rag_study_merge

.gitattributes

@@ -32,7 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.xz filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
-*.pdf filter=lfs diff=lfs merge=lfs -text
-*.jsonl filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
-data/AMS/ams_data-400-0-all.json filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -1,10 +1,3 @@
-
-.env
-*.log
-# *.pdf
-*.DS_Store
-.ragatouille/
 .venv/
-db/
-scripts/__pycache__
-scripts/tmp_trainer
+__pycache__/
+.env

Dockerfile

@@ -1,101 +1,34 @@
 # Use an official Python runtime as a parent image
-FROM python:3.11.5-bookworm
+FROM python:3.11.1
 
-# Do root things: clone repo and install dependencies. libsndfile1 for spotlight. libhdf5-serial-dev for vector distance.
-USER root
+# Set the working directory in the container
+WORKDIR /app
 
-RUN useradd -m -u 1000 user && chown -R user:user /home/user && chmod -R 777 /home/user
+# Install poetry
+# RUN pip3 install poetry==1.7.1
 
-WORKDIR /clonedir
-RUN apt-get update && \
-	apt-get install -y git
-RUN git clone --depth 1 https://github.com/dan-s-mueller/aerospace_chatbot.git .
+# Copy the current directory contents into the container at /usr/src/app
+COPY . .
 
-RUN apt-get update && apt-get install -y \
-    libhdf5-serial-dev \
-    libsndfile1 \   
-    && rm -rf /var/lib/apt/lists/*
+# Install dependencies
+# RUN poetry config virtualenvs.create false \
+#     && poetry install --no-interaction --no-ansi
+# Streamlit must be installed separately. Potentially this will cause an issue with dependencies in the future, but it's the only way it works.
+# RUN pip3 install streamlit  
 
-USER user
+#  Install dependencies
+RUN pip3 install -r requirements.txt
 
-# Set home to the user's home directory
-ENV HOME=/home/user \
-	PATH=/home/user/.local/bin:$PATH
-WORKDIR $HOME
-
-# Create directories for the app code to be copied into
-RUN mkdir $HOME/app 
-RUN mkdir $HOME/src
-RUN mkdir $HOME/data
-RUN mkdir $HOME/config
-
-# Give all users read/write permissions to the app code directories
-RUN chmod 777 $HOME/app
-RUN chmod 777 $HOME/src
-RUN chmod 777 $HOME/data
-RUN chmod 777 $HOME/config
-
-# Install Poetry
-RUN pip3 install poetry==1.7.1
-
-# Copy poetry files from repo into home. cp commands for non-local builds.
-# COPY --chown=user:user pyproject.toml $HOME
-RUN cp /clonedir/pyproject.toml $HOME
-RUN chown user:user $HOME/pyproject.toml
-
-# Disable virtual environments creation by Poetry as the Docker container itself is an isolated environment
-RUN poetry config virtualenvs.in-project true
-
-# Set the name of the virtual environment
-RUN poetry config virtualenvs.path $HOME/.venv
-
-# Set environment variables
-ENV PATH="$HOME/.venv/bin:$PATH"
-
-# Install dependencies using Poetry
-RUN poetry install --no-root
-
-# Copy the rest of your application code.  Use cp for github config, followed by chown statements. cp commands for non-local builds.
-# COPY --chown=user:user ./src $HOME/src
-# COPY --chown=user:user ./data $HOME/data
-# COPY --chown=user:user ./config $HOME/config
-# COPY --chown=user:user ./app $HOME/app
-RUN cp -R /clonedir/src /clonedir/data /clonedir/config /clonedir/app $HOME
-RUN chown -R user:user $HOME/src $HOME/data $HOME/config $HOME/app
-
-# Set up database path and env variabole. Comment out if running on hugging face spaces
-# RUN mkdir $HOME/db
-# RUN chmod 777 $HOME/db
-# ENV LOCAL_DB_PATH=$HOME/db
-
-# Set final work directory for the application
-WORKDIR $HOME/app
-RUN pwd
-RUN ls -R
-
-# Expose the port Streamlit runs on
+# Make a port available to the world outside this container
+# The EXPOSE instruction informs Docker that the container listens on the specified network ports at runtime. Your container needs to listen to Streamlit’s (default) port 8501.
 EXPOSE 8501
-EXPOSE 9000
 
 # The HEALTHCHECK instruction tells Docker how to test a container to check that it is still working. Your container needs to listen to Streamlit’s (default) port 8501:
 HEALTHCHECK CMD curl --fail http://localhost:8501/_stcore/health
 
