---
license: mit
---
# Instructions to Run the Model on Sample Dataset

## 1. Using Pegasus Model
- Open `code_pegasus.py`.
- Run the script. (We have provided the path to the sample data `9.txt`.)
- To test on other conversation files, provide the path to the desired data files.  
  *(You can find our sample dataset in the `Sample Dataset` directory.)*

## 2. Using BART-large Model
- Open `code_bart.py`.
- Run the script. (We have provided the path to the sample data `9.txt`.)
- To test on other conversation files, provide the path to the desired data files.  
  *(You can find our sample dataset in the `Sample Dataset` directory.)*

---

### Note:
The `Sample Dataset` includes:
- An anonymized our sample dataset named `conv_1`, `conv_2`, and `conv_3`.
- A Chinese language dataset translated to English named `conv_1001`, `conv_1002`, and `conv_1003` (D4).

---

### Important:
- This dataset is provided **only for research purposes**.  
  To get full access to the dataset, please fill out the Google form:  
  [Request Dataset Access](https://docs.google.com/forms/d/e/1FAIpQLSe_unAObNmRJjKrmluRC-UZZUTyKd9aujVLNUeTLgPtB-9BSw/viewform?usp=sf_link)

- For any queries related to dataset and model access, please contact:  
  **nilesh21@iiserb.ac.in**