Update app.py

app.py

@@ -1,15 +1,12 @@
 import streamlit as st
-from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline
+from transformers import pipeline
 import pandas as pd
 
 # Load the Question Answering model
 @st.cache_resource
 def load_qa_pipeline():
-    """Load the QA pipeline with deepset/roberta-base-squad2 model."""
-    model_name = "deepset/roberta-base-squad2"
-    tokenizer = AutoTokenizer.from_pretrained(model_name)
-    model = AutoModelForQuestionAnswering.from_pretrained(model_name)
-    return pipeline("question-answering", model=model, tokenizer=tokenizer)
+    """Load the QA pipeline with Rifky/Indobert-QA model."""
+    return pipeline("question-answering", model="Rifky/Indobert-QA", tokenizer="Rifky/Indobert-QA")
 
 qa_pipeline = load_qa_pipeline()
 
@@ -27,10 +24,8 @@ def find_best_context(question, dataset):
     best_score = 0
     best_context = None
 
-    for index, row in dataset.iterrows():
-        # Access the 'text' column in the row
+    for _, row in dataset.iterrows():
         context_text = row['text']
-        # Simple heuristic: Count the number of overlapping words
         overlap = len(set(question.lower().split()) & set(context_text.lower().split()))
         if overlap > best_score:
             best_score = overlap
@@ -39,26 +34,25 @@ def find_best_context(question, dataset):
     return best_context
 
 # Streamlit UI
-st.title("SOP Question Answering AI")
-st.markdown("Ask any question about Standard Operating Procedures:")
+st.title("Sistem Penjawab Pertanyaan SOP dengan IndoBERT")
+st.markdown("Ajukan pertanyaan seputar Prosedur Operasional Standar:")
 
 # User input
-question = st.text_area("Enter your question:", "")
+question = st.text_area("Masukkan pertanyaan Anda:", "")
 
 # Generate answer
-if st.button("Get Answer"):
+if st.button("Dapatkan Jawaban"):
     if question:
-        with st.spinner("Finding the best context..."):
-            # Automatically find the most relevant context
+        with st.spinner("Menemukan konteks yang paling relevan..."):
             context = find_best_context(question, dataset)
 
             if context:
-                with st.spinner("Answering your question..."):
+                with st.spinner("Menjawab pertanyaan Anda..."):
                     result = qa_pipeline(question=question, context=context)
-                    st.success("Answer:")
+                    st.success("Jawaban:")
                     st.write(result["answer"])
-                    st.write("Confidence Score:", result["score"])
+                    st.write("Skor Keyakinan:", result["score"])
             else:
-                st.warning("No relevant context found. Please try rephrasing your question.")
+                st.warning("Konteks yang relevan tidak ditemukan. Silakan coba pertanyaan lain.")
     else:
-        st.warning("Please enter a question.")
+        st.warning("Silakan masukkan pertanyaan.")