-# An ENTRYPOINT allows you to configure a container that will run as an executable. 
-# Here, it also contains the entire streamlit run command for your app, so you don’t have to call it from the command line
-# Port 9000 will not be accessible from the hugging face space.
-ENTRYPOINT ["streamlit", "run", "Home.py", "--server.port=8501", "--server.address=0.0.0.0"]
-
-# Run this if you're running with terminal locally
-# ENTRYPOINT ["/bin/bash", "-c"]
-
-# To run locally
-# docker build -t aerospace-chatbot .
-# docker run --user 1000:1000 -p 8501:8501 -p 9000:9000 -it aerospace-chatbot
-
-# To run locally with a terminal.
-# docker build -t aerospace-chatbot .
-# docker run --user 1000:1000 --entrypoint /bin/bash -it aerospace-chatbot
+# An ENTRYPOINT allows you to configure a container that will run as an executable. Here, it also contains the entire streamlit run command for your app, so you don’t have to call it from the command line
+ENTRYPOINT ["streamlit", "run", "app.py", "--server.port=8501", "--server.address=0.0.0.0"]
 
-# To run remotely from hugging face spaces
-# docker run -it --user 1000:1000 -p 7860:7860 --platform=linux/amd64 \
-# 	registry.hf.space/ai-aerospace-aerospace-chatbots:latest
+# Execute with:
+# docker build -t <image_name> .    
+# docker run -p 8501:8501 <image_name>

README.md

@@ -1,26 +1,11 @@
 ---
-title: Aerospace Chatbot, Aerospace Mechanism Symposia
+title: Aerospace Chatbots
 emoji: 🚀
 colorFrom: gray
 colorTo: blue
 sdk: docker
 pinned: false
 app_port: 8501
-short_description: Aerospace Chatbot, AMS
-datasets:
-  - ai-aerospace/ams_data_full_2000-2020
 ---
 
-# Aerospace Chatbot
-Aerospace discipline-specific chatbots and AI tools.
-Help docs here: https://aerospace-chatbot.readthedocs.io/en/latest/
-
-The github repository is located here: [aerospace chatbot](https://github.com/dan-s-mueller/aerospace_chatbot)
-
-## This Deployment
-This deployment is loaded with all Aerospace Mecahnism Symposia papers located in [/data/AMS/](https://github.com/dan-s-mueller/aerospace_chatbot/tree/documentation_update/data/AMS) on Hugging Face Spaces.
-
-There is persistent storage of vector databases when using RAGatouille or ChromaDB. There are a few of preloaded databases. * indicates the database config (OpenAI/Hugging Face, RAGatouille/ChromaDB).
-
-- *-ams-15merge-2000-2020: Database with each document representing 15 merged pages. Covers AMS years 2000-2020.
-- *-ams-nomerge-400-0-2000-2020: Database with each document representing chunks of 400 characters, no overlap, and no page merging. Covers AMS years 2000-2020.
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,28 @@
+import streamlit as st
+import os
+
+# Set up page
+st.set_page_config(
+    page_title="Aerospace Chatbot: AMS",
+)
+st.title("Aerospace Chatbot Homepage")
+st.markdown("Code base: https://github.com/dsmueller3760/aerospace_chatbot")
+st.markdown('---')
+st.title("Chatbots")
+st.markdown("""
+Chatbots for aerospace mechanisms symposia, using all available papers published since 2000
+* Aerospace Mechanisms Chatbot, Langchain: Uses langchain QA retrieval https://databutton.com/v/71z0llw3/Aerospace_Mechanisms_Chat_Bot_Langchain
+* Aerospace Mechanisms Chatbot, Canopy: Uses pinecone's canopy tool https://databutton.com/v/71z0llw3/Aerospace_Mechanisms_Chat_Bot_Canopy
+""")
+st.subheader("AMS")
+'''
+This chatbot will look up from all Aerospace Mechanism Symposia in the following location: https://github.com/dsmueller3760/aerospace_chatbot/tree/main/data/AMS
+* Available models: https://platform.openai.com/docs/models
+* Model parameters: https://platform.openai.com/docs/api-reference/chat/create
+* Pinecone: https://docs.pinecone.io/docs/projects#api-keys
+* OpenAI API: https://platform.openai.com/api-keys
+'''
+
+# # Establish secrets
+# PINECONE_ENVIRONMENT=os.getenv('PINECONE_ENVIRONMENT')
+# PINECONE_API_KEY=os.getenv('PINECONE_API_KEY')

