| # Data for Paper: *Dynamic Vocabularies for Learned Sparse Retrieval with Entities* |
|
|
| This guide explains how to download the data used in our paper. |
|
|
| --- |
|
|
| ## Setup Instructions |
|
|
| 1. **Clone the DyVo repository** |
| ``` |
| git clone https://github.com/thongnt99/DyVo |
| ``` |
|
|
| 2. **Create a data directory inside the project** |
| ``` |
| cd DyVo |
| mkdir dyvo_data |
| cd dyvo_data |
| ``` |
|
|
| 3. **Download the data from Hugging Face** |
|
|
| Make sure you have the Hugging Face CLI installed. If not, install it with: |
| ``` |
| pip install huggingface_hub |
| ``` |
|
|
| Then run: |
| ``` |
| huggingface-cli download lsr42/dyvo_data |
| ``` |
|
|
| --- |
|
|
| ## Notes |
|
|
| - You may need to log in to Hugging Face before downloading: |
| ``` |
| huggingface-cli login |
| ``` |
| - The downloaded files will be stored locally; check the Hugging Face CLI documentation for cache settings if needed. |
| - It is recommended to use Python 3.7 or later. |
|
|
| --- |
|
|
