Deborah.RebekahMiriam.PXvsBSPlotterRebekahMiriam

Deborah.RebekahMiriam.PXvsBSPlotterRebekahMiriam.build_flat_plot_dict — Method

build_flat_plot_dict(
    key::Symbol,
    orig_tag::Symbol,
    pred_tag1::Symbol,
    pred_tag2::Symbol,
    keyword::String,
    new_dict::Dict{Tuple{Symbol, Symbol, Symbol, String}, Array{Float64,2}}
) -> Dict{String, Array{Float64,2}}

Construct a flattened dictionary of average and error matrices for plotting, given a specific keyword and observable key.

This utility reformats selected entries from a 4-key dictionary new_dict—keyed by (observable, kind, tag, keyword)—into a flat Dict{String, Array} that mimics the naming convention used in Deborah.Rebekah.PXvsBSPlotter (e.g., "kurt:P_P2:avg").

It includes data from two prediction methods (pred_tag1, pred_tag2) as well as the original reference tag (orig_tag) for the specified observable key and criterion keyword.

Key naming convention in the returned dictionary:

Predictions: "$key:Y_<pred_tag>:avg" and "$key:Y_<pred_tag>:err"
Original: "$key:Y:<orig_tag>:avg" and "$key:Y:<orig_tag>:err" (Note the underscore for predictions vs colon for the original tag, matching Deborah.Rebekah.PXvsBSPlotter usage.)

Arguments

key: Observable symbol (e.g., :kurt).
orig_tag: Original/reference tag to include (previously hard-coded as :RWBS).
pred_tag1: First prediction tag to include (e.g., :RWP1).
pred_tag2: Second prediction tag to include (e.g., :RWP2).
keyword: Selector used in new_dict (e.g., "susp", "skew", "kurt" as interpolation criterion; or a $\kappa$ token like "13580" in measurement flows).
new_dict: Source dictionary keyed by (key, kind, tag, keyword) and valued with labels $\times$ trains matrices.

Returns

A flattened dictionary with keys like "kurt:Y_P2:avg" or "kurt:Y:RWBS:err" mapped to matrices, suitable for direct use in plotting.

source

Deborah.RebekahMiriam.PXvsBSPlotterRebekahMiriam.plot_PX_BS_vs_labels — Method

plot_PX_BS_vs_labels(
    key::String,
    suffix_orig::String,
    suffix1::String,
    suffix2::String,
    train_pct::Int,
    new_dict::Dict{String, Array{Float64,2}},
    labels_int::Vector{Int},
    trains_ext_int::Vector{Int},
    keyword::String,
    overall_name::String,
    figs_dir::String;
    key_tex::String = "",
    save_file::Bool = false
) -> Nothing

Plot P1, P2, and original values versus labeled set percentage for a fixed training set size, in a Deborah.Miriam analysis.

This function is specific to the Deborah.RebekahMiriam. It visualizes the behavior of a given observable (key) across varying label set sizes (labels_int), while fixing the training set percentage to train_pct.

The plotted values are extracted from new_dict, containing average and error arrays for:

baseline method (suffix_orig)
first estimation method (suffix1)
second estimation method (suffix2)

Each curve is plotted with its own marker style and color, and slightly shifted on the $x$-axis to improve readability. The $y$-axis reflects the observable evaluated at a specific interpolation point $\kappa_t$ determined by keyword.

Arguments

key::String: Name of the observable to be plotted (e.g., "skew", "kurt").
suffix_orig::String: Suffix used in the key for baseline values (e.g., "RWBS").
suffix1::String: Suffix for the first interpolation method (e.g., "RWP1").
suffix2::String: Suffix for the second interpolation method (e.g., "RWP2").
train_pct::Int: The training set percentage (used to select the column index from the 2D arrays).
new_dict::Dict{String, Array{Float64,2}}: Dictionary containing all relevant observable averages and error bars.
labels_int::Vector{Int}: Label set percentages to use for the $x$-axis.
trains_ext_int::Vector{Int}: Training set percentages, used to resolve the index for train_pct.
keyword::String: The origin cumulant that was used to determine the interpolation point $\kappa_t$. For example, "kurt" means that $\kappa_t$ was selected based on kurtosis behavior, and the observable in key was evaluated at that point.
overall_name::String: Suffix used for output filenames.
figs_dir::String: Output directory to save the figure.

Keyword Arguments

key_tex::String = "": Optional $\LaTeX$ string used as $y$-axis label.
save_file::Bool = false: If true, the figure is saved as a PDF and cropped using pdfcrop (if available).