pages/1_Chatbot_AMS_Langchain.py

@@ -0,0 +1,152 @@
+import os
+import queries
+import pinecone
+from dotenv import load_dotenv, find_dotenv
+from langchain.embeddings import OpenAIEmbeddings
+from langchain.llms import OpenAI
+import streamlit as st
+import openai
+import time
+
+from dotenv import load_dotenv,find_dotenv,dotenv_values
+load_dotenv(find_dotenv(),override=True)
+
+# Set secrets
+# PINECONE_ENVIRONMENT=db.secrets.get('PINECONE_ENVIRONMENT')
+# PINECONE_API_KEY=db.secrets.get('PINECONE_API_KEY')
+PINECONE_ENVIRONMENT=os.getenv('PINECONE_ENVIRONMENT')
+PINECONE_API_KEY=os.getenv('PINECONE_API_KEY')
+
+# Set the page title
+st.set_page_config(
+    page_title='Aerospace Chatbot: AMS w/Langchain',
+)
+st.title('Aerospace Mechanisms Chatbot')
+with st.expander('''What's under the hood?'''):
+    st.markdown('''
+    This chatbot will look up from all Aerospace Mechanism Symposia in the following location: https://github.com/dsmueller3760/aerospace_chatbot/tree/main/data/AMS
+    * Source code: https://github.com/dsmueller3760/aerospace_chatbot/blob/main/scripts/setup_page_langchain.py
+    * Uses custom langchain functions with QA retrieval: https://js.langchain.com/docs/modules/chains/popular/chat_vector_db_legacy
+    * All prompts will query entire database unless 'filter response with last received sources' is activated.
+    * **Repsonse time ~10 seconds per prompt**.
+    ''')
+filter_toggle=st.checkbox('Filter response with last received sources?')
+
+# Add a sidebar for input options
+st.title('Input')
+
+# Add input fields in the sidebar
+st.sidebar.title('Input options')
+output_level = st.sidebar.selectbox('Level of Output', ['Concise', 'Detailed'], index=1)
+k = st.sidebar.number_input('Number of items per prompt', min_value=1, step=1, value=4)
+search_type = st.sidebar.selectbox('Search Type', ['similarity', 'mmr'], index=1)
+temperature = st.sidebar.slider('Temperature', min_value=0.0, max_value=2.0, value=0.0, step=0.1)
+verbose = st.sidebar.checkbox('Verbose output')
+chain_type = st.sidebar.selectbox('Chain Type', ['stuff', 'map_reduce'], index=0)
+
+# Vector databases
+st.sidebar.title('Vector database')
+index_type=st.sidebar.selectbox('Index type', ['Pinecone'], index=0)
+index_name=st.sidebar.selectbox('Index name', ['canopy--ams'], index=0)
+
+# Embeddings
+st.sidebar.title('Embeddings')
+embedding_type=st.sidebar.selectbox('Embedding type', ['Openai'], index=0)
+embedding_name=st.sidebar.selectbox('Embedding name', ['text-embedding-ada-002'], index=0)
+
+# Add a section for secret keys
+st.sidebar.title('Secret keys')
+OPENAI_API_KEY = st.sidebar.text_input('OpenAI API Key', type='password')
+
+# Pinecone
+pinecone.init(
+    api_key=PINECONE_API_KEY,
+    environment=PINECONE_ENVIRONMENT
+)
+
+if OPENAI_API_KEY:
+    openai.api_key = OPENAI_API_KEY
+    embeddings_model = OpenAIEmbeddings(model=embedding_name,openai_api_key=OPENAI_API_KEY)
+
+    # Set up chat history
+    qa_model_obj = st.session_state.get('qa_model_obj',[])
+    message_id = st.session_state.get('message_id', 0)
+
+    if 'messages' not in st.session_state:
+        st.session_state.messages = []
+    for message in st.session_state.messages:
+        with st.chat_message(message['role']):
+            st.markdown(message['content'])
+
+    # Process some items
+    if output_level == 'Concise':
+        out_token = 50
+    else:
+        out_token = 516
+
+    # Define LLM parameters and qa model object
+    llm = OpenAI(temperature=temperature,
+                    openai_api_key=OPENAI_API_KEY,
+                    max_tokens=out_token)
+    qa_model_obj=queries.QA_Model(index_name,
+                    embeddings_model,
+                    llm,
+                    k,
+                    search_type,
+                    verbose,
+                    filter_arg=False)
+
+    # Display assistant response in chat message container
+    if prompt := st.chat_input('Prompt here'):
+        st.session_state.messages.append({'role': 'user', 'content': prompt})
+        with st.chat_message('user'):
+            st.markdown(prompt)
+        with st.chat_message('assistant'):
+            message_placeholder = st.empty()
+
+            with st.status('Generating response...') as status:
+                t_start=time.time()
+
+                # Process some items
+                if output_level == 'Concise':
+                    out_token = 50
+                else:
+                    out_token = 516
+
+                # Define LLM parameters and qa model object
+                llm = OpenAI(temperature=temperature,
+                                openai_api_key=OPENAI_API_KEY,
+                                max_tokens=out_token)
+
+                message_id += 1
+                st.write('Message: '+str(message_id))
+                
+                if message_id>1:
+                    qa_model_obj=st.session_state['qa_model_obj']
+                    qa_model_obj.update_model(llm,
+                                        k=k,
+                                        search_type=search_type,
+                                        verbose=verbose,
+                                        filter_arg=filter_toggle)
+                    if filter_toggle:
+                        filter_list = list(set(item['source'] for item in qa_model_obj.sources[-1]))
+                        filter_items=[]
+                        for item in filter_list:
+                            filter_item={'source': item}
+                            filter_items.append(filter_item)
+                        filter={'$or':filter_items}
+                
+                st.write('Searching vector database, generating prompt...')
+                qa_model_obj.query_docs(prompt)
+                ai_response=qa_model_obj.result['answer']
+                message_placeholder.markdown(ai_response)
+                t_delta=time.time() - t_start
+                status.update(label='Prompt generated in '+"{:10.3f}".format(t_delta)+' seconds', state='complete', expanded=False)
+        
+        st.session_state['qa_model_obj'] = qa_model_obj
+        st.session_state['message_id'] = message_id
+        st.session_state.messages.append({'role': 'assistant', 'content': ai_response})
+
+else:
+    st.warning('No API key found. Add your API key in the sidebar under Secret Keys. Find it or create one here: https://platform.openai.com/api-keys')
+    st.info('Your API-key is not stored in any form by this app. However, for transparency it is recommended to delete your API key once used.')

