Different results from allennlp tune
and allennlp retrain
with transformers
#45
Labels
bug
Something isn't working
allennlp tune
and allennlp retrain
with transformers
#45
When I am tuning a transformer model, I get different results from
allennlp tune
andallennlp retrain
with the same hyperparameters.I found this is caused by
allennlp.common.cached_transformers
module, which only constructs the model in the first trial (which would consume some random numbers), and uses the cached model in trials afterwards (which would not consume random numbers), leading to inconsistent results betweentune
andretrain
.The text was updated successfully, but these errors were encountered: