hasktorch/tokenizers
1
1
Fork 0
mirror of https://github.com/hasktorch/tokenizers.git synced 2024-10-26 14:09:09 +03:00
Code Issues Packages Projects Releases Wiki Activity

Python - Update tests for new encode

Browse Source
This commit is contained in:
Anthony MOI 2020-04-24 21:30:51 -04:00
parent 2e105c4258
commit dbc8e68c68
No known key found for this signature in database
GPG Key ID: CB646B1164C636A0
4 changed files with 13 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
bindings/python/tests/bindings/test_processors.py
View File

@ -19,7 +19,7 @@ class TestBertProcessing:
tokenizer.add_tokens(["my", "name", "is", "john", "pair"])
tokenizer.post_processor = BertProcessing(("[SEP]", 0), ("[CLS]", 1))
output = tokenizer.encode("my name", "pair")
output = tokenizer.encode(("my name", "pair"))
assert output.tokens == ["[CLS]", "my", "name", "[SEP]", "pair", "[SEP]"]
assert output.ids == [1, 2, 3, 0, 6, 0]
@ -37,7 +37,7 @@ class TestRobertaProcessing:
tokenizer.add_tokens(["my", "name", "is", "john", "pair"])
tokenizer.post_processor = RobertaProcessing(("</s>", 1), ("<s>", 0))
output = tokenizer.encode("my name", "pair")
output = tokenizer.encode(("my name", "pair"))
assert output.tokens == ["<s>", "my", "name", "</s>", "</s>", "pair", "</s>"]
assert output.ids == [0, 2, 3, 1, 1, 6, 1]

10
bindings/python/tests/bindings/test_tokenizer.py
View File

@ -102,9 +102,13 @@ class TestTokenizer:
assert type(output.overflowing) == list
# Can encode a pair of sequences
output = tokenizer.encode("my name is john", "pair")
output = tokenizer.encode(("my name is john", "pair"))
assert output.tokens == ["my", "name", "is", "john", "pair"]
# Can encode a single pre-tokenized sequence
output = tokenizer.encode(["my", "name", "is", "john"])
assert output.tokens == ["my", "name", "is", "john"]
# Can encode a batch with both a single sequence and a pair of sequences
output = tokenizer.encode_batch(["my name is john", ("my name is john", "pair")])
assert len(output) == 2
@ -136,7 +140,7 @@ class TestTokenizer:
assert output.tokens == ["my", "name"]
# Can truncate pair sequences as well
output = tokenizer.encode("my name is john", "pair")
output = tokenizer.encode(("my name is john", "pair"))
assert output.tokens == ["my", "pair"]
def test_padding(self):
@ -156,7 +160,7 @@ class TestTokenizer:
tokenizer.enable_padding(max_length=4)
output = tokenizer.encode("my name")
assert output.tokens == ["my", "name", "[PAD]", "[PAD]"]
output = tokenizer.encode("my name", "pair")
output = tokenizer.encode(("my name", "pair"))
assert output.tokens == ["my", "name", "pair", "[PAD]"]
def test_decode(self):

4
bindings/python/tests/implementations/test_bert_wordpiece.py
View File

@ -7,14 +7,14 @@ class TestBertWordPieceBPE:
tokenizer = BertWordPieceTokenizer(bert_files["vocab"])
# Encode with special tokens by default
output = tokenizer.encode("My name is John", "pair")
output = tokenizer.encode(("My name is John", "pair"))
assert output.ids == [101, 2026, 2171, 2003, 2198, 102, 3940, 102]
assert output.tokens == ["[CLS]", "my", "name", "is", "john", "[SEP]", "pair", "[SEP]"]
assert output.offsets == [(0, 0), (0, 2), (3, 7), (8, 10), (11, 15), (0, 0), (0, 4), (0, 0)]
assert output.type_ids == [0, 0, 0, 0, 0, 0, 1, 1]
# Can encode without the special tokens
output = tokenizer.encode("My name is John", "pair", add_special_tokens=False)
output = tokenizer.encode(("My name is John", "pair"), add_special_tokens=False)
assert output.ids == [2026, 2171, 2003, 2198, 3940]
assert output.tokens == ["my", "name", "is", "john", "pair"]
assert output.offsets == [(0, 2), (3, 7), (8, 10), (11, 15), (0, 4)]

4
bindings/python/tests/implementations/test_char_bpe.py
View File

@ -6,7 +6,7 @@ class TestBertWordPieceBPE:
def test_basic_encode(self, openai_files):
tokenizer = CharBPETokenizer(openai_files["vocab"], openai_files["merges"])
output = tokenizer.encode("My name is John", "pair")
output = tokenizer.encode(("My name is John", "pair"))
assert output.ids == [0, 253, 1362, 544, 0, 7, 12662, 2688]
assert output.tokens == [
"<unk>",
@ -32,7 +32,7 @@ class TestBertWordPieceBPE:
def test_lowercase(self, openai_files):
tokenizer = CharBPETokenizer(openai_files["vocab"], openai_files["merges"], lowercase=True)
output = tokenizer.encode("My name is John", "pair", add_special_tokens=False)
output = tokenizer.encode(("My name is John", "pair"), add_special_tokens=False)
assert output.ids == [547, 1362, 544, 2476, 2688]
assert output.tokens == ["my</w>", "name</w>", "is</w>", "john</w>", "pair</w>"]
assert output.offsets == [(0, 2), (3, 7), (8, 10), (11, 15), (0, 4)]