Update app.py

app.py

@@ -7,20 +7,38 @@ from transformers import pipeline
 #     output = classifier(text, candidate_labels)
 #     return output
 
+# def classify(text):
+#     classifier = pipeline("zero-shot-classification")
+#     candidate_labels = ["positive", "negative", "neutral"]
+#     output = classifier(text, candidate_labels)
+#     output_labels = [label['label'] for label in output['labels']]
+#     output_scores = [score for score in output['scores']]
+#     sorted_output = sorted(zip(output_labels, output_scores), key=lambda x: x[1], reverse=True)
+#     return sorted_output[:3]
+
+# demo = gr.Interface(fn=classify,
+#                     inputs=gr.Textbox(label="Enter text to classify"),
+#                     outputs=gr.Label(num_top_classes=3))
+    
+
+# demo.launch()
+
+
+
+classifier = pipeline("zero-shot-classification")
+
 def classify(text):
-    classifier = pipeline("zero-shot-classification")
     candidate_labels = ["positive", "negative", "neutral"]
     output = classifier(text, candidate_labels)
-    output_labels = [label['label'] for label in output['labels']]
-    output_scores = [score for score in output['scores']]
-    sorted_output = sorted(zip(output_labels, output_scores), key=lambda x: x[1], reverse=True)
-    return sorted_output[:3]
+    # Process the output to match Gradio's expected input format for gr.Label
+    labels = output['labels']
+    scores = output['scores']
+    # Construct a simple string representation of top classifications
+    top_classes = ', '.join([f"{labels[i]}: {scores[i]:.2f}" for i in range(len(labels))])
+    return top_classes
 
 demo = gr.Interface(fn=classify,
-                    inputs=gr.Textbox(label="Enter text to classify"),
-                    outputs=gr.Label(num_top_classes=3))
-    
+                    inputs=gr.Textbox(label="Enter something"),
+                    outputs=gr.Label())
 
 demo.launch()
-
-