Side Effects

Displays a PyPlot.jl figure comparing three estimation methods at the specified train_pct.
If save_file=true, saves a PDF named plot_<key>_<keyword>_<overall_name>_TRP_<train_pct>.pdf into figs_dir.

source

Deborah.RebekahMiriam.PXvsBSPlotterRebekahMiriam.plot_PX_BS_vs_labels_for_measurements — Method

plot_PX_BS_vs_labels_for_measurements(
    key::String,
    suffix_orig::String,
    suffix1::String,
    suffix2::String,
    train_pct::Int,
    new_dict::Dict{String, Array{Float64,2}},
    labels_int::Vector{Int},
    trains_ext_int::Vector{Int},
    keyword::String,
    overall_name::String,
    figs_dir::String;
    key_tex::String = "",
    save_file::Bool = false
) -> Nothing

Plot P1, P2, and original values versus labeled-set percentage at a fixed training-set size, for the single-ensemble-measurement-based workflow.

This variant mirrors plot_PX_BS_vs_labels but interprets keyword as a kappa token string (e.g., "13580"). The plot title shows $\kappa =$ 0.<keyword> (e.g., $\kappa = 0.13580$), indicating that all series are evaluated at that specific measurement point rather than an interpolation-derived $\kappa_t$.

The function visualizes a selected observable key across labels_int, while fixing the training-set percentage to train_pct. Three curves are drawn:

baseline/original (suffix_orig),
first estimation method (suffix1),
second estimation method (suffix2).

Values are fetched from new_dict, which stores average and error matrices indexed by (label_index, train_index) behind flattened keys:

"$key:Y_$suffix:avg" / "$key:Y_$suffix:err" for predictions,
"$key:Y:$suffix_orig:avg" / "$key:Y:$suffix_orig:err" for the original.

The special case key == "Deborah" omits the "$key:" prefix to match legacy naming.

Arguments

key::String: Observable to plot (e.g., "trM1", "Q2").
suffix_orig::String: Baseline/original tag suffix (e.g., "Y_BS").
suffix1::String: First method suffix (e.g., "Y_P1").
suffix2::String: Second method suffix (e.g., "Y_P2").
train_pct::Int: Training-set percentage; selects a column from the matrices.
new_dict::Dict{String, Array{Float64,2}}: Flattened dictionary containing avg/err arrays.
labels_int::Vector{Int}: Labeled-set percentages for the $x$-axis.
trains_ext_int::Vector{Int}: Training-set percentages; used to resolve train_pct.
keyword::String: $\kappa$ token for titling and output naming (e.g., "13580").
overall_name::String: Suffix used in the output filename.
figs_dir::String: Directory to save the figure.

Keyword Arguments

key_tex::String = "": Optional $\LaTeX$ $y$-axis label for the observable.
save_file::Bool = false: If true, saves plot_<key>_<keyword>_<overall_name>_TRP_<train_pct>.pdf and crops with pdfcrop if available.

Side Effects

Displays a PyPlot.jl figure comparing three methods at the specified train_pct.
Optionally writes a PDF to disk when save_file=true.

Notes

Assumes measurement summaries indexed by discrete $\kappa$ tokens; ensure new_dict was built accordingly.
Curves are slightly offset along $x$ for readability; error bars are shown for each series.

source

Deborah.RebekahMiriam.PXvsBSPlotterRebekahMiriam.plot_PX_BS_vs_trains — Method

plot_PX_BS_vs_trains(
    key::String,
    suffix_orig::String,
    suffix1::String,
    suffix2::String,
    label_pct::Int,
    new_dict::Dict{String, Array{Float64,2}},
    trains_ext_int::Vector{Int},
    labels_int::Vector{Int},
    keyword::String,
    overall_name::String,
    figs_dir::String;
    key_tex::String = "",
    save_file::Bool = false
) -> Nothing

Plot P1, P2, and original values versus training set percentage for a fixed label ratio, in a Deborah.Miriam analysis.

This function is specific to the Deborah.RebekahMiriam. It visualizes the behavior of a given observable (key) across varying training set sizes (trains_ext_int), while fixing the labeled set size to a specific value label_pct.

