Update app.py

app.py

@@ -28,26 +28,27 @@ import chainlit as cl
 
 system_template = """
 Use the following pieces of context to answer the user's question.
-Please respond as if you are a human female customer service representative for Daysoff, 
-a Norwegian company that provides welfare services by offering access to cottages and 
-apartments for employees of member companies.
-By default, you respond (in Norwegian language) using a warm, direct, and professional tone. 
-Your expertise covers FAQs, and privacy policies. 
-If you don't know the answer, just say that you don't know, don't try to make up an answer:
-politely redirect the user to customer service at [email protected] and remind them to always 
-include their booking id (bestillingskode). 
+You are an AI customer service assistant for Daysoff and respond to queries in Norwegian language (by default).
+Your expertise is in providing fast, accurate and on-brand answers that covers frequently asked questions about 
+Daysoff firmahytteordning and personvernspolicy.
+
+If you don't know the answer, just say "Hmm, I'm not sure." Don't try to make up an answer.
+If the question is not about either of these two topics, politely inform them that you are tuned to only answer 
+questions about this and that all other queries are best directed to [email protected].
+
 You can make inferences based on the context as long as it still faithfully represents the feedback.
 
-Example of how your response should be direct:
+Example of how your response should be, using a warm, direct, and professional tone:
 
 ```
-foo      
+foo
 ```
 
 Begin!
 ----------------
 {context}"""
 
+
 messages = [
     SystemMessagePromptTemplate.from_template(system_template),
     HumanMessagePromptTemplate.from_template("{question}"),
@@ -55,28 +56,31 @@ messages = [
 prompt = ChatPromptTemplate(messages=messages)
 chain_type_kwargs = {"prompt": prompt}
 
+
+
 @cl.author_rename
 def rename(orig_author: str):
     rename_dict = {"Just a moment": "Thinking.."}
     return rename_dict.get(orig_author, orig_author)
 
 @cl.on_chat_start
-async def init():
+async def start():
 
-    msg = cl.Message(content=f"Building vector store...")
+    msg = cl.Message(content="Building vector store...")
     await msg.send()
 
-    loader = CSVLoader(file_path="./data/total_faq.csv", source_column="Answer") 
+    loader = CSVLoader(file_path="./data/total_faq.csv", source_column="Answer")
     data = loader.load()
 
     for i, doc in enumerate(data):
-        doc.metadata["row_index"] = i + 1  
-        doc.metadata["source"] = doc.metadata.get("Info_Url", "") 
+        doc.metadata["row_index"] = i + 1
+        doc.metadata["source"] = doc.metadata.get("Info_Url", "")
 
     text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
-
     documents = text_splitter.transform_documents(data)
+
     store = LocalFileStore("./cache/")
+
     core_embeddings_model = OpenAIEmbeddings()
     embedder = CacheBackedEmbeddings.from_bytes_store(
         core_embeddings_model, store, namespace=core_embeddings_model.model
@@ -84,34 +88,36 @@ async def init():
     docsearch = await cl.make_async(FAISS.from_documents)(documents, embedder)
 
     chain = RetrievalQA.from_chain_type(
-        ChatOpenAI(model="gpt-4", temperature=0.0, streaming=True), # streaming=False
+        ChatOpenAI(model="gpt-4", temperature=0.0, streaming=True),
         chain_type="stuff",
         return_source_documents=True,
         retriever=docsearch.as_retriever(),
-        chain_type_kwargs = {"prompt": prompt}
+        chain_type_kwargs={"prompt": prompt}
+    )
+
+    props = {
+        "questions": [
+            "Jeg har lagt inn en bestilling hva skjer videre?",
+            "Hva er betingelser for utleie?",
+            "Jeg fikk en e-post om ny bestilling, men jeg finner den ikke i systemet?",
+            "Med hvem deler dere mine personlige opplysninger?",
+            "Hvordan beskytter dere dataene mine?",
+            "Kan jeg ta med hund eller katt?"
+        ]
+    }
+
+    custom_element = cl.CustomElement(
+        name="ExampleQuestions",
+        props=props,
+        display="inline"
     )
 
-    html_content = """
-    <table style="width: 80%; background-color: #1e1e1e; color: white; border-radius: 8px;">
-        <tr>
-            <td style="padding: 10px;">
-                <strong>Eksempler på spørsmål:</strong>
-                <ul>
-                    <li>Hvordan registrerer jeg meg som bruker?</li>
-                    <li>Kan jeg ha med kjæledyr på hytta?</li>
-                    <li>Adferdsmessig annonsering?</li>
-                    <li>Hvordan blir dataene mine beskyttet?</li>
-                </ul>
-            </td>
-        </tr>
-    </table>
-    """
-    
-    html_element = cl.Text(content=html_content, name="HTML Table", display="inline")
-    await cl.Message(content=f"FAISS ready. Her er noen eksempler på spørsmål:☕️", elements=[html_element]).send()
-    #msg.content = f"FAISS ready. Her er noen eksempler på spørsmål:\n\n{markdown}"
-    #await msg.send()
-    
+    await cl.Message(
+        content="FAISS ready. Her er noen eksempler på populære spørsmål omkring firmahytteordning og personvern: 🤔",
+        elements=[custom_element]
+    ).send()
+
+
     cl.user_session.set("chain", chain)
 
 @cl.on_message
@@ -156,4 +162,4 @@ async def main(message):
         return
 
     else:
-        await cl.Message(content=f"No sources found").send()
+        await cl.Message(content=f"No sources found").send()