Hi, I wrote some code to plot the results of some experiments the other day, and it worked fine, then I must have changed something because I am now getting a rather strange error:
ERROR: UndefVarError: select not defined
Stacktrace:
[1] extract_columns_and_names(::DataFrame, ::Symbol, ::Vararg{Any,N} where N) at /Users/harrisonwilde/.julia/packages/StatsPlots/Y1rGZ/src/df.jl:159
[2] (::var"#72#75"{Float64,Array{String,2},Array{String,1}})(::DataFrame) at ./none:0
[3] plot_real_α(::DataFrame, ::Float64, ::Array{String,2}, ::Array{String,1}, ::String) at /Users/harrisonwilde/Library/Mobile Documents/com~apple~CloudDocs/PhD/Synthetic Data/src/creditcard/plotting.jl:35
[4] top-level scope at none:0
This is after running this:
function plot_real_α(df, α, divergences, metric, t)
mkpath("src/creditcard/plots/$(t)/")
fdf = filter(row -> row[:real_α] == α, df)
print(fdf)
p = @df fdf plot(
:synth_α,
cols(Symbol("$(div)_$(metric)") for div in divergences),
title = "$(metric) divergence comparison, real alpha = $(α)",
label = [div for div in divergences],
xlabel = "Synthetic Alpha",
ylabel = metric,
legendtitle = "Divergence",
)
p = plot!(size=(1000, 700), legend=:outertopright)
png(p, "src/creditcard/plots/$(t)/real_alpha_$(α)__$(metric)")
end
plot_real_α(results, 0.1, ["beta" "weighted" "naive" "no_synth"], ["auc", "ll"], t)
Running select(df, :col)
works fine so I am not sure what the issue is here.
Thanks!