• Search by Module
• Search by Words
• Search Projects

• Java
• Python
• JavaScript
• TypeScript
• C++
• Scala
• Blog

• discomll-master
  • discomll
    • ensemble
      • distributed_weighted_forest_rand.py
      • distributed_random_forest.py
      • distributed_weighted_forest.py
      • __init__.py
      • core
        • __init__.py
        • k_medoids.py
        • decision_tree.py
        • measures.py
      • forest_distributed_decision_trees.py
    • datasets
      • regression_data_test1.txt
      • iris_meta.csv
      • breast_cancer_wisconsin_cont.txt
      • iris.txt
      • regression_data_test2.txt
      • chunk_testdata.sh
      • ex3_test.txt
      • regression_data2.txt
      • breast_cancer_wisconsin_cont_test.txt
      • iris_test.txt
      • breastcancer_meta.csv
      • breast_cancer_wisconsin.names.txt
      • ex4.txt
      • ex3.txt
      • breast_cancer_wisconsin_disc.txt
      • breast_cancer_wisconsin_disc_test.txt
      • __init__.py
      • iris.names.txt
      • regression_data1.txt
    • utils
      • model_view.py
      • accuracy.py
      • distribution.py
      • __init__.py
    • examples
      • ensemble
        • dwf_rand_breastcancer_url.py
        • dwf_breastcancer_url.py
        • __init__.py
        • drf_breastcancer_url.py
        • fddt_breastcancer_url.py
      • big_data
        • logreg_sonar.py
        • nb_lymphography.py
        • kmeans_segmentation.py
        • fddt_segmentation.py
        • lwlr_fraction.py
        • linreg_fraction.py
        • lwlr_linear.py
        • drf_segmentation.py
        • drf_lymphography.py
        • linreg_linear.py
        • svm_ionosphere.py
        • svm_sonar.py
        • fddt_lymphography.py
        • dwfr_lymphography.py
        • logreg_ionosphere.py
        • __init__.py
        • kmeans_linear.py
        • nb_segmentation.py
        • dwfr_segmentation.py
      • classification
        • logistic_regression_ddfs.py
        • linear_svm_ddfs.py
        • naivebayes_ddfs.py
        • __init__.py
      • chunk_testdata.sh
      • clustering
        • __init__.py
        • kmeans_ddfs.py
      • regression
        • linear_regression_ddfs.py
        • __init__.py
        • locally_weighted_linear_regression_ddfs.py
      • __init__.py
    • classification
      • linear_svm.py
      • naivebayes.py
      • __init__.py
      • logistic_regression.py
    • clustering
      • __init__.py
      • kmeans.py
    • regression
      • linear_regression.py
      • locally_weighted_linear_regression.py
      • __init__.py
    • __init__.py
    • tests
      • chunk_testdata.py
      • tests_clustering.py
      • tests_regression.py
      • locally_weighted_linear_regression.py
      • datasets.py
      • __init__.py
      • tests_classification.py
    • dataset.py
  • setup.py
  • LICENSE.md
  • setup.cfg
  • README.md
  • requirements.txt
  • .gitignore

import unittest

import numpy as np
import Orange
from disco.core import result_iterator

import datasets


class Tests_Classification(unittest.TestCase):
    @classmethod
    def setUpClass(self):
        import chunk_testdata
        from disco import ddfs
        ddfs = ddfs.DDFS()

        if not ddfs.exists("test:ex3"):
            print "Chunking test datasets to DDFS..."
            chunk_testdata.chunk_testdata()

    def test_naivebayes_breastcancer(self):
        # python -m unittest tests_classification.Tests_Classification.test_naivebayes_breastcancer
        from discomll.classification import naivebayes
        train_data1, test_data1 = datasets.breastcancer_disc_orange()
        train_data2, test_data2 = datasets.breastcancer_disc_discomll()

        for m in range(3):
            learner = Orange.classification.bayes.NaiveLearner(m=m)
            classifier = learner(train_data1)
            predictions1 = [classifier(inst, Orange.classification.Classifier.GetBoth) for inst in test_data1]
            predictions1_target = [v[0].value for v in predictions1]
            predictions1_probs = [v[1].values() for v in predictions1]

            fitmodel_url = naivebayes.fit(train_data2)
            predictions_url = naivebayes.predict(test_data2, fitmodel_url, m=m)
            predictions2_target = []
            predictions2_probs = []
            for k, v in result_iterator(predictions_url):
                predictions2_target.append(v[0])
                predictions2_probs.append(v[1])

            self.assertListEqual(predictions1_target, predictions2_target)
            self.assertTrue(np.allclose(predictions1_probs, predictions2_probs))

    def test_naivebayes_breastcancer_cont(self):
        # python -m unittest tests_classification.Tests_Classification.test_naivebayes_breastcancer_cont
        from sklearn.naive_bayes import GaussianNB
        from discomll.classification import naivebayes

        x_train, y_train, x_test, y_test = datasets.breastcancer_cont(replication=1)
        train_data, test_data = datasets.breastcancer_cont_discomll(replication=1)

        clf = GaussianNB()
        probs_log1 = clf.fit(x_train, y_train).predict_proba(x_test)

        fitmodel_url = naivebayes.fit(train_data)
        prediction_url = naivebayes.predict(test_data, fitmodel_url)
        probs_log2 = [v[1] for _, v in result_iterator(prediction_url)]

        self.assertTrue(np.allclose(probs_log1, probs_log2, atol=1e-8))

    def test_log_reg_thetas(self):
        # python tests_classification.py Tests_Classification.test_log_reg_thetas
        from discomll.classification import logistic_regression

        train_data1 = datasets.ex4_orange()
        train_data2 = datasets.ex4_discomll()

        lr = Orange.classification.logreg.LogRegFitter_Cholesky(train_data1)
        thetas1 = lr[1]

        thetas_url = logistic_regression.fit(train_data2)
        thetas2 = [v for k, v in result_iterator(thetas_url["logreg_fitmodel"]) if k == "thetas"]

        self.assertTrue(np.allclose(thetas1, thetas2))

    def test_log_reg(self):
        # python tests_classification.py Tests_Classification.test_log_reg
        from discomll.classification import logistic_regression

        train_data1, test_data1 = datasets.breastcancer_cont_orange()
        train_data2, test_data2 = datasets.breastcancer_cont_discomll()

        learner = Orange.classification.logreg.LogRegLearner(fitter=Orange.classification.logreg.LogRegFitter_Cholesky)
        classifier = learner(train_data1)
        thetas1 = classifier.beta

        predictions1 = []
        probabilities1 = []
        for inst in test_data1:
            target, probs = classifier(inst, Orange.classification.Classifier.GetBoth)
            predictions1.append(target.value)
            probabilities1.append(probs.values())

        thetas_url = logistic_regression.fit(train_data2, alpha=1e-8, max_iterations=10)
        thetas2 = [v for k, v in result_iterator(thetas_url["logreg_fitmodel"]) if k == "thetas"]
        results_url = logistic_regression.predict(test_data2, thetas_url)

        predictions2 = []
        probabilities2 = []
        for k, v in result_iterator(results_url):
            predictions2.append(v[0])
            probabilities2.append(v[1])
        self.assertTrue(np.allclose(thetas1, thetas2))
        self.assertTrue(np.allclose(probabilities1, probabilities2, atol=1e-5))
        self.assertListEqual(predictions1, predictions2)


if __name__ == '__main__':
    unittest.main()