LBBNN (version 0.1.2)

get_dataloaders: Wrapper around torch::dataloader

Description

Avoids users having to manually define their own dataloaders.

Usage

get_dataloaders(
  dataset,
  train_proportion,
  train_batch_size,
  test_batch_size,
  standardize = TRUE,
  shuffle_train = TRUE,
  shuffle_test = FALSE,
  seed = 1
)

Value

A list containing:

train_loader

A torch::dataloader for the training data.

test_loader

A torch::dataloader for the test data.

Arguments

dataset

A data.frame. The last column is assumed to be the dependent variable.

train_proportion

numeric, between 0 and 1. Proportion of data to be used for training.

train_batch_size

integer, number of samples per batch in the training dataloader.

test_batch_size

integer, number of sampels per batch in the testing dataloader.

standardize

logical, whether to standardize input-features, default is TRUE.

shuffle_train

logical, whether to shuffle the training data each epoch. default is TRUE

shuffle_test

logical, shuffle test data, default is FALSE. Usually not needed.

seed

integer. Used for reproducibility purposes in the train/test split.