洋红Tensorflow Melody\u RNN生成允许误差

Question

所以我一直在尝试用洋红的旋律来训练一些古典音乐，更确切地说是2小时的巴赫。我正在通过Anaconda使用带有GTX1080和Python 3.8的Tensorflow GPU。我用这个命令来训练它：

melody_rnn_train \ --config=lookback_rnn \ --run_dir=C:/Users/hudso/Downloads/bach_dataset/melody_rnn/logdir/run1 \ --sequence_example_file=C:/Users/hudso/Downloads/bach_dataset/melody_rnn/sequence_examples/training_melodies.tfrecord \ --hparams="batch_size=64,rnn_layer_sizes=[128,128]" \ --num_training_steps=5000

这是我用来创作音乐的：

melody_rnn_generate \ --config=lookback_rnn \ --run_dir=C:/Users/hudso/Downloads/bach_dataset/melody_rnn/logdir/run1 \ --output_dir=C:/Users/hudso/Downloads/bach_dataset/melody_rnn/generated \ --num_outputs=10 \ --num_steps=128 \ --hparams="batch_size=64,rnn_layer_sizes=[128,128]" \ --primer_melody="[60]"

但无论出于何种原因运行第二个命令时，我都会出现以下错误：

I1108 21:46:36.451361  5740 model.py:57] Checkpoint used: C:/Users/hudso/Downloads/bach_dataset/melody_rnn/logdir/run1\train\model.ckpt-5000
INFO:tensorflow:Restoring parameters from C:/Users/hudso/Downloads/bach_dataset/melody_rnn/logdir/run1\train\model.ckpt-5000
I1108 21:46:36.453360  5740 saver.py:1293] Restoring parameters from C:/Users/hudso/Downloads/bach_dataset/melody_rnn/logdir/run1\train\model.ckpt-5000
2020-11-08 21:46:36.729688: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library cublas64_10.dll
INFO:tensorflow:Beam search yields sequence with log-likelihood: -36.386410
I1108 21:46:37.818498  5740 events_rnn_model.py:380] Beam search yields sequence with log-likelihood: -36.386410
Traceback (most recent call last):
  File "c:\users\hudso\anaconda3\envs\artifintel\lib\runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "c:\users\hudso\anaconda3\envs\artifintel\lib\runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "C:\Users\hudso\anaconda3\envs\ArtifIntel\Scripts\melody_rnn_generate.exe\__main__.py", line 7, in <module>
  File "c:\users\hudso\anaconda3\envs\artifintel\lib\site-packages\magenta\models\melody_rnn\melody_rnn_generate.py", line 252, in console_entry_point
    tf.app.run(main)
  File "C:\Users\hudso\AppData\Roaming\Python\Python38\site-packages\tensorflow\python\platform\app.py", line 40, in run
    _run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
  File "C:\Users\hudso\AppData\Roaming\Python\Python38\site-packages\absl\app.py", line 299, in run
    _run_main(main, args)
  File "C:\Users\hudso\AppData\Roaming\Python\Python38\site-packages\absl\app.py", line 250, in _run_main
    sys.exit(main(argv))
  File "c:\users\hudso\anaconda3\envs\artifintel\lib\site-packages\magenta\models\melody_rnn\melody_rnn_generate.py", line 247, in main
    run_with_flags(generator)
  File "c:\users\hudso\anaconda3\envs\artifintel\lib\site-packages\magenta\models\melody_rnn\melody_rnn_generate.py", line 214, in run_with_flags
    note_seq.sequence_proto_to_midi_file(generated_sequence, midi_path)
  File "c:\users\hudso\anaconda3\envs\artifintel\lib\site-packages\note_seq\midi_io.py", line 371, in sequence_proto_to_midi_file
    return note_sequence_to_midi_file(sequence, output_file,
  File "c:\users\hudso\anaconda3\envs\artifintel\lib\site-packages\note_seq\midi_io.py", line 217, in note_sequence_to_midi_file
    copyfile(temp_file.name, output_file)
  File "c:\users\hudso\anaconda3\envs\artifintel\lib\shutil.py", line 261, in copyfile
    with open(src, 'rb') as fsrc, open(dst, 'wb') as fdst:
PermissionError: [Errno 13] Permission denied: 'C:\\Users\\hudso\\AppData\\Local\\Temp\\tmpbrmqzbjf'

我一直在试图找出原因，但我真的一事无成。我并没有太多地使用StackOverflow，所以如果我遗漏了一些东西，我应该只添加注释，然后编辑它。谢谢