Open ogrisel opened 1 year ago
Currently we fix n_iter. I think we should instead implement some kind of stochastic convergence criterion (either on the training data or on held out validation split) and instead configure a max_iter and some kind of patience parameter.
n_iter
max_iter
Currently we fix
n_iter. I think we should instead implement some kind of stochastic convergence criterion (either on the training data or on held out validation split) and instead configure amax_iterand some kind of patience parameter.