Create app.py

app.py

@@ -0,0 +1,32 @@
+import gradio as gr
+from setfit import SetFitModel
+
+# Load the pretrained model
+transactional_model = SetFitModel.from_pretrained("edugargar/transactional_model")
+
+def predict_and_score(text):
+    # Get predicted class
+    preds = transactional_model([text])
+    predicted_class = preds[0]
+
+    # Get probability distribution for each class
+    probs = transactional_model.predict_proba([text])
+    probability_distribution = probs[0]  # This is usually a list of probabilities for each class
+
+    # Convert probability distribution to a readable format
+    probability_str = ", ".join(f"{p:.4f}" for p in probability_distribution)
+
+    return predicted_class, probability_str
+
+# Build the Gradio interface
+with gr.Blocks() as demo:
+    gr.Markdown("## Transactional Model Prediction")
+
+    text_input = gr.Textbox(label="Enter your text here:")
+    predict_button = gr.Button("Predict")
+    prediction_output = gr.Textbox(label="Predicted Class:")
+    score_output = gr.Textbox(label="Score (Probability Distribution):")
+
+    predict_button.click(fn=predict_and_score, 
+                         inputs=[text_input], 
+                         outputs=[prediction_output, score_output])