output_dir: "data"

project:
    directory         : data
    file_extension    : csv
    submission_file   :
    submit_probas     : False

data:
    drop              : ['date', 'tag', 'open', 'high', 'low', 'close', 'volume', 'adjclose',
                         'low[1]', 'high[1]', 'net', 'close[1]', 'rmin_3', 'rmin_4', 'rmin_5',
                         'rmin_7', 'rmin_8', 'rmin_10', 'rmin_18', 'pval', 'mval', 'vma',
                         'rmax_2', 'rmax_3', 'rmax_4', 'rmax_5', 'rmax_6', 'rmax_7', 'rmax_10']
    features          : '*'
    sampling          :
        option        : True
        method        : under_random
        ratio         : 0.5
    sentinel          : -1
    separator         : ','
    shuffle           : True
    split             : 0.4
    target            : rrover
    target_value      : True

model:
    algorithms        : ['RF']
    balance_classes   : True
    calibration       :
        option        : False
        type          : isotonic
    cv_folds          : 3
    estimators        : 501
    feature_selection :
        option        : True
        percentage    : 50
        uni_grid      : [5, 10, 15, 20, 25]
        score_func    : f_classif
    grid_search       :
        option        : False
        iterations    : 100
        random        : True
        subsample     : True
        sampling_pct  : 0.25
    pvalue_level      : 0.01
    rfe               :
        option        : True
        step          : 10
    scoring_function  : 'roc_auc'
    type              : classification

features:
    clustering        :
        option        : False
        increment     : 3
        maximum       : 30
        minimum       : 3
    counts            :
        option        : False
    encoding          :
        rounding      : 3
        type          : factorize
    factors           : []
    interactions      :
        option        : True
        poly_degree   : 2
        sampling_pct  : 5
    isomap            :
        option        : False
        components    : 2
        neighbors     : 5
    logtransform      :
        option        : False
    numpy             :
        option        : False
    pca               :
        option        : False
        increment     : 3
        maximum       : 15
        minimum       : 3
        whiten        : False
    scaling           :
        option        : True
        type          : standard
    scipy             :
        option        : False
    text              :
        ngrams        : 1
        vectorize     : False
    tsne              :
        option        : False
        components    : 2
        learning_rate : 1000.0
        perplexity    : 30.0
    variance          :
        option        : True
        threshold     : 0.1

treatments:
    doji              : ['alphapy.features', 'runs_test', ['all'], 18]
    hc                : ['alphapy.features', 'runs_test', ['all'], 18]
    hh                : ['alphapy.features', 'runs_test', ['all'], 18]
    hl                : ['alphapy.features', 'runs_test', ['all'], 18]
    ho                : ['alphapy.features', 'runs_test', ['all'], 18]
    rrhigh            : ['alphapy.features', 'runs_test', ['all'], 18]
    rrlow             : ['alphapy.features', 'runs_test', ['all'], 18]
    rrover            : ['alphapy.features', 'runs_test', ['all'], 18]
    rrunder           : ['alphapy.features', 'runs_test', ['all'], 18]
    sephigh           : ['alphapy.features', 'runs_test', ['all'], 18]
    seplow            : ['alphapy.features', 'runs_test', ['all'], 18]
    trend             : ['alphapy.features', 'runs_test', ['all'], 18]

pipeline:
    number_jobs       : -1
    seed              : 10231
    verbosity         : 0

plots:
    calibration       : True
    confusion_matrix  : True
    importances       : True
    learning_curve    : True
    roc_curve         : True

xgboost:
    stopping_rounds   : 20