Persian_Models_Fine-tuning

Fine-tuning Parsbert for entailment, sentiment analysis, and named entity recognition, and Fine-tuning bert2bert for summarization

This repository contains Jupyter notebooks for fine-tuning large language models (LLMs) on various tasks including:

• Entailment
• Sentiment Analysis
• Summarization
• Named Entity Recognition (NER)

Notebooks

• entailment_fine_tuning.ipynb: Fine-tuning "persiannlp/parsbert-base-parsinlu-entailment" for entailment tasks.
• sentiment_analysis_fine_tuning.ipynb: Fine-tuning "nimaafshar/parsbert-fa-sentiment-twitter" for sentiment analysis.
• summarization_fine_tuning.ipynb: Fine-tuning "m3hrdadfi/bert2bert-fa-wiki-summary" for summarization tasks.
• ner_fine_tuning.ipynb: Fine-tuning "HooshvareLab/bert-base-parsbert-armanner-uncased" for named entity recognition.

Requirements

Ensure you have the following libraries installed:

• Transformers
• datasets
• torch
• tensorflow (if applicable)
• Any other dependencies used in your notebooks.

Usage

To run the notebooks, you can use Google Colab or Jupyter Notebook.

1. Open the notebook in Colab by selecting Open in Colab from the File menu.
2. Follow the instructions in each notebook to set up the environment and run the fine-tuning process.