torch_model_classif",
epochs = 1, batch_size = 32
)
glrn2 = as_learner(graph2)
glrn2$train(task)
#> A lit of torch error backtrace
#> This happened PipeOp torch_model_classif's $train()Created on 2025-04-02 with reprex v2.1.1 |
|
I confirm this works for me, thanks for the advice! |
|
Keeping this open as this is at least a TODO for documentation |
|
by the way, for others who are trying to do this, the code above runs without error, but does not learn, because the bce_with_logits_loss needs values in 0,1, but R labels are 1,2. nn_bce_loss3 = nn_module(c("nn_bce_with_logits_loss3", "nn_loss"),
initialize = function(weight = NULL, reduction = "mean", pos_weight = NULL) {
self$loss = nn_bce_with_logits_loss(weight, reduction, pos_weight)
},
forward = function(input, target) {
self$loss(input$reshape(-1), target$to(dtype = torch_float())-1)
}
) |
|
Thanks! I updated the code snippet |
|
This is now solved as torch classification learners are expected to output 1 column for binary classification problems |
Hi! @sebffischer
I am trying to implement a neural network with a custom loss function for binary classification.
It works just like the standard
torch::nn_bce_with_logits_loss-- https://torch.mlverse.org/docs/reference/nn_bce_with_logits_losscriterion(output, target)where both output and target are 1d tensors.Here is a MRE.
I tried the code above, and I observe a tensor size error,
in the output above we see that
loss=torch::nn_cross_entropy_lossworks without error, butloss=torch::nn_bce_with_logits_lossgives a tensor size error.I guess this is because MLP learner is always outputing a matrix with 2 columns? (even for the binary case where it could output 1 column or just a 1d tensor instead)
Is it possible in mlr3torch to have a neural network that outputs a 1d tensor instead?
I tried using graph learner code below, with nn_linear layer that has
out_features=1.I observe two different errors for the two different loss functions:
So I guess this means that it is not currently supported?
for positive control, I tried changing
out_features=2in the code above, and in that case I observe the same result as MLP learner (cross entropy loss works, error for bce_with_logits_loss).The text was updated successfully, but these errors were encountered: