| | --- |
| | dataset_info: |
| | features: |
| | - name: text |
| | dtype: string |
| | - name: label |
| | dtype: string |
| | splits: |
| | - name: train |
| | num_bytes: 87123 |
| | num_examples: 256 |
| | - name: test |
| | num_bytes: 162998 |
| | num_examples: 512 |
| | - name: val |
| | num_bytes: 19700 |
| | num_examples: 64 |
| | download_size: 160707 |
| | dataset_size: 269821 |
| | configs: |
| | - config_name: default |
| | data_files: |
| | - split: train |
| | path: data/train-* |
| | - split: test |
| | path: data/test-* |
| | - split: val |
| | path: data/val-* |
| | --- |
| | |
| | # EuroEval Test Evaluation Dataset |
| |
|
| | This is a dataset with a `euroeval_config.py` configuration, enabling evaluation of it directly with the [EuroEval framework](https://euroeval.com/). |
| |
|
| | You can evaluate it as follows: |
| |
|
| | 1. Install EuroEval with `pip install euroeval[all]` |
| | 2. Benchmark a model on this dataset as `euroeval --dataset EuroEval/test_dataset --model <model-id>` |
| |
|
| | Read more about how to use the EuroEval Python package in [this user guide](https://euroeval.com/python-package/). |
