python source code of problem

Project: ramp-workflow (GitHub Link)

ramp-workflow-master
- .circleci
  - config.yml
- testing-requirements.txt
- Makefile
- .coveragerc
- LICENSE
- ci_tools
  - circle
    - build_doc.sh
    - checkout_merge_commit.sh
    - push_doc.sh
- appveyor.yml
- .gitattributes
- CHANGELOG.md
- README.rst
- .bumpversion.cfg
- setup.py
- .travis.yml
- requirements.txt
- rampwf
  - utils
    - testing.py
    - notebook.py
    - command_line.py
    - cli
      - testing.py
      - show.py
      - __init__.py
      - tests
        test_cli.py
    - importing.py
    - scoring.py
    - io.py
    - __init__.py
    - combine.py
    - tests
      - test_command_line.py
      - test_submission.py
      - test_sanitize.py
    - sanitize.py
    - submission.py
    - pretty_print.py
  - score_types
    - classifier_base.py
    - mare.py
    - make_combined.py
    - clustering_efficiency.py
    - detection
      - precision_recall.py
      - util.py
      - scp.py
      - ospa.py
      - iou.py
      - average_precision.py
      - __init__.py
      - base.py
    - f1_above.py
    - combined.py
    - macro_averaged_recall.py
    - classification_error.py
    - balanced_accuracy.py
    - brier_score.py
    - accuracy.py
    - soft_accuracy.py
    - roc_auc.py
    - relative_rmse.py
    - rmse.py
    - __init__.py
    - README.md
    - tests
      - test_soft_accuracy.py
      - __init__.py
      - test_detection.py
    - negative_log_likelihood.py
    - normalized_rmse.py
    - normalized_gini.py
    - base.py
  - hyperopt
    - hyperopt.py
    - cli
      - hyperopt.py
      - __init__.py
    - __init__.py
    - tests
      - test_hyperparameter.py
      - __init__.py
      - interfaces
        header_in_files
        titanic
        problem.py
        submissions
        starting_kit
        feature_extractor.py
        classifier.py
        data
        train.csv
        test.csv
        README.md
        requirements.txt
  - __init__.py
  - workflows
    - grid_feature_extractor.py
    - air_passengers.py
    - regressor.py
    - feature_extractor_regressor.py
    - image_classifier.py
    - grid_feature_extractor_classifier.py
    - object_detector.py
    - drug_spectra.py
    - feature_extractor.py
    - clusterer.py
    - __init__.py
    - tests
      - __init__.py
    - feature_extractor_classifier.py
    - classifier.py
    - sklearn_pipeline.py
    - ts_feature_extractor.py
    - el_nino.py
    - simplified_image_classifier.py
  - tests
    - test_kits.py
    - __init__.py
    - kits
      - boston_housing_old
        problem.py
        submissions
        starting_kit
        regressor.py
        data
        train.csv
        test.csv
      - titanic
        problem.py
        submissions
        starting_kit
        estimator.py
        data
        train.csv
        test.csv
        titanic_starting_kit.ipynb
      - titanic_no_test_old
        titanic_no_test_old_starting_kit.ipynb
        problem.py
        submissions
        starting_kit
        feature_extractor.py
        classifier.py
        data
        train.csv
      - HEP_tracking
        problem.py
        submissions
        starting_kit
        clusterer.py
        data
        train.csv
        test.csv
      - air_passengers_old
        problem.py
        submissions
        starting_kit
        external_data_mini.csv
        regressor.py
        feature_extractor.py
        data
        test.csv.bz2
        train.csv.bz2
      - mars_craters
        problem.py
        submissions
        starting_kit
        object_detector.py
        data
        labels_test_mini.csv
        data_train_mini.npy
        data_test_mini.npy
        labels_train_mini.csv
      - drug_spectra
        problem.py
        submissions
        starting_kit
        regressor.py
        feature_extractor_clf.py
        feature_extractor_reg.py
        classifier.py
        data
      - titanic_old
        problem.py
        submissions
        starting_kit
        feature_extractor.py
        classifier.py
        data
        train.csv
        test.csv
        titanic_old_starting_kit.ipynb
      - air_passengers
        problem.py
        submissions
        starting_kit
        external_data.csv
        estimator.py
        data
        test.csv.bz2
        train.csv.bz2
      - digits
        problem.py
        submissions
        starting_kit
        image_preprocessor.py
        batch_classifier.py
      - iris
        iris_starting_kit.ipynb
        problem.py
        submissions
        starting_kit
        estimator.py
        random_forest_10_10
        estimator.py
        data
        train.csv
        test.csv
      - boston_housing
        problem.py
        submissions
        starting_kit
        estimator.py
        data
        train.csv
        test.csv
      - digits_simplified
        problem.py
        submissions
        starting_kit
        image_classifier.py
        parallel_load
        image_classifier.py
      - el_nino
        problem.py
        submissions
        starting_kit
        regressor.py
        ts_feature_extractor.py
        data
        train_mini.npy
        test_mini.npy
      - iris_old
        iris_old_starting_kit.ipynb
        problem.py
        submissions
        starting_kit
        classifier.py
        random_forest_10_10
        classifier.py
        data
        train.csv
        test.csv
      - titanic_no_test
        problem.py
        submissions
        starting_kit
        estimator.py
        data
        train.csv
        titanic_no_test_starting_kit.ipynb
  - _version.py
  - prediction_types
    - detection.py
    - combined.py
    - multiclass.py
    - regression.py
    - __init__.py
    - tests
      - test_detection_predictions.py
      - test_regression_predictions.py
      - test_mixed_predictions.py
      - __init__.py
      - test_multiclass_predictions.py
    - clustering.py
    - base.py
  - cvs
    - time_series.py
    - __init__.py
    - clustering.py
  - externals
    - colored
      - fore.py
      - back.py
      - hex.py
      - __init__.py
      - colored.py
      - colors.py
      - style.py
    - __init__.py
    - tabulate.py
- .gitignore
- MANIFEST.in
- doc
  - Makefile
  - using_kits.rst
  - make.bat
  - problem.rst
  - contribute.rst
  - images
  - command_line.rst
  - sphinxext
    - sphinx_issues.py
    - README.txt
    - github_link.py
    - LICENSE.txt
    - MANIFEST.in
  - data.rst
  - workflow.rst
  - who_are_you.rst
  - index.rst
  - conf.py
  - scoring.rst
  - _templates
    - breadcrumbs.html
    - class.rst
    - numpydoc_docstring.py
    - function.rst
  - _static
    - img
    - js
      - copybutton.js
    - css
      - ramp.css

