Update README.md

README.md

@@ -1,15 +1,40 @@
 ---
 license: cc-by-4.0
+language: en
+pretty_name: Primock-57
+task_categories:
+  - dialog
+  - healthcare
+size_categories:
+  - n<100
 ---
 
 # Primock-57
 
 This is a version of the [Primock-57](https://github.com/babylonhealth/primock57) dataset converted to the [`sdialog`](https://github.com/idiap/sdialog) `Dialog` JSON format.  
-It allows easy loading as dialog objects via `Dialog.from_file()`.
+It allows easy loading as dialog objects via `Dialog.from_file()` or `Dialog.from_huggingface`.
 
-## How to use
+The dataset contains **57** simulated primary-care style consultations (doctor ↔ patient). Each dialog is stored as an individual JSON file and includes a sequence of turns labelled by speaker.
 
-First, download the dataset from the Hugging Face Hub to a local directory (e.g., `"Primock-57"`):
+---
+## How to use (Automatically, using `Dialog.from_huggingface()`)
+
+Simply use the [`Dialog.from_huggingface()`](https://sdialog.readthedocs.io/en/latest/api/sdialog.html#sdialog.Dialog.from_huggingface) as in the following example:
+
+```python
+from sdialog import Dialog
+
+primock57_dialogs = Dialog.from_huggingface("sdialog/Primock-57")
+
+print("Number of dialogs:", len(primock57_dialogs))
+```
+
+---
+## How to use (Manually, using `Dialog.from_file()`)
+
+Alternatively, if you want to do it manually, you have to:
+
+1. First, download the dataset from the Hugging Face Hub to a local directory (e.g., `"Primock-57"`):
 
 ```python
 from huggingface_hub import snapshot_download
@@ -21,7 +46,7 @@ snapshot_download(
 )
 ```
 
-Then, load the dataset as `Dialog` objects:
+2. Then, load the dataset as `Dialog` objects:
 
 ```python
 from sdialog import Dialog