Update app.py

app.py

@@ -31,6 +31,8 @@ def predict(context,question):
     sparse_end_time = time.perf_counter()
     sparse_duration = (sparse_end_time - sparse_start_time) * 1000
     sparse_answer = sparse_predictions['answer']
+    sparse_score = sparse_predictions['score']
+    sparse_start = sparse_predictions['start']
     
     # dense_start_time = time.perf_counter()
     # dense_predictions = dense_qa_pipeline(context=context,question=question)
@@ -38,7 +40,7 @@ def predict(context,question):
     # dense_duration = (dense_end_time - dense_start_time) * 1000
     # dense_answer = dense_predictions['answer']
 
-    return sparse_answer,sparse_duration #,dense_answer,dense_duration
+    return sparse_answer,sparse_duration,sparse_score,sparse_start #,dense_answer,dense_duration
 
 md = """This prediction model is designed to answer a question about a given input text--reading comprehension. The model does not just answer questions in general -- it only works from the text that you provide. However, automated reading comprehension can be a valuable task.
 
@@ -53,9 +55,10 @@ Author of Hugging Face Space: Benjamin Consolvo, AI Solutions Engineer Manager a
 # predict()
 context=gr.Text(lines=10,label="Context")
 question=gr.Text(label="Question")
-sparse_answer=gr.Text(label="Sparse Answer")
-sparse_duration=gr.Text(label="Sparse latency (ms)")
-
+sparse_answer=gr.Text(label="Answer")
+sparse_duration=gr.Text(label="Latency (ms)")
+sparse_score=gr.Text(label="Probability score")
+sparse_start=gr.Text(label="Starting character")
 # dense_answer=gr.Text(label="Dense Answer")
 # dense_duration=gr.Text(label="Dense latency (ms)")
 
@@ -66,7 +69,7 @@ iface = gr.Interface(
     fn=predict, 
     inputs=[context,question],
     # outputs=[sparse_answer,sparse_duration,dense_answer,dense_duration],
-    outputs=[sparse_answer,sparse_duration],
+    outputs=[sparse_answer,sparse_score,sparse_start,sparse_duration],
     examples=[[apple_context,apple_question]],
     title = "Question & Answer with Sparse BERT using the SQuAD dataset",
     description = md,