test_feature_extraction_auto.py 2.73 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
# coding=utf-8
# Copyright 2021 the HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

16
import json
17
import os
18
import tempfile
19
20
import unittest

21
from transformers import AutoFeatureExtractor, Wav2Vec2Config, Wav2Vec2FeatureExtractor
22
23


24
SAMPLE_FEATURE_EXTRACTION_CONFIG_DIR = os.path.join(os.path.dirname(os.path.abspath(__file__)), "fixtures")
25
26
27
SAMPLE_FEATURE_EXTRACTION_CONFIG = os.path.join(
    os.path.dirname(os.path.abspath(__file__)), "fixtures/dummy_feature_extractor_config.json"
)
28
SAMPLE_CONFIG = os.path.join(os.path.dirname(os.path.abspath(__file__)), "fixtures/dummy-config.json")
29
30
31
32
33
34
35


class AutoFeatureExtractorTest(unittest.TestCase):
    def test_feature_extractor_from_model_shortcut(self):
        config = AutoFeatureExtractor.from_pretrained("facebook/wav2vec2-base-960h")
        self.assertIsInstance(config, Wav2Vec2FeatureExtractor)

36
    def test_feature_extractor_from_local_directory_from_key(self):
37
38
39
        config = AutoFeatureExtractor.from_pretrained(SAMPLE_FEATURE_EXTRACTION_CONFIG_DIR)
        self.assertIsInstance(config, Wav2Vec2FeatureExtractor)

40
41
42
43
44
45
    def test_feature_extractor_from_local_directory_from_config(self):
        with tempfile.TemporaryDirectory() as tmpdirname:
            model_config = Wav2Vec2Config()

            # remove feature_extractor_type to make sure config.json alone is enough to load feature processor locally
            config_dict = AutoFeatureExtractor.from_pretrained(SAMPLE_FEATURE_EXTRACTION_CONFIG_DIR).to_dict()
46

47
            config_dict.pop("feature_extractor_type")
48
            config = Wav2Vec2FeatureExtractor(**config_dict)
49
50
51
52
53
54
55

            # save in new folder
            model_config.save_pretrained(tmpdirname)
            config.save_pretrained(tmpdirname)

            config = AutoFeatureExtractor.from_pretrained(tmpdirname)

56
57
58
59
            # make sure private variable is not incorrectly saved
            dict_as_saved = json.loads(config.to_json_string())
            self.assertTrue("_processor_class" not in dict_as_saved)

60
61
        self.assertIsInstance(config, Wav2Vec2FeatureExtractor)

62
63
64
    def test_feature_extractor_from_local_file(self):
        config = AutoFeatureExtractor.from_pretrained(SAMPLE_FEATURE_EXTRACTION_CONFIG)
        self.assertIsInstance(config, Wav2Vec2FeatureExtractor)