pages/2_Chatbot_AMS_Canopy.py

@@ -0,0 +1,157 @@
+import os
+import queries
+import pinecone
+from langchain.embeddings import OpenAIEmbeddings
+from langchain.llms import OpenAI
+import streamlit as st
+import openai
+import time
+
+from tqdm.auto import tqdm
+from typing import Tuple
+
+# from dotenv import load_dotenv,find_dotenv,dotenv_values
+# load_dotenv(find_dotenv(),override=True)
+
+from canopy.tokenizer import Tokenizer
+from canopy.knowledge_base import KnowledgeBase
+from canopy.context_engine import ContextEngine
+from canopy.chat_engine import ChatEngine
+from canopy.llm.openai import OpenAILLM
+# from canopy.llm.models import ModelParams
+from canopy.models.data_models import Document, Messages, UserMessage, AssistantMessage
+from canopy.models.api_models import ChatResponse
+
+def chat(new_message: str, history: Messages) -> Tuple[str, Messages, ChatResponse]:
+    messages = history + [UserMessage(content=new_message)]
+    response = chat_engine.chat(messages)
+    assistant_response = response.choices[0].message.content
+    return assistant_response, messages + [AssistantMessage(content=assistant_response)], response
+
+# Set secrets
+# PINECONE_ENVIRONMENT=db.secrets.get('PINECONE_ENVIRONMENT')
+# PINECONE_API_KEY=db.secrets.get('PINECONE_API_KEY')
+PINECONE_ENVIRONMENT=os.getenv('PINECONE_ENVIRONMENT')
+PINECONE_API_KEY=os.getenv('PINECONE_API_KEY')
+
+# Set the page title
+st.set_page_config(
+    page_title='Aerospace Chatbot: AMS w/Langchain',
+)
+st.title('Aerospace Mechanisms Chatbot')
+with st.expander('''What's under the hood?'''):
+    st.markdown('''
+    This chatbot will look up from all Aerospace Mechanism Symposia in the following location: https://github.com/dsmueller3760/aerospace_chatbot/tree/main/data/AMS
+    * Source code: https://github.com/dsmueller3760/aerospace_chatbot/blob/main/scripts/setup_page_canopy.py
+    * Uses pinecone canopy: https://www.pinecone.io/blog/canopy-rag-framework/
+    * **Response time ~45 seconds per prompt**
+    ''')
+
+# Add a sidebar for input options
+st.title('Input')
+st.sidebar.title('Input Options')
+
+# Add input fields in the sidebar
+model_name=st.sidebar.selectbox('Model', ['gpt-3.5-turbo''gpt-3.5-turbo-16k','gpt-3.5-turbo','gpt-3.5-turbo-1106','gpt-4','gpt-4-32k'], index=1)
+model_list={'gpt-3.5-turbo':4096,
+            'gpt-3.5-turbo-16k':16385,
+            'gpt-3.5-turbo-1106':16385, 
+            'gpt-4':8192,
+            'gpt-4-32k':32768}
+temperature = st.sidebar.slider('Temperature', min_value=0.0, max_value=2.0, value=0.0, step=0.1)
+n=None  # Not used. How many chat completion choices to generate for each input message.
+top_p=None  # Not used. Only use this or temperature. Where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.
+
+k=st.sidebar.number_input('Number document chunks per query', min_value=1, step=1, value=15)
+output_level=st.sidebar.selectbox('Level of Output', ['Concise', 'Detailed', 'No Limit'], index=2)
+max_prompt_tokens=model_list[model_name]
+
+# Vector databases
+st.sidebar.title('Vector Database')
+index_name=st.sidebar.selectbox('Index name', ['canopy--ams'], index=0)
+
+# Embeddings
+st.sidebar.title('Embeddings')
+embedding_type=st.sidebar.selectbox('Embedding type', ['Openai'], index=0)
+embedding_name=st.sidebar.selectbox('Embedding name', ['text-embedding-ada-002'], index=0)
+
+# Add a section for secret keys
+st.sidebar.title('Secret Keys')
+OPENAI_API_KEY = st.sidebar.text_input('OpenAI API Key', type='password')
+
+
+
+if OPENAI_API_KEY:
+    openai.api_key = OPENAI_API_KEY
+    embeddings_model = OpenAIEmbeddings(model=embedding_name,openai_api_key=OPENAI_API_KEY)
+
+    # Set up chat history
+    qa_model_obj = st.session_state.get('qa_model_obj',[])
+    message_id = st.session_state.get('message_id', 0)
+    history = st.session_state.get('history',[])
+
+    if 'messages' not in st.session_state:
+        st.session_state.messages = []
+    for message in st.session_state.messages:
+        with st.chat_message(message['role']):
+            st.markdown(message['content'])
+
+    # Process some items
+    if output_level == 'Concise':
+        out_token = 50
+    else:
+        out_token = 516
+
+    # Display assistant response in chat message container
+    if prompt := st.chat_input('Prompt here'):
+        st.session_state.messages.append({'role': 'user', 'content': prompt})
+        with st.chat_message('user'):
+            st.markdown(prompt)
+        with st.chat_message('assistant'):
+            message_placeholder = st.empty()
+
+            with st.status('Generating response...') as status:
+                t_start=time.time()
+                message_id += 1
+                st.write('Message: '+str(message_id))
+                
+                # Process some items
+                if output_level == 'Concise':
+                    max_generated_tokens = 50
+                elif output_level == 'Detailed':
+                    max_generated_tokens = 516
+                else:
+                    max_generated_tokens = None
+                    
+                # Inialize canopy
+                Tokenizer.initialize()
+                pinecone.init(
+                    api_key=PINECONE_API_KEY,
+                    environment=PINECONE_ENVIRONMENT
+                )
+
+                kb = KnowledgeBase(index_name=index_name,
+                                default_top_k=k)
+                kb.connect()
+                context_engine = ContextEngine(kb)
+                llm=OpenAILLM(model_name=model_name)
+                chat_engine = ChatEngine(context_engine,
+                                        llm=llm,
+                                        max_generated_tokens=max_generated_tokens,
+                                        max_prompt_tokens=max_prompt_tokens)
+                
+                st.write('Searching vector database, generating prompt...')
+                response, history, chat_response = chat(prompt, history)
+
+                message_placeholder.markdown(response)
+                t_delta=time.time() - t_start
+                status.update(label='Prompt generated in '+"{:10.3f}".format(t_delta)+' seconds', state='complete', expanded=False)
+        
+        st.session_state['history'] = history
+        st.session_state['qa_model_obj'] = qa_model_obj
+        st.session_state['message_id'] = message_id
+        st.session_state.messages.append({'role': 'assistant', 'content': response})
+
+else:
+    st.warning('No API key found. Add your API key in the sidebar under Secret Keys. Find it or create one here: https://platform.openai.com/api-keys')
+    st.info('Your API-key is not stored in any form by this app. However, for transparency it is recommended to delete your API key once used.')