import os

import pandas as pd
from sklearn.model_selection import StratifiedShuffleSplit

import rampwf as rw


problem_title = 'Iris classification'
_target_column_name = 'species'
_prediction_label_names = ['setosa', 'versicolor', 'virginica']
# A type (class) which will be used to create wrapper objects for y_pred
Predictions = rw.prediction_types.make_multiclass(
    label_names=_prediction_label_names)
# An object implementing the workflow
workflow = rw.workflows.Classifier()

score_types = [
    rw.score_types.Accuracy(name='acc'),
    rw.score_types.ClassificationError(name='error'),
    rw.score_types.NegativeLogLikelihood(name='nll'),
    rw.score_types.F1Above(name='f1_70', threshold=0.7),
    # rw.score_types.BrierSkillScore(name="BSS", precision=3),
    # rw.score_types.BrierScore(name="BS", precision=3),
    # rw.score_types.BrierScoreReliability(name="BS Rel", precision=5),
    # rw.score_types.BrierScoreResolution(name="BS Res", precision=3),
    rw.score_types.NormalizedGini(name='ngini', precision=3),
    rw.score_types.BalancedAccuracy(name='bac', precision=3),
    rw.score_types.MacroAveragedRecall(name='mar', precision=3),
]


def get_cv(X, y):
    cv = StratifiedShuffleSplit(n_splits=2, test_size=0.2, random_state=57)
    return cv.split(X, y)


def _read_data(path, f_name):
    data = pd.read_csv(os.path.join(path, 'data', f_name))
    y_array = data[_target_column_name].values
    X_array = data.drop([_target_column_name], axis=1).values
    return X_array, y_array


def get_train_data(path='.'):
    f_name = 'train.csv'
    return _read_data(path, f_name)


def get_test_data(path='.'):
    f_name = 'test.csv'
    return _read_data(path, f_name)