toniwuest's picture
remove dataset2
ede0c80
|
raw
history blame
365 Bytes
metadata
language:
  - en
tags:
  - question-answering
datasets:
  - nq
  - natural-question
  - natural-question-short
metrics:
  - f1
  - em

Model based on distilbert-base-uncased model trained on natural question short dataset. Trained for one episode with AdamW optimizer and learning rate of 5e-03 and no warmup steps. We achieved a f1 score of ... and an em score of ...