Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -152,12 +152,20 @@ configs:
|
|
| 152 |
|
| 153 |
---
|
| 154 |
|
| 155 |
-
# Datasets
|
| 156 |
This directory includes all the downstream task datasets of the LucaOne paper (https://www.nature.com/articles/s42256-025-01044-4).
|
| 157 |
|
|
|
|
| 158 |
|
| 159 |
-
|
| 160 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
|
| 162 |
```python
|
| 163 |
from datasets import load_dataset
|
|
|
|
| 152 |
|
| 153 |
---
|
| 154 |
|
| 155 |
+
# LucaOne-Downstream-Task-Datasets
|
| 156 |
This directory includes all the downstream task datasets of the LucaOne paper (https://www.nature.com/articles/s42256-025-01044-4).
|
| 157 |
|
| 158 |
+
## Dataset Structure
|
| 159 |
|
| 160 |
+
This dataset includes three splits and a label info file:
|
| 161 |
+
|
| 162 |
+
- `train`(training dataset)
|
| 163 |
+
- `dev` (validation dataset)
|
| 164 |
+
- `test` (testing dataset)
|
| 165 |
+
- `label.txt` (label name to label index in `*.csv`, label_idx=${the line number of label_name - 1})
|
| 166 |
+
|
| 167 |
+
|
| 168 |
+
## Usage:
|
| 169 |
|
| 170 |
```python
|
| 171 |
from datasets import load_dataset
|