Edit on GitHub 

model = AutoAdapterModel.from_pretrained("facebook/bart-base")
config = AdapterConfig.load("houlsby", non_linearity="swish", reduction_factor=16)
model.load_adapter("nli/qnli@ukp", config=config)

Description

Adapter for bart-base in Houlsby architecture trained on the QNLI dataset for 15 epochs with early stopping and a learning rate of 1e-4.

Properties

Pre-trained model
facebook/bart-base
Adapter type
Task
Prediction Head
  Yes
Task
Natural Language Inference
Dataset
QNLI

Architecture

Name
houlsby
Non-linearity
swish
Reduction factor
16
{
  "ln_after": false,
  "ln_before": false,
  "mh_adapter": true,
  "output_adapter": true,
  "adapter_residual_before_ln": false,
  "non_linearity": "swish",
  "original_ln_after": true,
  "original_ln_before": false,
  "reduction_factor": 16,
  "residual_before_ln": true
}

Author

  Name
Clifton Poth
  E-Mail
calpt@mail.de
  Web
https://adapterhub.ml
  GitHub
calpt
  Twitter
@clifapt

Versions

Identifier Comment Score Download
1 DEFAULT

Citations

Architecture
@misc{houlsby2019parameterefficient,
  title={Parameter-Efficient Transfer Learning for NLP},
  author={Neil Houlsby and Andrei Giurgiu and Stanislaw Jastrzebski and Bruna Morrone and Quentin de Laroussilhe and Andrea Gesmundo and Mona Attariyan and Sylvain Gelly},
  year={2019},
  eprint={1902.00751},
  archivePrefix={arXiv},
  primaryClass={cs.LG}
}
Task
@article{wang2018glue,
  title={Glue: A multi-task benchmark and analysis platform for natural language understanding},
  author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R},
  journal={arXiv preprint arXiv:1804.07461},
  year={2018}
}