3. Logistic classifier
# This file was generated by PredictMD version 0.31.0, code name Basophil
# For help, please visit https://predictmd.net

import PredictMDExtra
PredictMDExtra.import_all()

import PredictMD
PredictMD.import_all()

### Begin project-specific settings

LOCATION_OF_PREDICTMD_GENERATED_EXAMPLE_FILES = "/home/travis/build/bcbi/PredictMD.jl/docs/src/examples"

PROJECT_OUTPUT_DIRECTORY = joinpath(
    LOCATION_OF_PREDICTMD_GENERATED_EXAMPLE_FILES,
    "cpu_examples",
    "breast_cancer_biopsy",
    "output",
    )

mkpath(PROJECT_OUTPUT_DIRECTORY)
mkpath(joinpath(PROJECT_OUTPUT_DIRECTORY, "data"))
mkpath(joinpath(PROJECT_OUTPUT_DIRECTORY, "models"))
mkpath(joinpath(PROJECT_OUTPUT_DIRECTORY, "plots"))



### End project-specific settings

### Begin logistic classifier code

Random.seed!(999)

trainingandtuning_features_df_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "trainingandtuning_features_df.csv",
    )
trainingandtuning_labels_df_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "trainingandtuning_labels_df.csv",
    )
testing_features_df_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "testing_features_df.csv",
    )
testing_labels_df_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "testing_labels_df.csv",
    )
training_features_df_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "training_features_df.csv",
    )
training_labels_df_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "training_labels_df.csv",
    )
tuning_features_df_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "tuning_features_df.csv",
    )
tuning_labels_df_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "tuning_labels_df.csv",
    )
trainingandtuning_features_df = DataFrames.DataFrame(
    FileIO.load(
        trainingandtuning_features_df_filename;
        type_detect_rows = 100,
        )
    )
trainingandtuning_labels_df = DataFrames.DataFrame(
    FileIO.load(
        trainingandtuning_labels_df_filename;
        type_detect_rows = 100,
        )
    )
testing_features_df = DataFrames.DataFrame(
    FileIO.load(
        testing_features_df_filename;
        type_detect_rows = 100,
        )
    )
testing_labels_df = DataFrames.DataFrame(
    FileIO.load(
        testing_labels_df_filename;
        type_detect_rows = 100,
        )
    )
training_features_df = DataFrames.DataFrame(
    FileIO.load(
        training_features_df_filename;
        type_detect_rows = 100,
        )
    )
training_labels_df = DataFrames.DataFrame(
    FileIO.load(
        training_labels_df_filename;
        type_detect_rows = 100,
        )
    )
tuning_features_df = DataFrames.DataFrame(
    FileIO.load(
        tuning_features_df_filename;
        type_detect_rows = 100,
        )
    )
tuning_labels_df = DataFrames.DataFrame(
    FileIO.load(
        tuning_labels_df_filename;
        type_detect_rows = 100,
        )
    )

smoted_training_features_df_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "smoted_training_features_df.csv",
    )
smoted_training_labels_df_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "smoted_training_labels_df.csv",
    )
smoted_training_features_df = DataFrames.DataFrame(
    FileIO.load(
        smoted_training_features_df_filename;
        type_detect_rows = 100,
        )
    )
smoted_training_labels_df = DataFrames.DataFrame(
    FileIO.load(
        smoted_training_labels_df_filename;
        type_detect_rows = 100,
        )
    )

categorical_feature_names_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "categorical_feature_names.jld2",
    )
continuous_feature_names_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "data",
    "continuous_feature_names.jld2",
    )
categorical_feature_names = FileIO.load(
    categorical_feature_names_filename,
    "categorical_feature_names",
    )
continuous_feature_names = FileIO.load(
    continuous_feature_names_filename,
    "continuous_feature_names",
    )
feature_names = vcat(categorical_feature_names, continuous_feature_names)

single_label_name = :Class
negative_class = "benign"
positive_class = "malignant"
single_label_levels = [negative_class, positive_class]

categorical_label_names = Symbol[single_label_name]
continuous_label_names = Symbol[]
label_names = vcat(categorical_label_names, continuous_label_names)

feature_contrasts = PredictMD.generate_feature_contrasts(
    smoted_training_features_df,
    feature_names,
    )



show(
    PredictMD.linearly_dependent_columns(df)
    )

show(
    PredictMD.linearly_dependent_columns(
        training_features_df,
        feature_names,
        )
    )

logistic_classifier =
        PredictMD.singlelabelbinaryclassdataframelogisticclassifier(
        feature_names,
        single_label_name,
        single_label_levels;
        package = :GLM,
        intercept = true,
        interactions = 1,
        name = "Logistic regression",
        )

