get_batch_constructor: Batch Constructor for a Task

In mlr3torch: Deep Learning with 'mlr3'

Batch Constructor for a Task

Description

Returns the function that builds a whole batch of a task, i.e. both the features x and the target y. This is how task_dataset loads its batches.

The returned function takes the arguments

• data, a data.table with the feature and target columns of the batch, and

• cache, a rather internal hashtab that can be used as a cache when loading multiple lazy_tensor columns.

It returns a list() with elements x (a named list() of torch_tensors) and y (a torch_tensor or NULL).

The default method applies the ingress tokens to obtain x and the target batchgetter to obtain y.

Usage

get_batch_constructor(
  task,
  feature_ingress_tokens,
  target_batchgetter = NULL,
  ...
)

Arguments

task	(Task) The task.
feature_ingress_tokens	(named list() of TorchIngressToken) The ingress tokens that define x. Their features are already resolved, i.e. they are character() vectors and not Selectors.
target_batchgetter	(⁠function()⁠ or NULL) Converts the target columns of a batch into the target tensor y that the loss is applied to. Takes an argument data, a data.table with only the target columns, and optionally an argument x, the named list of feature tensors of the batch, which is what a target that is a function of the input needs, see get_target_batchgetter(). If NULL (default), the batches have no y element.
...	(any) Additional arguments. Not used yet.

Value

⁠function(data, cache) -> list(x = list<torch_tensor>, y = torch_tensor | NULL)⁠

Examples

task = tsk("iris")
token = TorchIngressToken(task$feature_names, batchgetter_num, c(NA, 4))
# the token's features are a `Selector`; `task_dataset()` resolves them for you
token$features = token$features(task)
batch_constructor = get_batch_constructor(
  task,
  feature_ingress_tokens = list(input = token),
  target_batchgetter = get_target_batchgetter(task)
)
batch = batch_constructor(data = task$data(rows = 1:2))
batch$x$input
batch$y

mlr3torch documentation built on Oct. 30, 2026, 5:10 p.m.