Adapters
bert
adapterhub:comsense/siqa
calpt's picture
Add adapter bert-base-uncased-siqa_pfeiffer version AdapterFusion
aa4e831 verified
|
raw
history blame
1.58 kB
metadata
tags:
  - bert
  - adapterhub:comsense/siqa
  - adapter-transformers
datasets:
  - social_i_qa
license: apache-2.0

Adapter bert-base-uncased-siqa_pfeiffer for bert-base-uncased

Pfeiffer Adapter trained on Social IQA.

This adapter was created for usage with the Adapters library.

Usage

First, install adapters:

pip install -U adapters

Now, the adapter can be loaded and activated like this:

from adapters import AutoAdapterModel

model = AutoAdapterModel.from_pretrained("bert-base-uncased")
adapter_name = model.load_adapter("AdapterHub/bert-base-uncased-siqa_pfeiffer")
model.set_active_adapters(adapter_name)

Architecture & Training

  • Adapter architecture: pfeiffer
  • Prediction head: None
  • Dataset: Social IQA

Author Information

Citation

@article{Pfeiffer2020AdapterFusion,
author = {Pfeiffer, Jonas and Kamath, Aishwarya and R{\"{u}}ckl{\'{e}}, Andreas and Cho, Kyunghyun and Gurevych, Iryna},
journal = {arXiv preprint},
title = {{AdapterFusion}:  Non-Destructive Task Composition for Transfer Learning},
 url       = {https://arxiv.org/pdf/2005.00247.pdf},
year = {2020}
}

This adapter has been auto-imported from https://github.com/Adapter-Hub/Hub/blob/master/adapters/ukp/bert-base-uncased-siqa_pfeiffer.yaml.