kudkudak / common-sense-prediction

Common sense prediction using DL.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Grid to pick hyperparameter for each model

kudkudak opened this issue · comments

Proper fitting of models to old and new dev split. Right now hyperparameters are shared, which might be bias heavily results.

For each model tune learning rate and l2 regularization.

For NAACL we will just fix hyperparams to the ones tuned for DNNs. Performance of factorized and prototypical is approximately stable wrt to l2_a and lr hyperparameter. I haven't checked others.

If we submit it somewhere else I would tune these hyperparams, but closing for now

But it actually seems important for glove.. will try to run proper grid.