ImportError: Cannot load backend 'TkAgg' which requires the 'tk' interactive framework, as 'headless' is currently running

[harrygcoppock]

Hi,

First of all love the Library.

I am having problems with importing spec_augment_pytorch. The issue is with importing matplotlib same as this post: link

But as its in the package I cannot address this issue. Are there any quick fixes?

Running on python virtual env. Python 3.6. Linux.

pip list:
appdirs 1.4.4
audioread 2.1.9
certifi 2020.11.8
cffi 1.14.4
chardet 3.0.4
click 7.1.2
configparser 5.0.1
cycler 0.10.0
dataclasses 0.6
decorator 4.4.2
docker-pycreds 0.4.0
future 0.18.2
gitdb 4.0.5
GitPython 3.1.11
idna 2.10
joblib 0.17.0
kiwisolver 1.3.1
librosa 0.8.0
llvmlite 0.34.0
matplotlib 3.3.3
numba 0.51.2
numpy 1.19.4
packaging 20.7
pandas 1.1.4
pathtools 0.1.2
Pillow 8.0.1
pip 20.0.2
pkg-resources 0.0.0
pooch 1.3.0
promise 2.3
protobuf 3.14.0
psutil 5.7.3
pycparser 2.20
pyparsing 2.4.7
python-dateutil 2.8.1
pytz 2020.4
PyYAML 5.3.1
requests 2.25.0
resampy 0.2.2
scikit-learn 0.23.2
scipy 1.5.4
sentry-sdk 0.19.4
setuptools 44.0.0
shortuuid 1.0.1
six 1.15.0
sklearn 0.0
smmap 3.0.4
SoundFile 0.10.3.post1
SpecAugment 1.2.5
subprocess32 3.5.4
threadpoolctl 2.1.0
torch 1.7.0
torchvision 0.8.1
tqdm 4.54.0
typing-extensions 3.7.4.3
urllib3 1.26.2
wandb 0.10.12
watchdog 0.10.4

Thank you in advance for your time.

[PrathamSoni]

Perhaps try to just use the .py locally?

[actuy]

same error here...

[anpengjin]

me too, the same error

[harrygcoppock]

FYI I ended up using this function to mask the mel spec:

   def spec_augment(self,
               spec: np.ndarray,
               num_mask=2,
               freq_masking_max_percentage=0.15,
               time_masking_max_percentage=0.3):

    spec = spec.copy()
    for i in range(num_mask):
        all_frames_num, all_freqs_num = spec.shape
        freq_percentage = random.uniform(0.0, freq_masking_max_percentage)

        num_freqs_to_mask = int(freq_percentage * all_freqs_num)
        f0 = np.random.uniform(low=0.0,
                               high=all_freqs_num - num_freqs_to_mask)
        f0 = int(f0)
        spec[:, f0:f0 + num_freqs_to_mask] = 0

        time_percentage = random.uniform(0.0, time_masking_max_percentage)

        num_frames_to_mask = int(time_percentage * all_frames_num)
        t0 = np.random.uniform(low=0.0,
                               high=all_frames_num - num_frames_to_mask)
        t0 = int(t0)
        spec[t0:t0 + num_frames_to_mask, :] = 0

    return spec

[qihuifengrwth]

I solved the problem. mb you can try ;)
change "matplotlib.use('TkAgg')" into "matplotlib.use('Agg')" in site-packages/specAugment/spec_augment_pytorch.py
and lowercase the first character of any "SpecAugment" while loading packages.