PredictMD.fit!(logistic_classifier,
               smoted_training_features_df,
               smoted_training_labels_df) # TODO: fix this error

PredictMD.get_underlying(logistic_classifier) # TODO: fix this error

logistic_hist_training =
        PredictMD.plotsinglelabelbinaryclassifierhistogram( # TODO: fix this error
        logistic_classifier,
        smoted_training_features_df,
        smoted_training_labels_df,
        single_label_name,
        single_label_levels,
        );



display(logistic_hist_training)
PredictMD.save_plot(
    joinpath(
        PROJECT_OUTPUT_DIRECTORY,
        "plots",
        "logistic_hist_training.pdf",
        ),
    logistic_hist_training,
    )

logistic_hist_testing =
    PredictMD.plotsinglelabelbinaryclassifierhistogram(
        logistic_classifier,
        testing_features_df,
        testing_labels_df,
        single_label_name,
        single_label_levels,
        );



display(logistic_hist_testing)
PredictMD.save_plot(
    joinpath(
        PROJECT_OUTPUT_DIRECTORY,
        "plots",
        "logistic_hist_testing.pdf",
        ),
    logistic_hist_testing,
    )

show(
    PredictMD.singlelabelbinaryclassificationmetrics(
        logistic_classifier,
        smoted_training_features_df,
        smoted_training_labels_df,
        single_label_name,
        positive_class;
        sensitivity = 0.95,
        );
    allrows = true,
    allcols = true,
    splitcols = false,
    )

show(
    PredictMD.singlelabelbinaryclassificationmetrics(
        logistic_classifier,
        testing_features_df,
        testing_labels_df,
        single_label_name,
        positive_class;
        sensitivity = 0.95,
        );
    allrows = true,
    allcols = true,
    splitcols = false,
    )

logistic_calibration_curve =
    PredictMD.plot_probability_calibration_curve(
        logistic_classifier,
        smoted_training_features_df,
        smoted_training_labels_df,
        single_label_name,
        positive_class;
        window = 0.2,
        );



display(logistic_calibration_curve)
PredictMD.save_plot(
    joinpath(
        PROJECT_OUTPUT_DIRECTORY,
        "plots",
        "logistic_calibration_curve.pdf",
        ),
    logistic_calibration_curve,
    )

show(
    PredictMD.probability_calibration_metrics(
        logistic_classifier,
        testing_features_df,
        testing_labels_df,
        single_label_name,
        positive_class;
        window = 0.1,
        );
    allrows = true,
    allcols = true,
    splitcols = false,
    )

logistic_cutoffs, logistic_risk_group_prevalences =
    PredictMD.risk_score_cutoff_values(
        logistic_classifier,
        testing_features_df,
        testing_labels_df,
        single_label_name,
        positive_class;
        average_function = Statistics.mean,
        )
@info(
    string(
        "Low risk: 0 to $(logistic_cutoffs[1]).",
        " Medium risk: $(logistic_cutoffs[1]) to $(logistic_cutoffs[2]).",
        " High risk: $(logistic_cutoffs[2]) to 1.",
        )
    )
@info(logistic_risk_group_prevalences)
logistic_cutoffs, logistic_risk_group_prevalences =
    PredictMD.risk_score_cutoff_values(
        logistic_classifier,
        testing_features_df,
        testing_labels_df,
        single_label_name,
        positive_class;
        average_function = Statistics.median,
        )
@info(
    string(
        "Low risk: 0 to $(logistic_cutoffs[1]).",
        " Medium risk: $(logistic_cutoffs[1]) to $(logistic_cutoffs[2]).",
        " High risk: $(logistic_cutoffs[2]) to 1.",
        )
    )
@info(logistic_risk_group_prevalences)

logistic_classifier_filename = joinpath(
    PROJECT_OUTPUT_DIRECTORY,
    "models",
    "logistic_classifier.jld2",
    )

PredictMD.save_model(logistic_classifier_filename, logistic_classifier)

### End logistic classifier code



# This file was generated by PredictMD version 0.31.0, code name Basophil
# For help, please visit https://predictmd.net

This page was generated using Literate.jl.