Mirroristas/whisper
2
0
Fork 0
mirror of https://github.com/openai/whisper.git synced 2025-07-01 10:02:28 +00:00
Code Issues Packages Projects Releases Wiki Activity

remove tiktoken pin (#1759)

Browse Source
This commit is contained in:
Jong Wook Kim 2023-11-06 03:05:21 -08:00 committed by GitHub
parent b9f17e1f2d
commit 746aaaeafa
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 13 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
requirements.txt
View File

@ -3,4 +3,4 @@ numpy
torch
tqdm
more-itertools
tiktoken==0.3.3
tiktoken

14
tests/test_tokenizer.py
View File

@ -1,7 +1,17 @@
import pytest
from whisper.tokenizer import get_tokenizer
def test_tokenizer():
@pytest.mark.parametrize("multilingual", [True, False])
def test_tokenizer(multilingual):
tokenizer = get_tokenizer(multilingual=False)
assert tokenizer.sot in tokenizer.sot_sequence
assert len(tokenizer.all_language_codes) == len(tokenizer.all_language_tokens)
assert all(c < tokenizer.timestamp_begin for c in tokenizer.all_language_tokens)
def test_multilingual_tokenizer():
gpt2_tokenizer = get_tokenizer(multilingual=False)
multilingual_tokenizer = get_tokenizer(multilingual=True)
@ -20,5 +30,5 @@ def test_split_on_unicode():
tokens = [8404, 871, 287, 6, 246, 526, 3210, 20378]
words, word_tokens = multilingual_tokenizer.split_tokens_on_unicode(tokens)
assert words == [" elle", " est", " l", "'", "<EFBFBD>", "é", "rit", "oire"]
assert words == [" elle", " est", " l", "'", "\ufffd", "é", "rit", "oire"]
assert word_tokens == [[8404], [871], [287], [6], [246], [526], [3210], [20378]]