Update app.py

app.py

@@ -108,11 +108,11 @@ def infer(video_file):
         logits = outputs.logits
     softmax_scores = torch.nn.functional.softmax(logits, dim=-1).squeeze(0)
     confidences = {LABELS[i]: float(softmax_scores[i]) for i in range(len(LABELS))}
-    return confidences
+    return {confidences}
 
 gr.Interface(
     fn=infer,
-    inputs=gr.Video(type="file"),
+    inputs=gr.Video(),
     outputs=gr.Label(num_top_classes=3),
     title="VideoMAE fine-tuned on a subset of UCF-101",
     description=(