visualizePrediction: This is a Shiny web application for the visualization of...

In hiddengenome/altum: deepG

Description

This is a Shiny web application for the visualization of hidden states of LSTM models from the package deepG. It also comes with the possibility to use own trained models.

Usage

visualizePrediction(
  sample = "",
  states = FALSE,
  model.path = "",
  fasta.path = "",
  states.path = "",
  start_position = 0,
  end_position = 800,
  batch.size = 200,
  vocabulary = c("l", "a", "g", "c", "t"),
  layer_depth = 1,
  cell_number = 1,
  step = 1,
  padding = TRUE
)

Arguments

sample	character input string of text
states	if states = TRUE, then load states from states_path
model.path	path to trained model
fasta.path	path to fasta files
states.path	path to the .h5 file with the calculated states
start_position	start from the dygraph
end_position	end from the dygraph
batch.size	number of samples to evaluate at once
vocabulary	used vocabulary
layer_depth	depth of layer to evaluate
cell_number	cell_number showed in the dygraph
step	frequency of sampling steps
padding	TRUE/FALSE generate states for the first maxlen nucleotides

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