Convert C matrix array to data.frame

Source: R/c_array_as_df.R

Converts the C history array from the mELO output to a data.frame. Useful for plotting diagnostics.

c_array_as_df(c_mat_array)

Arguments

c_mat_array

The history of C matrix values for a mELO rating model.

Value

A data.frame.

Examples

# rock paper scissors fire water model
rpsfw_model <- mELO(rpsfw_df, k=2)
c_mat_df <- c_array_as_df(rpsfw_model$c_mat_history)
head(c_mat_df)
#>   Player time      c_1      c_2      c_3       c_4
#> 1   FIRE    1 8.807558 4.656891 4.043954 -4.179362
#> 2   FIRE    2 8.807558 4.656891 4.043954 -4.179362
#> 3   FIRE    3 2.628271 9.329291 3.968360 -8.138085
#> 4   FIRE    4 2.628271 9.329291 3.968360 -8.138085
#> 5   FIRE    5 2.628271 9.329291 3.968360 -8.138085
#> 6   FIRE    6 2.628271 9.329291 3.968360 -8.138085

# Plot the evolution of C vectors for each throw
# Make data a bit 'tidyer'
tidy_c_mat_df <- tidyr::gather(
    c_mat_df,
    key = "C",
    value = "value",
    -time, -Player
    )
ggplot2::ggplot(tidy_c_mat_df, ggplot2::aes(x=time,y=value,colour=C)) +
    ggplot2::geom_line(size=1.25) +
    ggplot2::facet_wrap(~Player)