Test error in unittest_windows #741
Description
🐛 Bug
There are test errors (as shown below) in PR #738. I also saw it in my PR #735 .
To Reproduce
This is the error message.
#!bash.exe
.circleci/unittest/windows/scripts/run_test.sh
Collecting environment information...
PyTorch version: 1.6.0.dev20200623
Is debug build: No
CUDA used to build PyTorch: None
OS: Microsoft Windows Server 2019 Datacenter
GCC version: Could not collect
CMake version: Could not collect
Python version: 3.6
Is CUDA available: No
CUDA runtime version: No CUDA
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
Versions of relevant libraries:
[pip3] numpy==1.18.5
[pip3] torch==1.6.0.dev20200623
[pip3] torchaudio==0.6.0a0+bbf3003
[conda] Could not collect
============================= test session starts =============================
platform win32 -- Python 3.6.10, pytest-5.4.3, py-1.8.2, pluggy-0.13.1 -- C:\Users\circleci\project\env\python.exe
cachedir: .pytest_cache
rootdir: C:\Users\circleci\project
plugins: cov-2.10.0
collecting ... collected 1790 items / 1 error / 1789 selected
=================================== ERRORS ====================================
_____________ ERROR collecting test/test_librosa_compatibility.py _____________
test\test_librosa_compatibility.py:15: in <module>
import librosa
env\lib\site-packages\librosa\__init__.py:12: in <module>
from . import core
env\lib\site-packages\librosa\core\__init__.py:125: in <module>
from .time_frequency import * # pylint: disable=wildcard-import
env\lib\site-packages\librosa\core\time_frequency.py:11: in <module>
from ..util.exceptions import ParameterError
env\lib\site-packages\librosa\util\__init__.py:77: in <module>
from .utils import * # pylint: disable=wildcard-import
env\lib\site-packages\librosa\util\utils.py:5: in <module>
import scipy.ndimage
env\lib\site-packages\scipy\__init__.py:104: in <module>
from . import _distributor_init
env\lib\site-packages\scipy\_distributor_init.py:61: in <module>
WinDLL(os.path.abspath(filename))
env\lib\ctypes\__init__.py:348: in __init__
self._handle = _dlopen(self._name, mode)
E OSError: [WinError 126] The specified module could not be found
============================== warnings summary ===============================
env\lib\site-packages\_pytest\junitxml.py:417
C:\Users\circleci\project\env\lib\site-packages\_pytest\junitxml.py:417: PytestDeprecationWarning: The 'junit_family' default value will change to 'xunit2' in pytest 6.0.
Add 'junit_family=xunit1' to your pytest.ini file to keep the current format in future versions of pytest and silence this warning.
_issue_warning_captured(deprecated.JUNIT_XML_DEFAULT_FAMILY, config.hook, 2)
torchaudio\extension\extension.py:14
C:\Users\circleci\project\torchaudio\extension\extension.py:14: UserWarning: torchaudio C++ extension is not available.
warnings.warn('torchaudio C++ extension is not available.')
-- Docs: https://docs.pytest.org/en/latest/warnings.html
---- generated xml file: C:\Users\circleci\project\test-results\junit.xml -----
---------- coverage: platform win32, python 3.6.10-final-0 -----------
Name Stmts Miss Cover
-------------------------------------------------------------
torchaudio\__init__.py 14 4 71%
torchaudio\_internal\__init__.py 0 0 100%
torchaudio\_internal\misc_ops.py 17 12 29%
torchaudio\_internal\module_utils.py 25 4 84%
torchaudio\backend\__init__.py 5 0 100%
torchaudio\backend\common.py 32 11 66%
torchaudio\backend\no_backend.py 17 4 76%
torchaudio\backend\soundfile_backend.py 61 29 52%
torchaudio\backend\sox_backend.py 98 67 32%
torchaudio\backend\sox_io_backend.py 5 1 80%
torchaudio\backend\utils.py 41 15 63%
torchaudio\compliance\__init__.py 1 0 100%
torchaudio\compliance\kaldi.py 307 269 12%
torchaudio\datasets\__init__.py 10 0 100%
torchaudio\datasets\cmuarctic.py 48 30 38%
torchaudio\datasets\commonvoice.py 50 31 38%
torchaudio\datasets\gtzan.py 54 35 35%
torchaudio\datasets\librispeech.py 50 34 32%
torchaudio\datasets\ljspeech.py 41 24 41%
torchaudio\datasets\speechcommands.py 44 28 36%
torchaudio\datasets\utils.py 176 138 22%
torchaudio\datasets\vctk.py 63 43 32%
torchaudio\datasets\yesno.py 46 30 35%
torchaudio\extension\__init__.py 3 0 100%
torchaudio\extension\extension.py 30 10 67%
torchaudio\functional.py 793 739 7%
torchaudio\kaldi_io.py 28 9 68%
torchaudio\models\__init__.py 2 0 100%
torchaudio\models\_wavernn.py 101 81 20%
torchaudio\models\wav2letter.py 18 11 39%
torchaudio\sox_effects\__init__.py 4 1 75%
torchaudio\sox_effects\sox_effects.py 87 62 29%
torchaudio\transforms.py 331 250 24%
torchaudio\version.py 2 0 100%
-------------------------------------------------------------
TOTAL 2604 1972 24%
=========================== short test summary info ===========================
ERROR test/test_librosa_compatibility.py - OSError: [WinError 126] The specif...
!!!!!!!!!!!!!!!!!!! Interrupted: 1 error during collection !!!!!!!!!!!!!!!!!!!!
======================== 2 warnings, 1 error in 2.96s =========================
Expected behavior
Test passed.