LIPI FinSim3 Hypernym

🚀 Financial Hypernym Detection Model

This project provides a financial hypernym detection model, which may have some differences in performance due to different training settings compared to the original model.

🚀 Quick Start

Import the Model

Use the following code to import this in Transformers:

from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("sohomghosh/LIPI_FinSim3_Hypernym")

model = AutoModel.from_pretrained("sohomghosh/LIPI_FinSim3_Hypernym")


#Using SentenceTransformers
from sentence_transformers import SentenceTransformer
model_finlipi = SentenceTransformer('sohomghosh/LIPI_FinSim3_Hypernym')

📚 Documentation

Note: This model may not perfectly replicate the numbers mentioned in the paper (\cite{chopra-ghosh-2021-term}) as unlike the original one it has been trained with lower batches sizes for fewer epochs.

The source code for training will son be made available in https://github.com/sohomghosh/FinSim_Financial_Hypernym_detection

📄 License

This project is licensed under the MIT license.

📖 Citation

@inproceedings{chopra-ghosh-2021-term,
    title = "Term Expansion and {F}in{BERT} fine-tuning for Hypernym and Synonym Ranking of Financial Terms",
    author = "Chopra, Ankush  and
      Ghosh, Sohom",
    booktitle = "Proceedings of the Third Workshop on Financial Technology and Natural Language Processing (FinNLP@IJCAI 2021)",
    month = "Aug ",
    year = "2021",
    address = "Online",
    publisher = "-",
    url = "https://aclanthology.org/2021.finnlp-1.8",
    pages = "46--51",
}