MCPcopy Create free account
hub / github.com/ankane/tokenizers-ruby / test_from_pretrained_bert

Method test_from_pretrained_bert

test/tokenizer_test.rb:4–16  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

2
3class TokenizerTest < Minitest::Test
4 def test_from_pretrained_bert
5 tokenizer = Tokenizers.from_pretrained("bert-base-cased")
6
7 # encode
8 encoded = tokenizer.encode("I can feel the magic, can you?")
9 expected_ids = [101, 146, 1169, 1631, 1103, 3974, 117, 1169, 1128, 136, 102]
10 expected_tokens = ["[CLS]", "I", "can", "feel", "the", "magic", ",", "can", "you", "?", "[SEP]"]
11 assert_equal expected_ids, encoded.ids
12 assert_equal expected_tokens, encoded.tokens
13
14 # decode
15 assert_equal "I can feel the magic, can you?", tokenizer.decode(encoded.ids)
16 end
17
18 def test_from_pretrained_gpt2
19 tokenizer = Tokenizers.from_pretrained("gpt2")

Callers

nothing calls this directly

Calls 3

from_pretrainedMethod · 0.45
encodeMethod · 0.45
decodeMethod · 0.45

Tested by

no test coverage detected