initializeGenerators: Initializes generators defined by labelByFolderGenerator...

In hiddengenome/altum: deepG

Usage

initializeGenerators(
  directories,
  format = "fasta",
  batch.size = 256,
  maxlen = 250,
  max_iter = 10000,
  vocabulary = c("a", "c", "g", "t"),
  verbose = FALSE,
  randomFiles = FALSE,
  step = 1,
  showWarnings = FALSE,
  seed = 1234,
  shuffleFastaEntries = FALSE,
  numberOfFiles = NULL,
  fileLog = NULL,
  reverseComplements = FALSE,
  val = FALSE
)

Arguments

directories	Vector of paths to folder containing fasta files. Files in one folder should belong to one class.
format	File format.
batch.size	Number of batches, will get rounded to be multiple of number of targets if necessary.
maxlen	Length of predictor sequence.
max_iter	Stop after max_iter number of iterations failed to produce a new batch.
vocabulary	Vector of allowed characters, character outside vocabulary get encoded as 0-vector.
verbose	Whether to show message.
randomFiles	Logical, whether to go through files randomly or sequential.
step	How often to take a sample.
showWarnings	Logical, give warning if character outside vocabulary appears.
seed	Sets seed for set.seed function, for reproducible results when using randomFiles or shuffleFastaEntries
shuffleFastaEntries	Logical, shuffle fasta entries.
numberOfFiles	Use only specified number of files, ignored if greater than number of files in directories.
fileLog	Write name of files to csv file if path is specified.
reverseComplements	Logical, half of batch contains sequences and other its reverse complements. Reverse complement is given by reversed order of sequence and switching A/T and C/G. batch.size argument has to be even, otherwise 1 will be added to batch.size
val	Logical, call initialized generarator "genY" or "genValY" where Y is an integer between 1 and length of directories.

hiddengenome/altum documentation built on April 22, 2020, 9:33 p.m.