Hi @astorfi
In the development step, i tried running train_softmax.py over VoxCeleb2 dataset and then i got nan output (logits variable in code). how to solve this?
@hamiGH It is hard to say as I do not know the details of your implementation. My guess is the learning rate must be adjusted. Furthermore, please make sure the batch-normalization is active.
Hi @astorfi In the development step, i tried running train_softmax.py over VoxCeleb2 dataset and then i got nan output (logits variable in code). how to solve this?