Open chen955 opened 4 years ago
You have to set up the labels.csv
with multiple labels, and your train.csv
and val.csv
as described in the readme for multilabel classification. Then you need to set the multilabel
argument to True
in the learner builder. I currently have a pull request (#190) that edits the modeling.py
script to use the correct definitions for Albert - you can do this yourself by following issue #172.
How can I run Multi-Label text classification with albert using fast-bert?