Climate Change Fact Checker

Using machine learning to accurately analyze climate change statements

Using the model

Prerequesites

• Install datasets
• Install transformers
• Install huggingface_hub

Executing the program

Copy GitHub program code OR access the Colab notebook here: https://colab.research.google.com/drive/1yrw8CbFNBQWkqip3FXEuqNznbHdWUxPs?usp=sharing

1. Run code in order provided
2. Copy Hugging Face token from tokens page when prompt accessible
3. To directly test the model - run the last code cell with your own input in the pipeline

Authors

Rayyan Mohamed

• student
• contact at: rayyanym17@gmail.com

License

This project is licensed under the MIT License - see the LICENSE.md file for details

Acknowledgments

Sources (also listed in "main.py"):

Access final model: https://huggingface.co/IntroToProgramming/model

climate_fever dataset: https://huggingface.co/datasets/climate_fever

Research paper for background of dataset: https://www.arxiv-vanity.com/papers/2012.00614/

Implementing DistilBERT: https://www.youtube.com/watch?v=s3LBdmZb00g

Plotting graphs: https://www.geeksforgeeks.org/graph-plotting-in-python-set-1/

Creating an ML project: https://www.youtube.com/watch?v=nKW8Ndu7Mjw

Using Colab: https://www.youtube.com/watch?v=inN8seMm7UI

Tokenization: https://www.youtube.com/watch?v=VFp38yj8h3A

Hugging Face notebook login hub: https://huggingface.co/docs/hub/models-adding-libraries