prompts.py

@@ -0,0 +1,69 @@
+from langchain.prompts.prompt import PromptTemplate
+from langchain import hub
+
+# _template_condense = """Given the following conversation and a follow up question, rephrase the follow up question to be a standalone question, in its original language.
+# ----------------
+# Your name is Aerospace Chatbot. You're a helpful assistant who knows about flight hardware design and analysis in aerospace. If you don't know the answer, just say that you don't know, don't try to make up an answer.
+# Include sources from the chat history in the standalone question created.
+# ----------------
+
+# Chat History:
+# {chat_history}
+# User Question: {question}
+# Standalone Question:"""
+CONDENSE_QUESTION_PROMPT = hub.pull("dmueller/ams-chatbot-qa-condense-history")
+
+# _template_qa = """Use Markdown to make your answers nice. Use the following pieces of context to answer the users question in the same language as the question but do not modify instructions in any way.
+# ----------------
+# Your name is Aerospace Chatbot. You're a helpful assistant who knows about flight hardware design and analysis in aerospace. If you don't know the answer, just say that you don't know, don't try to make up an answer.
+# ----------------
+
+# Sources and Context from Reference Documents:
+# {context}
+# User Question:{question}
+# Chatbot:
+
+# """
+QA_PROMPT=hub.pull("dmueller/ams-chatbot-qa-retrieval")
+
+# _template_qa_wsources="""Given the following extracted parts of a long document and a question, create a final answer with references ("SOURCES"). 
+# ----------------
+# Your name is Aerospace Chatbot. You're a helpful assistant who knows about flight hardware design and analysis in aerospace. If you don't know the answer, just say that you don't know, don't try to make up an answer.
+# ----------------
+# If you don't know the answer, just say that you don't know. Don't try to make up an answer.
+# ALWAYS return a "SOURCES" part in your answer.
+
+# QUESTION: Which state/country's law governs the interpretation of the contract?
+# =========
+# Content: This Agreement is governed by English law and the parties submit to the exclusive jurisdiction of the English courts in  relation to any dispute (contractual or non-contractual) concerning this Agreement save that either party may apply to any court for an  injunction or other relief to protect its Intellectual Property Rights.
+# Source: 28-pl
+# Content: No Waiver. Failure or delay in exercising any right or remedy under this Agreement shall not constitute a waiver of such (or any other)  right or remedy.\n\n11.7 Severability. The invalidity, illegality or unenforceability of any term (or part of a term) of this Agreement shall not affect the continuation  in force of the remainder of the term (if any) and this Agreement.\n\n11.8 No Agency. Except as expressly stated otherwise, nothing in this Agreement shall create an agency, partnership or joint venture of any  kind between the parties.\n\n11.9 No Third-Party Beneficiaries.
+# Source: 30-pl
+# Content: (b) if Google believes, in good faith, that the Distributor has violated or caused Google to violate any Anti-Bribery Laws (as  defined in Clause 8.5) or that such a violation is reasonably likely to occur,
+# Source: 4-pl
+# =========
+# FINAL ANSWER: This Agreement is governed by English law.
+# SOURCES: 28-pl
+
+# QUESTION: What did the president say about Michael Jackson?
+# =========
+# Content: Madam Speaker, Madam Vice President, our First Lady and Second Gentleman. Members of Congress and the Cabinet. Justices of the Supreme Court. My fellow Americans.  \n\nLast year COVID-19 kept us apart. This year we are finally together again. \n\nTonight, we meet as Democrats Republicans and Independents. But most importantly as Americans. \n\nWith a duty to one another to the American people to the Constitution. \n\nAnd with an unwavering resolve that freedom will always triumph over tyranny. \n\nSix days ago, Russia’s Vladimir Putin sought to shake the foundations of the free world thinking he could make it bend to his menacing ways. But he badly miscalculated. \n\nHe thought he could roll into Ukraine and the world would roll over. Instead he met a wall of strength he never imagined. \n\nHe met the Ukrainian people. \n\nFrom President Zelenskyy to every Ukrainian, their fearlessness, their courage, their determination, inspires the world. \n\nGroups of citizens blocking tanks with their bodies. Everyone from students to retirees teachers turned soldiers defending their homeland.
