[Discussion] Something about training spleeter model

[Spencer19990618]

I tried to train my model with musdb18 dataset with CLI and find out the total required time of training. I implemented the command "time spleeter train -p ./configs/musdb_config.json -d ~/data/musdb18/train/". I have modified the json file and checked all of the directories, but It seems nothing happened. The terminal only printed out:

INFO:spleeter:Start model training INFO:spleeter:Model training done

real 0m37.829s user 0m34.643s sys 0m1.982s

[romi1502]

Hi @Spencer19990618, something seems to have indeed gone wrong as the full training time seems quite small. You could try to use --verbose option to have a bit more info about what's happening under the hood.

[Spencer19990618]

@romi1502 I had tried --verbose before. it seems that the files in my dataset were not loaded.

INFO:tensorflow:Saving 'checkpoint_path' summary for global step 0: musdb_model/model.ckpt-0
WARNING:tensorflow:Training with estimator made no steps. Perhaps input is empty or misspecified.

I think it is the same issue as #612 that haven't been solved yet.

I also tried to perform training using Python API , but the instruction website in wiki is invalid now.

update:

Error: ERROR:spleeter:An error occurs while loading audio Traceback (most recent call last): File "/home/cclun/miniconda3/envs/envspleeter/lib/python3.8/site-packages/spleeter/audio/ffmpeg.py", line 101, in load probe = ffmpeg.probe(path) File "/home/cclun/miniconda3/envs/envspleeter/lib/python3.8/site-packages/ffmpeg/_probe.py", line 23, in probe raise Error('ffprobe', out, err) ffmpeg._run.Error: ffprobe error (see stderr output for detail)

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "/tmp/tmppfj0a383.py", line 21, in safeload (data, ) = ag.converted_call(ag.ld(self).load, (ag.converted_call(ag.ld(path).numpy, (), None, fscope_1), ag.converted_call(ag__.ld(offset).numpy, (), None, fscope_1), ag.converted_call(ag.ld(duration).numpy, (), None, fscope_1), ag__.converted_call(ag.ld(sample_rate).numpy, (), None, fscope_1)), dict(dtype=ag__.converted_call(ag__.ld(dtype).numpy, (), None, fscope_1)), fscope_1) File "/home/cclun/miniconda3/envs/envspleeter/lib/python3.8/site-packages/tensorflow/python/autograph/impl/api.py", line 340, in converted_call return _call_unconverted(f, args, kwargs, options, False) File "/home/cclun/miniconda3/envs/envspleeter/lib/python3.8/site-packages/tensorflow/python/autograph/impl/api.py", line 463, in _call_unconverted return f(*args, **kwargs) File "/home/cclun/miniconda3/envs/envspleeter/lib/python3.8/site-packages/spleeter/audio/ffmpeg.py", line 103, in load raise SpleeterError( spleeter.SpleeterError: An error occurs with ffprobe (see ffprobe output below)

ffprobe version 9c33b2f Copyright (c) 2007-2021 the FFmpeg developers built with gcc 9.3.0 (crosstool-NG 1.24.0.133_b0863d8_dirty) configuration: --prefix=/home/cclun/miniconda3/envs/envspleeter --cc=/home/conda/feedstock_root/build_artifacts/ffmpeg_1627813612080/_build_env/bin/x86_64-conda-linux-gnu-cc --disable-doc --disable-openssl --enable-avresample --enable-gnutls --enable-gpl --enable-hardcoded-tables --enable-libfreetype --enable-libopenh264 --enable-libx264 --enable-pic --enable-pthreads --enable-shared --enable-static --enable-version3 --enable-zlib --enable-libmp3lame --pkg-config=/home/conda/feedstock_root/build_artifacts/ffmpeg_1627813612080/_build_env/bin/pkg-config libavutil 56. 51.100 / 56. 51.100 libavcodec 58. 91.100 / 58. 91.100 libavformat 58. 45.100 / 58. 45.100 libavdevice 58. 10.100 / 58. 10.100 libavfilter 7. 85.100 / 7. 85.100 libavresample 4. 0. 0 / 4. 0. 0 libswscale 5. 7.100 / 5. 7.100 libswresample 3. 7.100 / 3. 7.100 libpostproc 55. 7.100 / 55. 7.100 /media/labhdd/data/cclun/spleeter/train/train/Young Griffo - Pennies/mixture.wav: No such file or directory

and my terminal is like : ~/data/spleeter$ time spleeter train --verbose -p configs/musdb_config.json -d ./train

I am now handling the directory problem. it will be thankful, if you can tell me which .py file is responsible for loading audio file.