Merge pull request #3 from MachineLearningReply/other/AI-audit-agent

app.py

@@ -38,7 +38,7 @@ def manage_files(modal, document_store):
     if modal.is_open():
         with modal.container():
             uploaded_file = st.file_uploader(
-                "Upload a CV in PDF format",
+                "Upload a document in PDF format",
                 type=("pdf",),
                 on_change=new_file(),
                 disabled=st.session_state['document_qa_model'] is None,
@@ -57,7 +57,7 @@ def manage_files(modal, document_store):
             if uploaded_file:
                 st.session_state['file_uploaded'] = True
                 st.session_state['files'] = pd.concat([st.session_state['files'], edited_df])
-                with st.spinner('Processing the CV content...'):
+                with st.spinner('Processing the document content...'):
                     store_file_in_table(document_store, uploaded_file)
                     ingest_document(uploaded_file)
 
@@ -103,7 +103,7 @@ def init_session_state():
 
 def set_page_config():
     st.set_page_config(
-        page_title="CV Insights AI Assistant",
+        page_title="AI Audit Assistant",
         page_icon=":shark:",
         initial_sidebar_state="expanded",
         layout="wide",

document_qa_engine.py

@@ -76,12 +76,13 @@ def create_inference_pipeline(document_store, model_name, api_key):
         generator = OpenAIChatGenerator(api_key=Secret.from_token("<local LLM doesn't need an API key>"),
                                         model=model_name,
                                         api_base_url="http://localhost:1234/v1",
-                                        generation_kwargs={"max_tokens": MAX_TOKENS}
+                                        generation_kwargs={"max_tokens": MAX_TOKENS},
                                         )
     elif "gpt" in model_name:
         generator = OpenAIChatGenerator(api_key=Secret.from_token(api_key), model=model_name,
-                                        generation_kwargs={"max_tokens": MAX_TOKENS},
+                                        generation_kwargs={"max_tokens": MAX_TOKENS, "temperature": 0},
                                         streaming_callback=lambda chunk: print(chunk.content, end="", flush=True),
+                                        
                                         )
     else:
         generator = HuggingFaceTGIChatGenerator(token=Secret.from_token(api_key), model=model_name,
@@ -118,7 +119,7 @@ class DocumentQAEngine:
 
     def inference(self, query, input_messages: List[dict]):
         system_message = ChatMessage.from_system(
-            "You are a professional HR recruiter that answers questions based on the content of the uploaded CV. in 1 or 2 sentences.")
+            "You are a professional analyzer of git repos, having access to the repo content. In 1-3 sentences")
         messages = [system_message]
         for message in input_messages:
             if message["role"] == "user":
@@ -127,7 +128,7 @@ class DocumentQAEngine:
                 messages.append(
                     ChatMessage.from_user(message["content"]))
         messages.append(ChatMessage.from_user("""
-        Relevant information from the uploaded CV:
+        Relevant information from the uploaded repo:
             {% for doc in documents %}
                 {{ doc.content }}
             {% endfor %}

utils.py

@@ -50,7 +50,7 @@ def append_documentation_to_sidebar():
     with st.expander("Documentation"):
         st.markdown(
             """
-            Upload a CV as PDF document. Once the spinner stops, you can proceed to ask your questions. The answers will
+            Upload document as PDF document. Once the spinner stops, you can proceed to ask your questions. The answers will
             be displayed in the right column. The system will answer your questions using the content of the document
             and mark refrences over the PDF viewer.
             """)