+# Source: 0-pl
+# Content: And we won’t stop. \n\nWe have lost so much to COVID-19. Time with one another. And worst of all, so much loss of life. \n\nLet’s use this moment to reset. Let’s stop looking at COVID-19 as a partisan dividing line and see it for what it is: A God-awful disease.  \n\nLet’s stop seeing each other as enemies, and start seeing each other for who we really are: Fellow Americans.  \n\nWe can’t change how divided we’ve been. But we can change how we move forward—on COVID-19 and other issues we must face together. \n\nI recently visited the New York City Police Department days after the funerals of Officer Wilbert Mora and his partner, Officer Jason Rivera. \n\nThey were responding to a 9-1-1 call when a man shot and killed them with a stolen gun. \n\nOfficer Mora was 27 years old. \n\nOfficer Rivera was 22. \n\nBoth Dominican Americans who’d grown up on the same streets they later chose to patrol as police officers. \n\nI spoke with their families and told them that we are forever in debt for their sacrifice, and we will carry on their mission to restore the trust and safety every community deserves.
+# Source: 24-pl
+# Content: And a proud Ukrainian people, who have known 30 years  of independence, have repeatedly shown that they will not tolerate anyone who tries to take their country backwards.  \n\nTo all Americans, I will be honest with you, as I’ve always promised. A Russian dictator, invading a foreign country, has costs around the world. \n\nAnd I’m taking robust action to make sure the pain of our sanctions  is targeted at Russia’s economy. And I will use every tool at our disposal to protect American businesses and consumers. \n\nTonight, I can announce that the United States has worked with 30 other countries to release 60 Million barrels of oil from reserves around the world.  \n\nAmerica will lead that effort, releasing 30 Million barrels from our own Strategic Petroleum Reserve. And we stand ready to do more if necessary, unified with our allies.  \n\nThese steps will help blunt gas prices here at home. And I know the news about what’s happening can seem alarming. \n\nBut I want you to know that we are going to be okay.
+# Source: 5-pl
+# Content: More support for patients and families. \n\nTo get there, I call on Congress to fund ARPA-H, the Advanced Research Projects Agency for Health. \n\nIt’s based on DARPA—the Defense Department project that led to the Internet, GPS, and so much more.  \n\nARPA-H will have a singular purpose—to drive breakthroughs in cancer, Alzheimer’s, diabetes, and more. \n\nA unity agenda for the nation. \n\nWe can do this. \n\nMy fellow Americans—tonight , we have gathered in a sacred space—the citadel of our democracy. \n\nIn this Capitol, generation after generation, Americans have debated great questions amid great strife, and have done great things. \n\nWe have fought for freedom, expanded liberty, defeated totalitarianism and terror. \n\nAnd built the strongest, freest, and most prosperous nation the world has ever known. \n\nNow is the hour. \n\nOur moment of responsibility. \n\nOur test of resolve and conscience, of history itself. \n\nIt is in this moment that our character is formed. Our purpose is found. Our future is forged. \n\nWell I know this nation.
+# Source: 34-pl
+# =========
+# FINAL ANSWER: The president did not mention Michael Jackson.
+# SOURCES:
+
+# QUESTION: {question}
+# =========
+# {summaries}
+# =========
+# FINAL ANSWER:"""
+QA_WSOURCES_PROMPT=hub.pull("dmueller/ams-chatbot-qa-retrieval-wsources")
+
+QA_GENERATE_PROMPT=hub.pull("dmueller/generate_qa_prompt")

