valueError about audio-classification HOT 6 CLOSED

I've ran your data. It works just fine as long as you have two directories for each class in the wavfiles directory for example. water and no_water. This error suggests you did not clean your audio first and are passing several seconds of data.

from audio-classification.

Ok, but seth i have cleaned my data the file which i have sent you, you can see clean directory where i have my data but one confusion is when i pass
self.wav_paths = wav_paths
then i got
X[i,] = wav.reshape(1, -1)
ValueError: could not broadcast input array from shape (1,3424256) into shape (1,16000)

but when i pass :-
self.wav_paths =
('C:/Users/i00504285/Desktop/Aditya/Sound_Analysis/CSF48/wave_files/clean')

where i have my clean data then i got
IndexError: index 56 is out of bounds for axis 0 with size 54
the above error

from audio-classification.

wav_paths should be a list of file paths to all the wav files in the src_root from argparser at the bottom. The src_root should be your clean directory here. That second way of manually setting it to a directory is wrong. I think it's just looking in the wrong directory. Check your src_root.

from audio-classification.

Ya it solved thank you, but i have one new issue which is:-
tensorflow.python.framework.errors_impl.InvalidArgumentError: logits and labels must be broadcastable: logits_size=[16,10] labels_size=[16,2]

[[node loss/softmax_loss/softmax_cross_entropy_with_logits (defined at C:\Users\i00504285\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow_core\python\framework\ops.py:1751) ]] [Op:__inference_distributed_function_4563]

Function call stack:
distributed_function.

why this happened and how to solve this?

from audio-classification.

It looks like you are using a model with the softmax output for 10 classes while you have 2 classes.

What file are you running that creates this error? line number?

from audio-classification.

I solved it thank you 👍

from audio-classification.