def test_two_word_seq(self):
data_source = getStream(self.test_words_long)
ts = split_corpus_dataset.DomainAdaptationSplit(
data_source, self.vocab, 2, 0.5)
tokens_strings = list(iter(ts))
expectation = [(torch.LongTensor([0, 1]), torch.LongTensor([2]))]
self.assertEqual(tokens_strings, expectation)
def test_single_word(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.TokenizedSplit(data_source, self.vocab, 1)
tokens_string = next(iter(ts))
expectation = (torch.LongTensor([0]), torch.LongTensor([1])
) # input, target
self.assertEqual(tokens_string, expectation)
def test_two_word_seq_long(self):
data_source = getStream(self.test_words_long)
ts = split_corpus_dataset.TokenizedSplit(data_source, self.vocab, 2)
tokens_strings = list(iter(ts))
expectation = [(torch.LongTensor([0, 1]), torch.LongTensor([1, 2])),
(torch.LongTensor([2, 0]), torch.LongTensor([0, 0]))]
self.assertEqual(tokens_strings, expectation)
def test_single_word_retrieval(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.DomainAdaptationSplit(data_source,
self.vocab,
1,
end_portion=0.5)
words = list(ts.input_words())
self.assertEqual(words, ['a']) # we expect the input words
def get_tokenized_splits(self, word_seqs, unroll):
files = [getStream(seq) for seq in word_seqs]
tss = [
split_corpus_dataset.TokenizedSplit(f, self.vocab, unroll)
for f in files
]
return tss
def test_single_word_seq(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.TokenizedSplit(data_source, self.vocab, 1)
tokens_strings = list(iter(ts))
expectation = [(torch.LongTensor([0]), torch.LongTensor([1])),
(torch.LongTensor([1]), torch.LongTensor([2])),
(torch.LongTensor([2]), torch.LongTensor([0]))]
self.assertEqual(tokens_strings, expectation)
def test_single_word(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.DomainAdaptationSplitFFMultiTarget(
data_source, self.vocab, 1, 1, end_portion=0.5)
tokens_string = next(iter(ts))
expectation = (torch.LongTensor([0]), torch.LongTensor([1])
) # input, target
self.assertEqual(tokens_string, expectation)
def test_two_word_seq_long_mt(self):
data_source = getStream(self.test_words_long)
ts = split_corpus_dataset.DomainAdaptationSplitFFMultiTarget(
data_source, self.vocab, 2, 2, end_portion=0.25)
tokens_strings = list(iter(ts))
expectation = [
(torch.LongTensor([0, 1, 2]), torch.LongTensor([2, 0])),
]
self.assertEqual(tokens_strings, expectation)
def test_iter_ends(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.TokenizedSplit(data_source, self.vocab, 1)
appender = ivec_appenders.CheatingIvecAppender(ts, self.ivec_eetor)
appender = iter(appender)
next(appender)
next(appender)
next(appender)
self.assertRaises(StopIteration, next, appender)
def test_single_data(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.TokenizedSplit(data_source, self.vocab, 1)
appender = ivec_appenders.CheatingIvecAppender(ts, self.ivec_eetor)
# cannot acces ts._tokens, it's an implementation
tokens = [self.vocab[w] for w in self.test_words_short]
expectation = self.ivec_eetor(" ".join(self.test_words_short[:-1]))
seqs = next(iter(appender))
first = seqs[2]
self.assertEqual(first, expectation)
def test_whole_seq(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.TokenizedSplit(data_source, self.vocab, 1)
appender = ivec_appenders.HistoryIvecAppender(ts, self.ivec_eetor)
# cannot acces ts._tokens, it's an implementation
tokens = [self.vocab[w] for w in self.test_words_short]
expectation = [
self.ivec_eetor(" ".join(self.test_words_short[:0])),
self.ivec_eetor(" ".join(self.test_words_short[:1])),
self.ivec_eetor(" ".join(self.test_words_short[:2])),
]
seqs = [x[2] for x in (iter(appender))]
self.assertEqual(seqs, expectation)
def test_whole_seq_with_next(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.TokenizedSplit(data_source, self.vocab, 1)
appender = ivec_appenders.CheatingIvecAppender(ts, self.ivec_eetor)
appender = iter(appender)
# cannot acces ts._tokens, it's an implementation
tokens = [self.vocab[w] for w in self.test_words_short]
expectation = [
self.ivec_eetor(" ".join(self.test_words_short[:-1])),
self.ivec_eetor(" ".join(self.test_words_short[:-1])),
self.ivec_eetor(" ".join(self.test_words_short[:-1]))
]
seq0 = next(appender)[2]
self.assertEqual(seq0, expectation[0])
seq1 = next(appender)[2]
self.assertEqual(seq1, expectation[1])
seq2 = next(appender)[2]
self.assertEqual(seq2, expectation[2])
def test_single_word_len(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.TokenizedSplitSingleTarget(
data_source, self.vocab, 1)
self.assertEqual(len(ts), len(self.test_words_short) - 1)
def test_two_word_retrieval(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.TokenizedSplit(data_source, self.vocab, 2)
words = list(ts.input_words())
self.assertEqual(words, ['a b']) # we expect the input words
def test_len_no_output(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.TokenizedSplit(data_source, self.vocab, 5)
self.assertEqual(len(ts), 0)
def test_two_word_retrieval(self):
data_source = getStream(self.test_words_long)
ts = split_corpus_dataset.DomainAdaptationSplitFFMultiTarget(
data_source, self.vocab, 2, 1, end_portion=0.5)
words = list(ts.input_words())
self.assertEqual(words, ['a a']) # we expect the input words
def test_single_word_len(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.DomainAdaptationSplit(
data_source, self.vocab, 1, 0.5)
self.assertEqual(len(ts), 2)
def test_len_no_output(self):
data_source = getStream(self.test_words_short)
ts = split_corpus_dataset.DomainAdaptationSplitFFMultiTarget(
data_source, self.vocab, 5, 1, end_portion=0.5)
self.assertEqual(len(ts), 0)
