SENN_train cannot run

[Franky0620]

I set the Path of data, but it has the problem on " images, targets = SE_Net.inputs(data_frames)"

[Huer-H]

Hello, I have the same problem as you. How do you solve it?

[zhr1201]

I think it is caused by different versions of python and its dependencies. You should check the error details to get that working on your versions of software

[Franky0620]

I think it is caused by different versions of python and its dependencies. You should check the error details to get that working on your versions of software

I test on Python 3.7 3.6 3.5, while, not working . The issue is happed on SENN.py, for i in range(0, self.batch_size, 1)])....

[Franky0620]

Hello, I have the same problem as you. How do you solve it?

still not working....

[zhr1201]

Please please please read the error message carefully otherwise you are not moving in the right direction

[Huer-H]

targets = tf.concat(0,[tf.reshape(speech_f[i][self.N_IN - 1][0:self.NEFF],[1, self.NEFF]) for i in range(0, self.batch_size, 1)])
There is a problem with this part of the code (senn.py)
----ValueError: Shapes (20, 1, 129) and () are incompatible---
20 is batch_size

[Huer-H]

I have modified this part of the code

targets = tf.concat(0,[tf.reshape(speech_f[i][self.N_IN - 1][0:self.NEFF],[1, self.NEFF]) for i in range(0, self.batch_size, 1)])------------------------>
targets = tf.concat([tf.reshape(speech_f[i][self.N_IN - 1][0:self.NEFF],[1, self.NEFF]) for i in range(0, self.batch_size, 1)],0)

The error message is now gone

[zhr1201]

@JH2018-2021 Thx! Could you please list your tf version so that other people with the same version of tensorflow can refer to this issue to solve their problems.

[Luciferhy]

@JYH9351 May I ask which version of tensorflow are you using?

[Luciferhy]

/usr/local/lib/python2.7/dist-packages/librosa/init.py:35: FutureWarning: You are using librosa with Python 2. Please note that librosa 0.7 will be the last version to support Python 2, after which it will require Python 3 or later.
FutureWarning)
11572 speech found
11572 noise found
824 speech found
824 noise found
Traceback (most recent call last):
File "SENN_train.py", line 191, in
train()
File "SENN_train.py", line 122, in train
sess.run(init)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 717, in run
run_metadata_ptr)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 915, in _run
feed_dict_string, options, run_metadata)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 965, in _do_run
target_list, options, run_metadata)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 985, in _do_call
raise type(e)(node_def, op, message)
InvalidArgumentError: No OpKernel was registered to support Op 'FFT' with these attrs. Registered kernels:

     [[Node: FFT_1 = FFT[](Cast_1)]]

Caused by op u'FFT_1', defined at:
File "SENN_train.py", line 191, in
train()
File "SENN_train.py", line 105, in train
images, targets = SE_Net.inputs(data_frames)
File "SENN.py", line 98, in inputs
speech_f0 = tf.fft(tf.cast(raw_speech, tf.complex64))
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/gen_math_ops.py", line 798, in fft
result = _op_def_lib.apply_op("FFT", input=input, name=name)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/op_def_library.py", line 749, in apply_op
op_def=op_def)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 2380, in create_op
original_op=self._default_original_op, op_def=op_def)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 1298, in init
self._traceback = _extract_stack()

InvalidArgumentError (see above for traceback): No OpKernel was registered to support Op 'FFT' with these attrs. Registered kernels:

     [[Node: FFT_1 = FFT[](Cast_1)]]

我遇到一个这个问题，查了一下是由于CPU版上没有FFT操作对象之类的，请问是必须要在GPU上运行吗？ @zhr1201

[Lerry123]

@Luciferhy 你这个问题解决了吗？我也遇到同样的问题。