Added a default value to the interactive example

app.py

@@ -1,6 +1,51 @@
 import evaluate
-from evaluate.utils import launch_gradio_widget
+from evaluate.utils import parse_readme, infer_gradio_input_types, json_to_string_type,parse_gradio_data
+import re
+import sys
+from pathlib import Path
+
+from evaluate.utils.logging import get_logger
+
+logger = get_logger(__name__)
+REGEX_YAML_BLOCK = re.compile(r"---[\n\r]+([\S\s]*?)[\n\r]+---[\n\r]")
 
 module = evaluate.load("nDCG.py")
 
-launch_gradio_widget(module)
+def launch_gradio_widget(metric):
+    """Launches `metric` widget with Gradio."""
+
+    try:
+        import gradio as gr
+    except ImportError as error:
+        logger.error("To create a metric widget with Gradio make sure gradio is installed.")
+        raise error
+
+    local_path = Path(sys.path[0])
+    # if there are several input types, use first as default.
+    if isinstance(metric.features, list):
+        (feature_names, feature_types) = zip(*metric.features[0].items())
+    else:
+        (feature_names, feature_types) = zip(*metric.features.items())
+    gradio_input_types = infer_gradio_input_types(feature_types)
+
+    def compute(data):
+        return metric.compute(**parse_gradio_data(data, gradio_input_types))
+
+    iface = gr.Interface(
+        fn=compute,
+        inputs=gr.inputs.Dataframe(
+            headers=feature_names,
+            col_count=len(feature_names),
+            row_count=2,
+            datatype=json_to_string_type(gradio_input_types),
+            default=[['[1,2,3]','[1,2,3]'],['[1,1,0]','[0,1,1]']]
+        ),
+        outputs=gr.outputs.Textbox(label=metric.name),
+        description=metric.info.description,
+        title=f"Metric: {metric.name}",
+        article=parse_readme(local_path / "README.md"),
+        )
+
+    iface.launch()
+
+x = launch_gradio_widget(module)

nDCG.py

@@ -18,6 +18,7 @@ scores in the ranking
 
 References
     ----------
+    
     `Wikipedia entry for Discounted Cumulative Gain
     <https://en.wikipedia.org/wiki/Discounted_cumulative_gain>`_
 
@@ -27,7 +28,7 @@ References
 
     Wang, Y., Wang, L., Li, Y., He, D., Chen, W., & Liu, T. Y. (2013, May).
     A theoretical analysis of NDCG ranking measures. In Proceedings of the 26th
-    Annual Conference on Learning Theory (COLT 2013)
+    Annual Conference on Learning Theory (COLT 2013).
 
     McSherry, F., & Najork, M. (2008, March). Computing information retrieval
     performance measures efficiently in the presence of tied scores. In