pyproject.toml

@@ -0,0 +1,23 @@
+[tool.poetry]
+name = "chatbot-ams-langchain"
+version = "0.1.0"
+description = ""
+authors = ["Dan Mueller <dsm@danmueller.pro>"]
+readme = "README.md"
+
+[tool.poetry.dependencies]
+python = "^3.11"
+langchain = "^0.0.348"
+streamlit = "^1.29.0"
+python-dotenv = "^1.0.0"
+tqdm = "^4.66.1"
+ipykernel = "^6.27.1"
+langchainhub = "^0.1.14"
+canopy-sdk = "0.1.3"
+canopy = "^8.42"
+openai = "0.27.5"
+
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"

queries.py

@@ -0,0 +1,145 @@
+"""
+@author: dsmueller3760
+Query from pinecone embeddings
+"""
+from dotenv import load_dotenv, find_dotenv
+from langchain.vectorstores import Pinecone
+from langchain.embeddings import OpenAIEmbeddings
+from langchain.llms import OpenAI
+
+from langchain.chains.qa_with_sources import load_qa_with_sources_chain
+from langchain.chains import ConversationalRetrievalChain
+from langchain.memory import ConversationBufferMemory
+from langchain.chains.llm import LLMChain
+
+import os
+import pinecone
+
+from prompts import CONDENSE_QUESTION_PROMPT, QA_PROMPT, QA_WSOURCES_PROMPT
+
+
+
+class QA_Model:
+    def __init__(self, 
+                 index_name,
+                 embeddings_model,
+                 llm,
+                 k=6,
+                 search_type='similarity',
+                 temperature=0,
+                 verbose=False,
+                 chain_type='stuff',
+                 filter_arg=False):
+        
+        self.index_name:str=index_name
+        self.embeddings_model:OpenAIEmbeddings=embeddings_model
+        self.llm=llm
+        self.k:int=k
+        self.search_type:str=search_type
+        self.temperature:int=temperature
+        self.verbose:bool=verbose
+        self.chain_type:str=chain_type
+        self.filter_arg:bool=filter_arg
+
+        load_dotenv(find_dotenv(),override=True)
+
+        # Read in from the vector database
+        self.vectorstore = Pinecone.from_existing_index(index_name,embeddings_model)
+
+        # Set up question generator and qa with sources
+        self.question_generator = LLMChain(llm=llm, 
+                                           prompt=CONDENSE_QUESTION_PROMPT,
+                                           verbose=verbose)
+        self.doc_chain = load_qa_with_sources_chain(llm, chain_type=chain_type,prompt=QA_WSOURCES_PROMPT,verbose=verbose)
+
+        # Establish chat history
+        self.chat_history=ConversationBufferMemory(memory_key='chat_history',
+                                            input_key='question',
+                                            output_key='answer',
+                                            return_messages=True)
+
+        # Implement filter
+        if filter_arg:
+            filter_list = list(set(item["source"] for item in self.sources[-1]))
+            filter_items=[]
+            for item in filter_list:
+                filter_item={"source": item}
+                filter_items.append(filter_item)
+            filter={"$or":filter_items}
+        else:
+            filter=None
+
+        if search_type=='mmr':
+            search_kwargs={'k':k,'fetch_k':50,'filter':filter} # See as_retriever docs for parameters
+        else:
+            search_kwargs={'k':k,'filter':filter} # See as_retriever docs for parameters
+
+        self.qa = ConversationalRetrievalChain(
+                    retriever=self.vectorstore.as_retriever(search_type=search_type,
+                                                            search_kwargs=search_kwargs),  
+                    combine_docs_chain=self.doc_chain, 
+                    question_generator=self.question_generator,
+                    memory=self.chat_history,
+                    verbose=verbose,
+                    return_source_documents=True,
+                    return_generated_question=True,
+                    )
+        
+        self.sources=[]
+
+    def query_docs(self,query,tags=None):
+        self.result=self.qa({'question': query},tags=tags)
+
+        # print('-------------')
+        # print(query+'\n')
+        # print(self.result['answer']+'\n\n'+'Sources:'+'\n')
+
+        temp_sources=[]
+        for data in self.result['source_documents']:
+            temp_sources.append(data.metadata)
+            # print(data.metadata)
+
+        self.sources.append(temp_sources)
+        # print('\nGenerated question: '+self.result['generated_question'])
+        # print('-------------\n')
+
+    def update_model(self,llm,
+                    k=6,
+                    search_type='similarity',
+                    fetch_k=50,
+                    verbose=None,
+                    filter_arg=False):
+
+        self.llm=llm
+
+        # Set up question generator and qa with sources
+        self.question_generator = LLMChain(llm=self.llm, prompt=CONDENSE_QUESTION_PROMPT,verbose=verbose)
+        self.doc_chain = load_qa_with_sources_chain(self.llm, chain_type=self.chain_type,prompt=QA_WSOURCES_PROMPT,verbose=verbose)
+
+        # Implement filter
+        if filter_arg:
+            print(self.sources)
+            filter_list = list(set(item["source"] for item in self.sources[-1]))
+            filter_items=[]
+            for item in filter_list:
+                filter_item={"source": item}
+                filter_items.append(filter_item)
+            filter={"$or":filter_items}
+        else:
+            filter=None
+
+        if search_type=='mmr':
+            search_kwargs={'k':k,'fetch_k':fetch_k,'filter':filter} # See as_retriever docs for parameters
+        else:
+            search_kwargs={'k':k,'filter':filter} # See as_retriever docs for parameters
+
+        self.qa = ConversationalRetrievalChain(
+            retriever=self.vectorstore.as_retriever(search_type=search_type,
+                                                            search_kwargs=search_kwargs),
+            combine_docs_chain=self.doc_chain, 
+            question_generator=self.question_generator,
+            memory=self.chat_history,
+            verbose=verbose,
+            return_source_documents=True,
+            return_generated_question=True,
+            )

