I am new on this forum and i could not find a topic about my issue,
I am not a programmer and i am not completley sure what problem i have,
I have TensorFlow version: 2.10.0. And python 3
The problem seems to be insid the generarion itself, my latent vector seems to be working but all the tensors are having the exact same value and it is very close to 0, during training all the tensors are having different values so i believe the model is working, anyone have any ideas or have exerienced the same problem?
I am training with one code then save the model as .keras and then loading with another code.
Am greatful for any ideas
Hi @the_vicke, This may be due to the data you are passing to the model, Here are a few things you can consider make sure your data is properly labeled, normalizing the data, using different model architecture, etc. Thank You
This is int16 files converted to float32 and it is normalized to between 1 and -1, and regarding labeling this is only waveform data…
Hi @the_vicke, If possible could you please share the stand alone code to reproduce the issue. Thank You!
Went through my codes and found the problem, i made a mistake where i missmatched the definition for length so it did fit the input shape but did not work.
Since i am new to coding i am learning on the way