def test_orange_with_tweets_kmeans(self): import time start = time.time() from_date = datetime.datetime(2011, 1, 25, 0, 0, 0) to_date = datetime.datetime(2011, 1, 26, 0, 0, 0) items = ws.get_documents_by_date(from_date, to_date, limit=1000) oc = OrangeKmeansClusterer(k=20, ngram=1) oc.add_documents(items) oc.run("orange_clustering_test", pca=True) print time.time() - start oc.plot_growth_timeline(cumulative=True) oc.plot_scatter() oc.dump_clusters_to_file("kmeans_with_tweets_orange")
''' Created on 26 Jan 2012 @author: george ''' import unittest, numpy from analysis.clustering.kmeans import OrangeKmeansClusterer from tests.test_document import get_test_documents ########################################### # GLOBALS # ########################################### ignore, ignore, samples = get_test_documents() oc = OrangeKmeansClusterer(k=2) for sample in samples: oc.add_document(sample) class Test(unittest.TestCase): def test_orange_cluster_term_document_matrix(self): oc.construct_term_doc_matrix() calculated = oc.td_matrix expected = numpy.array( [[0.31388923, 0.11584717, 0, 0, 0, 0, 0.47083384], [0, 0.13515504, 0.3662041, 0, 0.3662041, 0, 0], [0, 0, 0, 0.54930614, 0, 0.549306140, 0]]) self.assertEqual(expected.all(), calculated.all()) def test_orange_save_matrix_to_tab_file(self): oc.construct_term_doc_matrix()