Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -104,7 +104,14 @@ To use the raw data, you will need to use the `huggingface_hub`:
|
|
104 |
```python
|
105 |
from huggingface_hub import snapshot_download
|
106 |
|
107 |
-
snapshot_download(repo_id="McGill-NLP/WebLINX-full", repo_type="dataset", local_dir="./
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
108 |
```
|
109 |
|
110 |
For more information on how to use this data using our [official library](https://github.com/McGill-NLP/WebLINX), please refer to the [WebLINX documentation](https://mcgill-nlp.github.io/weblinx/docs).
|
|
|
104 |
```python
|
105 |
from huggingface_hub import snapshot_download
|
106 |
|
107 |
+
snapshot_download(repo_id="McGill-NLP/WebLINX-full", repo_type="dataset", local_dir="./wl_data")
|
108 |
+
|
109 |
+
# You can download specific demos, for example
|
110 |
+
demo_names = ['saabwsg', 'ygprzve', 'iqaazif'] # 3 random demo from valid
|
111 |
+
patterns = [f"demonstrations/{name}/*" for name in demo_names]
|
112 |
+
snapshot_download(
|
113 |
+
repo_id="McGill-NLP/WebLINX-full", repo_type="dataset", local_dir="./wl_data", allow_patterns=patterns
|
114 |
+
)
|
115 |
```
|
116 |
|
117 |
For more information on how to use this data using our [official library](https://github.com/McGill-NLP/WebLINX), please refer to the [WebLINX documentation](https://mcgill-nlp.github.io/weblinx/docs).
|