Create app.py

app.py

@@ -0,0 +1,31 @@
+import gradio as gr
+from transformers import BertForQuestionAnswering
+from transformers import BertTokenizerFast
+import torch
+
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+tokenizer = BertTokenizerFast.from_pretrained('bert-base-uncased')
+model = BertForQuestionAnswering.from_pretrained("CountingMstar/ai-tutor-bert-model").to(device)
+
+def get_prediction(context, question):
+    inputs = tokenizer.encode_plus(question, context, return_tensors='pt').to(device)
+    outputs = model(**inputs)
+    
+    answer_start = torch.argmax(outputs.start_logits)
+    answer_end = torch.argmax(outputs.end_logits) + 1
+    
+    answer = tokenizer.convert_tokens_to_string(tokenizer.convert_ids_to_tokens(inputs['input_ids'][0][answer_start:answer_end]))
+    
+    return answer
+
+def question_answer(context, question):
+    prediction = get_prediction(context, question)
+    return prediction
+
+iface = gr.Interface(
+    fn=question_answer,
+    inputs=[gr.Textbox("Context"), gr.Textbox("Question")],
+    outputs=gr.Textbox("Answer")
+)
+
+iface.launch()