The music_genre_classification's discuss from cvxtz

cvxtz / music_genre_classification Goto Github PK

View Code? Open in Web Editor NEW

music genre classification : LSTM vs Transformer

License: MIT License

Python 99.39% Shell 0.61%

music_genre_classification's Issues

having problems with rnn.h5 and transformer.h5 model files

In training part, it requires h5_name which is "rnn.h5" and "transformer.h5",
so I created h5 files with models.py using model.save(), but the following error occurs:

_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
input_1 (InputLayer)         [(None, None, 128)]       0         
_________________________________________________________________
bidirectional (Bidirectional (None, None, 256)         198144    
_________________________________________________________________
bidirectional_1 (Bidirection (None, None, 256)         296448    
_________________________________________________________________
dropout (Dropout)            (None, None, 256)         0         
_________________________________________________________________
global_average_pooling1d (Gl (None, 256)               0         
_________________________________________________________________
dense (Dense)                (None, 652)               167564    
_________________________________________________________________
dense_1 (Dense)              (None, 163)               106439    
=================================================================
Total params: 768,595
Trainable params: 768,595
Non-trainable params: 0
_________________________________________________________________

model's layer properties are same as original post, but

File "C:\miniconda\lib\site-packages\tensorflow_core\python\framework\tensor_shape.py", line 1115, in assert_is_compatible_with
    raise ValueError("Shapes %s and %s are incompatible" % (self, other))
ValueError: Shapes (256, 652) and (256, 64) are incompatible

I think these error started with wrong h5 files but I can't find way to solve these errors.

Recommend Projects