search

KhalilMrini / LAL-Parser

Neural Adobe-UCSD Parser, the current State of the Art in Constituency and Dependency Parsing.
138 stars 24 forks source link

How many epochs did the training take to converge? #23

Open huberemanuel opened 3 years ago

huberemanuel commented 3 years ago

The script to train the best parser mentions 100 epochs. The trained ended at the 100th epoch or an early stopping was triggered at a certain epoch?