A issue about vad_extract.py

[wolfwaytx]

Hello, thank you for your outstanding work. I encountered the following issue when using the vad_extract.py provided in your code:

Traceback (most recent call last):
  File "/private/CPJ/vadnet-master/vad_extract.py", line 141, in <module>
    extract_voice(args.model, args.files, n_batch=args.n_batch)
  File "/private/CPJ/vadnet-master/vad_extract.py", line 98, in extract_voice
    sess.run(init, feed_dict = { x : input, y : labels, ph_n_shuffle : 1, ph_n_repeat : 1, ph_n_batch : n_batch})
  File "/private/CPJ/anaconda3/envs/vad/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 969, in run
    run_metadata_ptr)
  File "/private/CPJ/anaconda3/envs/vad/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 1192, in _run
    feed_dict_tensor, options, run_metadata)
  File "/private/CPJ/anaconda3/envs/vad/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 1372, in _do_run
    run_metadata)
  File "/private/CPJ/anaconda3/envs/vad/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 1397, in _do_call
    raise type(e)(node_def, op, message)  # pylint: disable=no-value-for-parameter
tensorflow.python.framework.errors_impl.InternalError: Graph execution error:

Detected at node 'ds/initializer' defined at (most recent call last):
Node: 'ds/initializer'
AddDataset: Failed to build BatchDataset op with error More Input() calls than the 2 input_args while building NodeDef 'BatchDataset/_14' using Op<name=BatchDataset; signature=input_dataset:variant, batch_size:int64 -> handle:variant; attr=output_types:list(type),min=1; attr=output_shapes:list(shape),min=1; attr=metadata:string,default="">
     [[{{node ds/initializer}}]]

Original stack trace for 'ds/initializer':

Here it prompts that two extra parameters are given, but I executed it according to the way you provided in the readme. What could be the reason for this? Looking forward to your reply.