def test_to_vw(self):
stream = TextFileStreamer(path_list=[self.doc1, self.doc2], tokenizer=self.tokenizer)
result = StringIO()
stream.to_vw(result)
benchmark = " 1 doc1| failure:1 doomed:1\n 1 doc2| set:1 success:1\n"
self.assertEqual(benchmark, result.getvalue())
def test_to_vw(self):
stream = TextFileStreamer(path_list=[self.doc1, self.doc2],
tokenizer=self.tokenizer)
result = StringIO()
stream.to_vw(result)
benchmark = " 1 doc1| failure:1 doomed:1\n 1 doc2| set:1 success:1\n"
self.assertEqual(benchmark, result.getvalue())
def test_to_scipyspare(self):
stream = TextFileStreamer(path_list=[self.doc1, self.doc2], tokenizer=self.tokenizer)
result = stream.to_scipysparse()
benchmark = sparse.csr_matrix([[1, 1, 0, 0], [0, 0, 1, 1]])
compare = result.toarray() == benchmark.toarray()
self.assertTrue(compare.all())
def test_to_scipyspare(self):
stream = TextFileStreamer(path_list=[self.doc1, self.doc2],
tokenizer=self.tokenizer)
result = stream.to_scipysparse()
benchmark = sparse.csr_matrix([[1, 1, 0, 0], [0, 0, 1, 1]])
compare = result.toarray() == benchmark.toarray()
self.assertTrue(compare.all())
def test_token_stream(self):
stream = TextFileStreamer(path_list=[self.doc1, self.doc2], tokenizer=self.tokenizer)
token_benchmark = [["doomed", "failure"], ["set", "success"]]
id_benchmark = ["doc1", "doc2"]
token_result = []
for each in stream.token_stream(cache_list=["doc_id"]):
token_result.append(each)
self.assertEqual(token_benchmark, token_result)
self.assertEqual(id_benchmark, stream.__dict__["doc_id_cache"])
def test_token_stream(self):
stream = TextFileStreamer(path_list=[self.doc1, self.doc2],
tokenizer=self.tokenizer)
token_benchmark = [['doomed', 'failure'], ['set', 'success']]
id_benchmark = ['doc1', 'doc2']
token_result = []
for each in stream.token_stream(cache_list=['doc_id']):
token_result.append(each)
self.assertEqual(token_benchmark, token_result)
self.assertEqual(id_benchmark, stream.__dict__['doc_id_cache'])
def test_token_stream(self):
stream = TextFileStreamer(path_list = [self.doc1, self.doc2],
tokenizer=self.tokenizer)
token_benchmark = [['doomed', 'failure'],
['set', 'success']]
id_benchmark = ['doc1', 'doc2']
token_result = []
for each in stream.token_stream(cache_list=['doc_id']):
token_result.append(each)
self.assertEqual(token_benchmark, token_result)
self.assertEqual(id_benchmark, stream.__dict__['doc_id_cache'])
def test_info_stream(self):
stream = TextFileStreamer(path_list=[self.doc1, self.doc2], tokenizer=self.tokenizer)
token_benchmark = [["doomed", "failure"], ["set", "success"]]
text_benchmark = ["doomed to failure\n", "set for success\n"]
token_result = []
text_result = []
for each in stream.info_stream():
token_result.append(each["tokens"])
text_result.append(each["text"])
self.assertEqual(token_benchmark, token_result)
self.assertEqual(text_benchmark, text_result)
def test_info_stream(self):
stream = TextFileStreamer(path_list=[self.doc1, self.doc2],
tokenizer=self.tokenizer)
token_benchmark = [['doomed', 'failure'], ['set', 'success']]
text_benchmark = ['doomed to failure\n', 'set for success\n']
token_result = []
text_result = []
for each in stream.info_stream():
token_result.append(each['tokens'])
text_result.append(each['text'])
self.assertEqual(token_benchmark, token_result)
self.assertEqual(text_benchmark, text_result)
def test_info_stream(self):
stream = TextFileStreamer(path_list = [self.doc1, self.doc2],
tokenizer=self.tokenizer)
token_benchmark = [['doomed', 'failure'],
['set', 'success']]
text_benchmark = ['doomed to failure\n', 'set for success\n']
token_result = []
text_result = []
for each in stream.info_stream():
token_result.append(each['tokens'])
text_result.append(each['text'])
self.assertEqual(token_benchmark, token_result)
self.assertEqual(text_benchmark, text_result)
def test_to_scipyspare(self):
stream = TextFileStreamer(path_list = [self.doc1, self.doc2],
tokenizer=self.tokenizer)
result = stream.to_scipysparse()
benchmark = sparse.csr_matrix([[1, 1, 0, 0], [0, 0, 1, 1]])
def test_to_scipyspare(self):
stream = TextFileStreamer(path_list=[self.doc1, self.doc2],
tokenizer=self.tokenizer)
result = stream.to_scipysparse()
benchmark = sparse.csr_matrix([[1, 1, 0, 0], [0, 0, 1, 1]])
