Improve dataset card: Add task category, library name, and sample usage

This PR enhances the dataset card for the `FreeGraspData` dataset by:
- Adding the `task_categories: robotics` field to the YAML metadata, which improves discoverability on the Hugging Face Hub. The existing `robotics` tag has been moved to this new field.
- Including the `library_name: datasets` field in the metadata to indicate compatibility with the Hugging Face `datasets` library.
- Adding a "Sample Usage" section in the content to provide users with a quick guide on how to load and inspect the dataset, along with a pointer to the main GitHub repository for more detailed examples.

README.md

@@ -1,10 +1,14 @@
 ---
+language:
+- en
 license: cc-by-nc-sa-3.0
-pretty_name: FreeGraspData
+library_name: datasets
 size_categories:
 - n<1K
-tags:
+pretty_name: FreeGraspData
+task_categories:
 - robotics
+tags:
 - grasping
 - language
 - images
@@ -28,9 +32,8 @@ dataset_info:
     dtype: bool
   - name: split
     dtype: string
-language:
-- en
 ---
+
 <h1><img height="50px" width="50px" style="display: inline;" src="https://tev-fbk.github.io/FreeGrasp/static/images/robotic_icon.png"> Free-form language-based robotic reasoning and grasping Dataset: FreeGraspData</h1>
 
 - Homepage: https://tev-fbk.github.io/FreeGrasp/
@@ -106,6 +109,28 @@ Interestingly, despite these variations, the consistently high GPT scores across
 (2) Reimers, Nils, and Iryna Gurevych. "Sentence-bert: Sentence embeddings using siamese bert-networks." (2019).
 (3) Lin, Chin-Yew, and Franz Josef Och. "Automatic evaluation of machine translation quality using longest common subsequence and skip-bigram statistics." Proceedings of the 42nd annual meeting of the association for computational linguistics (ACL-04). 2004.(3)
 
+## Sample Usage
+
+You can easily load and use the dataset with the Hugging Face `datasets` library:
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("FBK-TeV/FreeGraspData")
+
+# Print the dataset structure
+print(dataset)
+
+# Access an example from the 'train' split
+example = dataset["train"][0]
+
+# Print an image and its annotation
+print(f"Image type: {type(example['image'])}")
+print(f"Annotation: {example['annotation']}")
+```
+
+For more detailed usage examples, including installation and running demos, please refer to the [FreeGrasp GitHub repository](https://github.com/tev-fbk/FreeGrasp_code).
+
 ## Data Fields
 ```
 - **__index_level_0__**: An integer representing the unique identifier for each example.