Hugging Face's logo

AdapterHub
/
bert-base-uncased_nli_rte_houlsby

Text Classification
Adapters
bert
adapterhub:nli/rte
Model card Files Community
bert-base-uncased_nli_rte_houlsby / README.md
calpt's picture
calpt
Add adapter bert-base-uncased_nli_rte_houlsby version 1
86971ee verified
preview code
|
raw
history blame contribute delete
1.81 kB
metadata
tags:
  - bert
  - text-classification
  - adapter-transformers
  - adapterhub:nli/rte
license: apache-2.0

Adapter bert-base-uncased_nli_rte_houlsby for bert-base-uncased

Adapter in Houlsby architecture trained on the RTE task for 20 epochs with early stopping and a learning rate of 1e-4. See https://arxiv.org/pdf/2007.07779.pdf.

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_nli_rte_houlsby")
model.set_active_adapters(adapter_name)

Architecture & Training

  • Adapter architecture: houlsby
  • Prediction head: classification
  • Dataset: RTE

Author Information

Citation 

@article{pfeiffer2020AdapterHub,
    title={AdapterHub: A Framework for Adapting Transformers},
    author={Jonas Pfeiffer and
            Andreas R\"uckl\'{e} and
            Clifton Poth and
            Aishwarya Kamath and
            Ivan Vuli\'{c} and
            Sebastian Ruder and
            Kyunghyun Cho and
            Iryna Gurevych},
    journal={arXiv preprint},
    year={2020},
    url={https://arxiv.org/abs/2007.07779}
}

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