plot_prediction_samples: Plot multiple splits from list with forecast results

In thfuchs/tsRNN: Time Series Forecasting using Recurrent Neural Networks by Keras

Description

Plot results from cv_arima (by ggplot2)

Usage

plot_prediction_samples(
  splits,
  col_date = "index",
  col_value = "value",
  col_group = "key",
  title = NULL,
  date_type = "datetime",
  ncol = 3L,
  scale = NULL,
  PI = FALSE,
  col_pi_high = "hi95",
  col_pi_low = "lo95"
)

Arguments

splits	list of prediction data.frames
col_date	Date column, default to "index"
col_value	Value column, default to "value"
col_group	Group column , default to "key" ("actual" and "predict") - see example
title	diagram title
date_type	string vector - one of "datetime", "date" or "character"
ncol	Number of columns
scale	NULL or two Date values to scale x-axis
PI	add prediction interval?
col_pi_high	Upper confidence bound
col_pi_low	Lower confidence bound

Value

ggplot2 object

Examples

data <- tsRNN::fc_arima
plot_prediction_samples(
  splits = list(data, data),
  ncol = 2L
)

# with Prediction Interval
plot_prediction_samples(
  splits = list(data, data),
  ncol = 2L,
  PI = TRUE
)

thfuchs/tsRNN documentation built on April 17, 2021, 11:03 p.m.