There could be multiple reasons for the reduction of the VAE training loss function to nan, some of them are:
Vanishing gradient: During the training process, if the gradient for a particular weight becomes too small, it can lead to a failure in weight updates during backpropagation, resulting in a nan loss value.
Invalid input: If the input data contains invalid or superfluous values, it can result in nan values.
Overfitting: If the model is overfitted to the training data, the VAE loss value may become too small and may eventually result in a nan value.
Numerical instability: VAEs involve complex math operations, such as log probability calculations, which can cause numerical instability during training and result in nan values.
Incorrect model architecture or hyperparameters: If the model architecture or hyperparameters are not correctly set, it can lead to nan values during training, particularly during validation.
Asked: 2023-05-13 20:44:52 +0000
Seen: 8 times
Last updated: May 13 '23