search

ybracke / transnormer

A lexical normalizer for historical spelling variants using a transformer architecture.
GNU General Public License v3.0
6 stars 1 forks source link

Exclude long input sequences during training #78

Closed ybracke closed 8 months ago

ybracke commented 8 months ago

Use the sort function of datasets as described in #76
Set padding to "longest" in training_config.toml Specify truncation in training_config.toml not hard-coded in train_model.py

ybracke commented 8 months ago

Closed by #80