Twitter Sarcasm Detection Using Transformers

This repository is based on the Transformers library by HuggingFace. It is intended as a starting point for anyone who wishes to use Transformer models in text classification tasks.

Table of contents

• Setup
  • With Conda
• Usage
  • Twitter Sarcasm Dataset
  • Current Pretrained Models
  • Custom Datasets
  • Evaluation Metrics
• Acknowledgements

Setup

With Conda

1. Install Anaconda or Miniconda Package Manager from here
2. Create a new virtual environment and install packages.
   conda create -n transformers python pandas tqdm jupyter
conda activate transformers
   If using cuda:
   conda install pytorch cudatoolkit=10.0 -c pytorch
   else:
   conda install pytorch cpuonly -c pytorch
conda install -c anaconda scipy
conda install -c anaconda scikit-learn
pip install transformers or download source code from Transformers*
3. Clone repo. git clone https://github.com/muhammadadyl/SarcasmDetection.git

*Important if you wanted to run DistilRoBERTa (soft release)

Usage

Twitter Sarcasm Dataset

If you are doing it manually;

Files are already available for use in data/ folder with name train.csv and test.csv.

Once the download is complete, you can run the data_prep_sarcasm.ipynb notebook to get the data ready for training.

Finally, you can run the run_model.ipynb notebook to fine-tune a Transformer model on the Twitter Dataset and evaluate the results.

Current Pretrained Models

The table below shows the currently available model types and their models. You can use any of these by setting the model_type and model_name in the args dictionary. For more information about pretrained models, see HuggingFace docs.

Architecture	Model Type	Model Name	Details
BERT	bert	bert-base-cased	12-layer, 768-hidden, 12-heads, 110M parameters. Trained on cased English text.
XLNet	xlnet	xlnet-base-cased	12-layer, 768-hidden, 12-heads, 110M parameters. XLNet English model
RoBERTa	roberta	roberta-base	125M parameters RoBERTa using the BERT-base architecture
DistilBERT	distilbert	distilbert-base-uncased	6-layer, 768-hidden, 12-heads, 66M parameters DistilBERT uncased base model
DistilRoBERTa	distilroberta	distilroberta-base	6-layer, 768-hidden, 12-heads, 82M parameters DistilRoBERTa-base model.

Note: DistilRoBERTa is in a soft release as of the day this repo published, to run this model you need to explicitly install Transformer library from Hugging Face's Repository. Installing through pip won't work here.

Custom Datasets

When working with your own datasets, you can create a script/notebook similar to data_prep_sarcasm.ipynb that will convert the dataset to a Transformer ready format.

The data needs to be in tsv format, with four columns, and no header.

This is the required structure.

• guid: An ID for the row.
• label: The label for the row (should be an int).
• alpha: A column of the same letter for all rows. Not used in classification but still expected by the DataProcessor.
• text: The sentence or sequence of text.

Evaluation Metrics

The evaluation process in the run_model.ipynb notebook outputs the confusion matrix, and the Matthews correlation coefficient. If you wish to add any more evaluation metrics, simply edit the get_eval_reports() function in the notebook. This function takes the predictions and the ground truth labels as parameters, therefore you can add any custom metrics calculations to the function as required.

Acknowledgements

None of this would have been possible without the hard work by the HuggingFace team in developing the Transformers library. Further, I would also like to thanks Thilina Rajapakse for his Text Classification Code and blog