#!/usr/bin/env python # -*- coding: utf-8 -*- # # Copyright (C) 2010 Radim Rehurek # Licensed under the GNU LGPL v2.1 - http://www.gnu.org/licenses/lgpl.html """ Automated tests for checking transformation algorithms (the models package). """ import logging import unittest import numpy as np import scipy.linalg from gensim import matutils from gensim.corpora.mmcorpus import MmCorpus from gensim.models import lsimodel from gensim.test import basetmtests from gensim.test.utils import datapath, get_tmpfile class TestLsiModel(unittest.TestCase, basetmtests.TestBaseTopicModel): def setUp(self): self.corpus = MmCorpus(datapath('testcorpus.mm')) self.model = lsimodel.LsiModel(self.corpus, num_topics=2) def test_transform(self): """Test lsi[vector] transformation.""" # create the transformation model model = self.model # make sure the decomposition is enough accurate u, s, vt = scipy.linalg.svd(matutils.corpus2dense(self.corpus, self.corpus.num_terms), full_matrices=False) self.assertTrue(np.allclose(s[:2], model.projection.s)) # singular values must match # transform one document doc = list(self.corpus)[0] transformed = model[doc] vec = matutils.sparse2full(transformed, 2) # convert to dense vector, for easier equality tests expected = np.array([-0.6594664, 0.142115444]) # scaled LSI version # expected = np.array([-0.1973928, 0.05591352]) # non-scaled LSI version self.assertTrue(np.allclose(abs(vec), abs(expected))) # transformed entries must be equal up to sign def test_transform_float32(self): """Test lsi[vector] transformation.""" # create the transformation model model = lsimodel.LsiModel(self.corpus, num_topics=2, dtype=np.float32) # make sure the decomposition is enough accurate u, s, vt = scipy.linalg.svd(matutils.corpus2dense(self.corpus, self.corpus.num_terms), full_matrices=False) self.assertTrue(np.allclose(s[:2], model.projection.s)) # singular values must match self.assertEqual(model.projection.u.dtype, np.float32) self.assertEqual(model.projection.s.dtype, np.float32) # transform one document doc = list(self.corpus)[0] transformed = model[doc] vec = matutils.sparse2full(transformed, 2) # convert to dense vector, for easier equality tests expected = np.array([-0.6594664, 0.142115444]) # scaled LSI version # transformed entries must be equal up to sign self.assertTrue(np.allclose(abs(vec), abs(expected), atol=1.e-5)) def test_corpus_transform(self): """Test lsi[corpus] transformation.""" model = self.model got = np.vstack([matutils.sparse2full(doc, 2) for doc in model[self.corpus]]) expected = np.array([ [0.65946639, 0.14211544], [2.02454305, -0.42088759], [1.54655361, 0.32358921], [1.81114125, 0.5890525], [0.9336738, -0.27138939], [0.01274618, -0.49016181], [0.04888203, -1.11294699], [0.08063836, -1.56345594], [0.27381003, -1.34694159] ]) self.assertTrue(np.allclose(abs(got), abs(expected))) # must equal up to sign def test_online_transform(self): corpus = list(self.corpus) doc = corpus[0] # use the corpus' first document for testing # create the transformation model model2 = lsimodel.LsiModel(corpus=corpus, num_topics=5) # compute everything at once # start with no documents, we will add them later model = lsimodel.LsiModel(corpus=None, id2word=model2.id2word, num_topics=5) # train model on a single document model.add_documents([corpus[0]]) # transform the testing document with this partial transformation transformed = model[doc] vec = matutils.sparse2full(transformed, model.num_topics) # convert to dense vector, for easier equality tests expected = np.array([-1.73205078, 0.0, 0.0, 0.0, 0.0]) # scaled LSI version self.assertTrue(np.allclose(abs(vec), abs(expected), atol=1e-6)) # transformed entries must be equal up to sign # train on another 4 documents model.add_documents(corpus[1:5], chunksize=2) # train on 4 extra docs, in chunks of 2 documents, for the lols # transform a document with this partial transformation transformed = model[doc] vec = matutils.sparse2full(transformed, model.num_topics) # convert to dense vector, for easier equality tests expected = np.array([-0.66493785, -0.28314203, -1.56376302, 0.05488682, 0.17123269]) # scaled LSI version self.assertTrue(np.allclose(abs(vec), abs(expected), atol=1e-6)) # transformed entries must be equal up to sign # train on the rest of documents model.add_documents(corpus[5:]) # make sure the final transformation is the same as if we had decomposed the whole corpus at once vec1 = matutils.sparse2full(model[doc], model.num_topics) vec2 = matutils.sparse2full(model2[doc], model2.num_topics) # the two LSI representations must equal up to sign self.assertTrue(np.allclose(abs(vec1), abs(vec2), atol=1e-5)) def test_persistence(self): fname = get_tmpfile('gensim_models_lsi.tst') model = self.model model.save(fname) model2 = lsimodel.LsiModel.load(fname) self.assertEqual(model.num_topics, model2.num_topics) self.assertTrue(np.allclose(model.projection.u, model2.projection.u)) self.assertTrue(np.allclose(model.projection.s, model2.projection.s)) tstvec = [] self.assertTrue(np.allclose(model[tstvec], model2[tstvec])) # try projecting an empty vector def test_persistence_compressed(self): fname = get_tmpfile('gensim_models_lsi.tst.gz') model = self.model model.save(fname) model2 = lsimodel.LsiModel.load(fname, mmap=None) self.assertEqual(model.num_topics, model2.num_topics) self.assertTrue(np.allclose(model.projection.u, model2.projection.u)) self.assertTrue(np.allclose(model.projection.s, model2.projection.s)) tstvec = [] self.assertTrue(np.allclose(model[tstvec], model2[tstvec])) # try projecting an empty vector def test_large_mmap(self): fname = get_tmpfile('gensim_models_lsi.tst') model = self.model # test storing the internal arrays into separate files model.save(fname, sep_limit=0) # now load the external arrays via mmap model2 = lsimodel.LsiModel.load(fname, mmap='r') self.assertEqual(model.num_topics, model2.num_topics) self.assertTrue(isinstance(model2.projection.u, np.memmap)) self.assertTrue(isinstance(model2.projection.s, np.memmap)) self.assertTrue(np.allclose(model.projection.u, model2.projection.u)) self.assertTrue(np.allclose(model.projection.s, model2.projection.s)) tstvec = [] self.assertTrue(np.allclose(model[tstvec], model2[tstvec])) # try projecting an empty vector def test_large_mmap_compressed(self): fname = get_tmpfile('gensim_models_lsi.tst.gz') model = self.model # test storing the internal arrays into separate files model.save(fname, sep_limit=0) # now load the external arrays via mmap return # turns out this test doesn't exercise this because there are no arrays # to be mmaped! self.assertRaises(IOError, lsimodel.LsiModel.load, fname, mmap='r') def test_docs_processed(self): self.assertEqual(self.model.docs_processed, 9) self.assertEqual(self.model.docs_processed, self.corpus.num_docs) def test_get_topics(self): topics = self.model.get_topics() vocab_size = len(self.model.id2word) for topic in topics: self.assertTrue(isinstance(topic, np.ndarray)) self.assertEqual(topic.dtype, np.float64) self.assertEqual(vocab_size, topic.shape[0]) # LSI topics are not probability distributions # self.assertAlmostEqual(np.sum(topic), 1.0, 5) if __name__ == '__main__': logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.DEBUG) unittest.main()