6 ms·
It's a hyperparameter much like learning rate. If the learning rate is too high, the training process would not work either. Addressing this is just a matter of
by hodgehog11 1y ago
It's a hyperparameter much like learning rate. If the learning rate is too high, the training process would not work either. Addressing this is just a matter of a grid search.