The plotted values are extracted from new_dict, containing average and error arrays for:

baseline method (suffix_orig)
first estimation method (suffix1)
second estimation method (suffix2)

Arguments

key::String: Name of the observable to be plotted (e.g., "skew", "kurt").
suffix_orig::String: Suffix used in the key for baseline values (e.g., "RWBS").
suffix1::String: Suffix for the first interpolation method (e.g., "RWP1").
suffix2::String: Suffix for the second interpolation method (e.g., "RWP2").
label_pct::Int: The label set percentage (used to select the row index from the 2D arrays).
new_dict::Dict{String, Array{Float64,2}}: Dictionary containing all relevant observable averages and error bars.
trains_ext_int::Vector{Int}: Training set percentages to use for the $x$-axis.
labels_int::Vector{Int}: Labeled set percentages, used to resolve the index for label_pct.
keyword::String: The origin cumulant that was used to determine the interpolation point $\kappa_t$. For example, "kurt" means that $\kappa_t$ was selected based on kurtosis behavior, and the observable in key was evaluated at that point.
overall_name::String: Suffix used for output filenames.
figs_dir::String: Output directory to save the figure.

Keyword Arguments

key_tex::String = "": Optional $\LaTeX$ string used as $y$-axis label.
save_file::Bool = false: If true, the figure is saved as a PDF and cropped using pdfcrop (if available).

Side Effects

Displays a PyPlot.jl figure comparing three estimation methods at the specified label_pct.
If save_file=true, saves a PDF named plot_<key>_<keyword>_<overall_name>_LBP_<label_pct>.pdf into figs_dir.

source

Deborah.RebekahMiriam.PXvsBSPlotterRebekahMiriam.plot_PX_BS_vs_trains_for_measurements — Method

plot_PX_BS_vs_trains_for_measurements(
    key::String,
    suffix_orig::String,
    suffix1::String,
    suffix2::String,
    label_pct::Int,
    new_dict::Dict{String, Array{Float64,2}},
    trains_ext_int::Vector{Int},
    labels_int::Vector{Int},
    keyword::String,
    overall_name::String,
    figs_dir::String;
    key_tex::String = "",
    save_file::Bool = false
) -> Nothing

Plot P1, P2, and original values versus training-set percentage at a fixed label ratio, for the single-ensemble-measurement-based workflow.

This variant mirrors plot_PX_BS_vs_trains but interprets keyword as a kappa token string (e.g., "13580"). The plot title explicitly shows $\kappa$ = 0.<keyword> (e.g., $\kappa =$ "0.13580"), indicating that all values are evaluated at that specific measurement point rather than an interpolation-derived $\kappa_t$.

The function visualizes a selected observable key across trains_ext_int, while holding the labeled-set size at label_pct. Three curves are shown:

baseline/original (suffix_orig),
first estimation method (suffix1),
second estimation method (suffix2).

Data are loaded from new_dict, which contains average and error matrices indexed by (label_index, train_index) using flattened keys such as:

"$key:Y_$suffix:avg" / "$key:Y_$suffix:err" for predictions,
"$key:Y:$suffix_orig:avg" / "$key:Y:$suffix_orig:err" for the original.

The special case key == "Deborah" omits the "$key:" prefix to match Deborah.DeborahCore naming.

Arguments

key::String: Observable name to plot (e.g., "trM1", "Q2").
suffix_orig::String: Baseline/original tag suffix (e.g., "Y_BS").
suffix1::String: First method suffix (e.g., "Y_P1").
suffix2::String: Second method suffix (e.g., "Y_P2").
label_pct::Int: Labeled-set percentage; selects the row from the matrices.
new_dict::Dict{String, Array{Float64,2}}: Flattened dictionary with avg/err arrays.
trains_ext_int::Vector{Int}: Training-set percentages for the $x$-axis.
labels_int::Vector{Int}: Available labeled-set percentages; used to resolve label_pct.
keyword::String: $\kappa$ token used for figure titling and output naming (e.g., "13580").
overall_name::String: Suffix used in the output filename.
figs_dir::String: Directory for saving the figure.

Keyword Arguments

key_tex::String = "": Optional $\LaTeX$ $y$-axis label for the observable.
save_file::Bool = false: If true, saves plot_<key>_<keyword>_<overall_name>_LBP_<label_pct>.pdf and crops with pdfcrop if available.

Side Effects

Displays a PyPlot.jl figure comparing three methods at the specified label_pct.
Optionally writes the figure to disk when save_file=true.

Notes

This function assumes the measurement pipeline where results are indexed by discrete $\kappa$ values. Ensure that new_dict was built accordingly (e.g., via build_flat_plot_dict-style utilities).
Curves are horizontally offset slightly to improve readability; error bars are drawn for each series.

source