mii/tokenizers
1
0
Fork 0
mirror of https://github.com/mii443/tokenizers.git synced 2025-12-08 13:48:19 +00:00
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
4 changed files with 13 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

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)]