View on huggingface.co 

model = AutoAdapterModel.from_pretrained("roberta-base")
model.load_adapter("AdapterHub/roberta-base-pf-emo", source="hf")

Description

Adapter AdapterHub/roberta-base-pf-emo for roberta-base

An adapter for the roberta-base model that was trained on the emo dataset and includes a prediction head for classification.

This adapter was created for usage with the adapter-transformers library.

Usage

First, install adapter-transformers:

pip install -U adapter-transformers

Note: adapter-transformers is a fork of transformers that acts as a drop-in replacement with adapter support. More

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

from transformers import AutoModelWithHeads

model = AutoModelWithHeads.from_pretrained("roberta-base")
adapter_name = model.load_adapter("AdapterHub/roberta-base-pf-emo", source="hf")
model.active_adapters = adapter_name

Architecture & Training

The training code for this adapter is available at https://github.com/adapter-hub/efficient-task-transfer. In particular, training configurations for all tasks can be found here.

Evaluation results

Refer to the paper for more information on results.

Citation

If you use this adapter, please cite our paper "What to Pre-Train on? Efficient Intermediate Task Selection":

@inproceedings{poth-etal-2021-pre,
    title = "{W}hat to Pre-Train on? {E}fficient Intermediate Task Selection",
    author = {Poth, Clifton  and
      Pfeiffer, Jonas  and
      R{"u}ckl{'e}, Andreas  and
      Gurevych, Iryna},
    booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2021",
    address = "Online and Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.emnlp-main.827",
    pages = "10585--10605",
}

Properties

Pre-trained model
roberta-base
Adapter type
Task
Prediction Head
  Yes
Task
Emotion Classification
Dataset
EmoContext

Architecture

{
  "adapter_residual_before_ln": false,
  "cross_adapter": false,
  "inv_adapter": null,
  "inv_adapter_reduction_factor": null,
  "leave_out": [],
  "ln_after": false,
  "ln_before": false,
  "mh_adapter": false,
  "non_linearity": "relu",
  "original_ln_after": true,
  "original_ln_before": true,
  "output_adapter": true,
  "reduction_factor": 16,
  "residual_before_ln": true
}

Citations

Task
@inproceedings{chatterjee-etal-2019-semeval,
    title={SemEval-2019 Task 3: EmoContext Contextual Emotion Detection in Text},
    author={Ankush Chatterjee and Kedhar Nath Narahari and Meghana Joshi and Puneet Agrawal},
    booktitle={Proceedings of the 13th International Workshop on Semantic Evaluation},
    year={2019},
    address={Minneapolis, Minnesota, USA},
    publisher={Association for Computational Linguistics},
    url={https://www.aclweb.org/anthology/S19-2005},
    doi={10.18653/v1/S19-2005},
    pages={39--48},
    abstract={In this paper, we present the SemEval-2019 Task 3 - EmoContext: Contextual Emotion Detection in Text. Lack of facial expressions and voice modulations make detecting emotions in text a challenging problem. For instance, as humans, on reading ''Why don't you ever text me!'' we can either interpret it as a sad or angry emotion and the same ambiguity exists for machines. However, the context of dialogue can prove helpful in detection of the emotion. In this task, given a textual dialogue i.e. an utterance along with two previous turns of context, the goal was to infer the underlying emotion of the utterance by choosing from four emotion classes - Happy, Sad, Angry and Others. To facilitate the participation in this task, textual dialogues from user interaction with a conversational agent were taken and annotated for emotion classes after several data processing steps. A training data set of 30160 dialogues, and two evaluation data sets, Test1 and Test2, containing 2755 and 5509 dialogues respectively were released to the participants. A total of 311 teams made submissions to this task. The final leader-board was evaluated on Test2 data set, and the highest ranked submission achieved 79.59 micro-averaged F1 score. Our analysis of systems submitted to the task indicate that Bi-directional LSTM was the most common choice of neural architecture used, and most of the systems had the best performance for the Sad emotion class, and the worst for the Happy emotion class}
}