Skip to contents

Smooth the signal and remove the hooks

Source: R/df_ccsig.R
df_ccsig.Rd

This function smooths the signal and removes the hooks by running a t-test in the boundaries to see if there are large 'hooks'.

Usage

df_ccsig(
  sig_df,
  span1 = 400,
  span2 = 40,
  breaks_q = c(0, 0.025, 0.05, 0.95, 0.975, 1),
  outlier_cycles = 1,
  ifplot = FALSE
)

Arguments

sig_df

A signal data frame with two columns: x and sig.

span1

A positive parameter to control the degree of smoothing in bullterxtrctr::cc_get_signature.

span2

A positive parameter to control the degree of smoothing in bullterxtrctr::cc_get_signature.

breaks_q

A vector of quantiles to split the x values into five groups for t-test of large hooks.

outlier_cycles

a positive integer for the number of times we check for outliers along the boundaries.

ifplot

A Boolean flag indicating whether to save ggplot lists in the output attributes.

Value

A data frame with two columns:

  • x: The x values from the x3p object.

  • sig: The extracted signal.

Examples

x3p <- x3p_subsamples[[2]]
insidepoly_df <- x3p_insidepoly_df(x3p, mask_col = "#FF0000", concavity = 1.5, b = 1)
x3p_inner_nomiss_res <- df_rmtrend_x3p(insidepoly_df)
x3p_inner_impute <- x3p_impute(x3p_inner_nomiss_res,
  ifout = FALSE, ifsave = FALSE, dir_name = NULL, ifplot = FALSE
)
x3p_bin_rotate <- x3p_vertical(x3p_inner_impute, min_score_cut = 0.1, ifplot = FALSE)
raw_sig_df <- x3p_raw_sig_df(x3p_bin_rotate, ifplot = FALSE)

raw_ccsig_df <- df_ccsig(raw_sig_df, ifplot = TRUE)
#> Warning: k-d tree limited by memory. ncmax= 207

attr(raw_ccsig_df, "sig_df_plot")