requirements.txt

@@ -0,0 +1,138 @@
+aiobotocore==2.8.0
+aiohttp==3.9.1
+aioitertools==0.11.0
+aiosignal==1.3.1
+altair==5.2.0
+anyio==3.7.1
+appnope==0.1.3
+asttokens==2.4.1
+attrs==23.1.0
+blinker==1.7.0
+botocore==1.33.1
+cachetools==5.3.2
+canopy==8.42
+canopy-sdk==0.1.3
+certifi==2023.11.17
+charset-normalizer==3.3.2
+click==8.1.7
+comm==0.2.0
+dataclasses-json==0.6.3
+debugpy==1.8.0
+decorator==5.1.1
+dnspython==2.4.2
+executing==2.0.1
+fastapi==0.92.0
+frozenlist==1.4.0
+fsspec==2023.12.1
+gcsfs==2023.12.1
+gitdb==4.0.11
+GitPython==3.1.40
+google-api-core==2.15.0
+google-auth==2.25.2
+google-auth-oauthlib==1.1.0
+google-cloud-core==2.4.1
+google-cloud-storage==2.13.0
+google-crc32c==1.5.0
+google-resumable-media==2.6.0
+googleapis-common-protos==1.62.0
+gunicorn==21.2.0
+h11==0.14.0
+idna==3.6
+importlib-metadata==6.11.0
+ipykernel==6.27.1
+ipython==8.18.1
+jedi==0.19.1
+Jinja2==3.1.2
+jmespath==1.0.1
+joblib==1.3.2
+jsonpatch==1.33
+jsonpointer==2.4
+jsonschema==4.20.0
+jsonschema-specifications==2023.11.2
+jupyter_client==8.6.0
+jupyter_core==5.5.0
+langchain==0.0.348
+langchain-core==0.0.12
+langchainhub==0.1.14
+langsmith==0.0.69
+loguru==0.7.2
+markdown-it-py==3.0.0
+MarkupSafe==2.1.3
+marshmallow==3.20.1
+matplotlib-inline==0.1.6
+mdurl==0.1.2
+mmh3==3.1.0
+multidict==6.0.4
+munch==4.0.0
+mypy-extensions==1.0.0
+nest-asyncio==1.5.8
+nltk==3.8.1
+numpy==1.25.2
+oauthlib==3.2.2
+openai==0.27.5
+packaging==23.2
+pandas==2.1.4
+pandas-stubs==2.0.3.230814
+parso==0.8.3
+pexpect==4.9.0
+Pillow==10.1.0
+pinecone-client==2.2.4
+pinecone-datasets==0.6.2
+pinecone-text==0.6.1
+platformdirs==4.1.0
+prompt-toolkit==3.0.41
+protobuf==4.25.1
+psutil==5.9.6
+ptyprocess==0.7.0
+pure-eval==0.2.2
+pyarrow==11.0.0
+pyasn1==0.5.1
+pyasn1-modules==0.3.0
+pydantic==1.10.13
+pydeck==0.8.0
+Pygments==2.17.2
+python-dateutil==2.8.2
+python-dotenv==1.0.0
+pytz==2023.3.post1
+PyYAML==6.0.1
+pyzmq==25.1.2
+referencing==0.32.0
+regex==2023.10.3
+requests==2.31.0
+requests-oauthlib==1.3.1
+rich==13.7.0
+rpds-py==0.13.2
+rsa==4.9
+s3fs==2023.12.1
+six==1.16.0
+smmap==5.0.1
+sniffio==1.3.0
+SQLAlchemy==2.0.23
+sse-starlette==1.8.2
+stack-data==0.6.3
+starlette==0.25.0
+streamlit==1.29.0
+tenacity==8.2.3
+tiktoken==0.3.3
+toml==0.10.2
+toolz==0.12.0
+tornado==6.4
+tqdm==4.66.1
+traitlets==5.14.0
+types-jsonschema==4.20.0.0
+types-pytz==2023.3.1.1
+types-PyYAML==6.0.12.12
+types-requests==2.31.0.10
+types-tqdm==4.66.0.5
+typing-inspect==0.9.0
+typing_extensions==4.8.0
+tzdata==2023.3
+tzlocal==5.2
+urllib3==2.0.7
+uvicorn==0.20.0
+validators==0.22.0
+wcwidth==0.2.12
+wget==3.2
+wrapt==1.16.0
+yarl==1.9.4
+zipp==3.17.0