Streamlit application

02_csrd_search.py

@@ -138,4 +138,4 @@
 # COMMAND ----------
 
 # MAGIC %md
-# MAGIC Through this simple example, we have guided our model to formulate a point of view solely based on facts we provided upfront (through our vector store), facts that we know we can trust. For the purpose of that demo, we represented output as a form of an HTML notebook. In real life scenario, one should offer that capability as a chat interface outside of a notebook based environment (outside of the scope here). 
+# MAGIC Through this simple example, we have guided our model to formulate a point of view solely based on facts we provided upfront (through our vector store), facts that we know we can trust. For the purpose of that demo, we represented output as a form of an HTML notebook. In real life scenario, one should offer that capability as a chat interface outside of a notebook based environment.

04_csrd_assistant.py

@@ -301,7 +301,7 @@ def search_reference(article_reference: str) -> str:
 
 # COMMAND ----------
 
-query = 'Quelles sont les conditions pur que mon entreprise soit sujette au statut de micro entreprise? Dans le case de micro enterprise, quelles sont les dispositions particulieres en termes de normes de reporting?'
+query = 'Quelles sont les conditions pour que mon entreprise soit sujette au statut de micro entreprise? Dans le case de micro enterprise, quelles sont les dispositions particulieres en termes de normes de reporting?'
 
 answer = agent.invoke({"input": query})
 displayHTML(llm_html(query, answer['output']))

app/app.py

@@ -0,0 +1,157 @@
+import streamlit as st
+import time
+import os
+import configparser
+
+from langchain_core.vectorstores import VectorStoreRetriever
+from typing import List
+from langchain.docstore.document import Document
+from langchain_community.vectorstores import DatabricksVectorSearch
+from databricks.vector_search.client import VectorSearchClient
+from langchain_community.chat_models import ChatDatabricks
+from langchain.chains import ConversationalRetrievalChain
+from langchain.memory import ConversationBufferMemory
+from langchain.prompts import PromptTemplate
+
+
+class VectorStoreRetrieverFilter(VectorStoreRetriever):
+    def _get_relevant_documents(self, query: str, **kwargs) -> List[Document]:
+        docs = self.vectorstore.similarity_search(query, filters={"group": "PARAGRAPH"}, **self.search_kwargs)
+        return docs
+
+
+def load_model(endpoint, index):
+
+    template = """Given the context information and the chat history.
+    Answer compliance issue related to the CSRD directive only.
+    If the question is not related to regulatory compliance, kindly decline to answer. 
+    If you don't know the answer, just say that you don't know, don't try to make up an answer. 
+    Keep the answer as concise as possible, always citing articles and chapters whenever applicable.
+    Please do not repeat the answer and do not add any additional information. 
+    Context: {context}
+
+    Chat History: {chat_history}
+    Follow up question: {question}
+    Assistant:
+    """
+
+    prompt = PromptTemplate(template=template, input_variables=["context", "chat_history", "question"])
+
+    memory = ConversationBufferMemory(
+        memory_key='chat_history',
+        return_messages=True,
+        output_key='answer'
+    )
+
+    return ConversationalRetrievalChain.from_llm(
+        llm=load_chat_model(),
+        retriever=load_vector_store(endpoint, index),
+        combine_docs_chain_kwargs={'prompt': prompt},
+        memory=memory,
+        return_source_documents=True
+    )
+
+
+def load_chat_model():
+    return ChatDatabricks(
+        workspace_url=os.environ['DATABRICKS_HOST'], 
+        personal_access_token=os.environ['DATABRICKS_TOKEN'],
+        endpoint="databricks-dbrx-instruct", 
+        max_tokens=300, 
+        temperature=0, 
+        extra_body={"enable_safety_filter": True}
+        )
+
+
+def load_vector_store(endpoint, index):
+    vsc = VectorSearchClient(
+        workspace_url=os.environ['DATABRICKS_HOST'], 
+        personal_access_token=os.environ['DATABRICKS_TOKEN'],
+        disable_notice=True
+        )
+
+    vector_index = vsc.get_index(endpoint, index)
+    vector_store = DatabricksVectorSearch(vector_index, text_column="content", columns=["id", "label"])
+    return VectorStoreRetrieverFilter(vectorstore=vector_store)
+
+
+def bootstrap():
+    config = configparser.ConfigParser()
+    config.read('../config/environment.ini')
+    config = config['DEMO']
+    vector_endpoint = config['vector_endpoint']
+    vector_index = "{}.{}.{}".format(config['catalog'], config['schema'], config['vector_index'])
+    return load_model(vector_endpoint, vector_index)
+
+
+def response_generator(answer_str):
+    for word in answer_str.split():
+        yield word + " "
+        time.sleep(0.01)
+
+
+def parse_history(messages):
+    history = []
+    for i, m in enumerate(messages):
+        if i == len(messages) - 1:
+            break
+        history.append([m, messages[i + 1]])
+    return history
+
+
+# ------------------
+# CORE STREAMLIT APP
+# ------------------
+
+st.image('european_commision.svg', width=300)
+st.caption('''The Corporate Sustainability Reporting Directive (CSRD) is a European Union initiative 
+aimed at enhancing corporate accountability regarding sustainability matters. It mandates certain companies to 
+disclose non-financial information (such as environmental, social, and governance factors) in their annual reports 
+and other public disclosures. This assistant provides users with the ability to navigate the complexity of the CSRD directive, its numerous chapters, articles and paragraphs.''')
+
+# Initialize chat history and load model if necessary
+if "messages" not in st.session_state:
+    st.session_state.messages = []
+    st.session_state.llm = bootstrap()
+
+# Display chat messages from history
+for message in st.session_state.messages:
+    if message['role'] == 'assistant':
+        with st.chat_message(message["role"], avatar='assistant.png'):
+            st.markdown(message["content"])
+    else:
+        with st.chat_message(message["role"], avatar='user.png'):
+            st.markdown(message["content"])
+
+# Assign the user's input to the question variable and checked if it's not None in the same line
+if question := st.chat_input("Ask me anything about the CSRD directive"):
+
+    # Display user message in chat message container
+    with st.chat_message("user", avatar='user.png'):
+        st.markdown(question)
+
+    # Define our model input
+    model_input = {'question': question}
+
+    # Query our model
+    model_output = st.session_state.llm.invoke(model_input)
+
+    # Add user message to chat history
+    st.session_state.messages.append({"role": "user", "content": question})
+
+    # Display response in chat message container
+    with st.chat_message("assistant", avatar='assistant.png'):
+        response = st.write_stream(response_generator(model_output['answer']))
+        source_documents = [[doc.metadata['label'], doc.page_content] for doc in model_output['source_documents']]
+        for article_id, article_content in source_documents:
+            st.markdown(f":gray[**{article_id}**]")
+            pre_text = f"""
+                <body style="background: lightgray;">
+                    <pre style="white-space: break-spaces;color:#9aa2b1;font-size: 0.6em">{article_content}</pre>
+                </body>
+            """
+            st.html(pre_text)
+
+    # Add assistant response to chat history
+    st.session_state.messages.append({"role": "assistant", "content": response})
+